An energy–trust balanced and attack-resilient cluster head election scheme for resource-constrained wireless sensor networks Authors: R. Gopi, C. Gunasundari, B. Bazeer Ahamed, G. Sathya, K. B. Bhaskar, T. R. Priyadharshini Published: 2026 ABSTRACT Wireless Sensor Networks (WSNs) demand optimized Cluster Head (CH) election mechanisms to balance energy efficiency and communication security. Traditional protocols such as LEACH and ISSA rely on probabilistic, or metaheuristic approaches that do not fully integrate security metrics into CH selection, leading to increased vulnerability and uneven energy depletion. This paper proposes an Energy-Driven Cluster Head Secure Selection (ED-CHSS) algorithm that integrates residual-energy modeling, node trust-index evaluation, and lightweight symmetric cryptographic primitives for secure intra-cluster communication. The election process of CH is designed based on an optimization model of multiple parameters, including the residual energy ratio, communication cost and trust weight coefficient. CH candidacy probability is dynamically adapted by the node density and link stability. The simulation results show that ED-CHSS decreases average energy dissipation by 28–32%, increases network lifetime by about 38–42%, and increases the attack resistance of the network against selective forwarding and sinkhole attacks. The proposed model provides secure and energy balanced clustering for mission critical WSN deployments. Introduction Background WSN has now been widely used in different fields, including smart cities, medical field, and environment monitoring1. Security and CH selection are major aspects of WSNs2. Selecting the proper CH has a positive effect on data aggregation, network lifetime, and energy consumption3. Restrictive security policies protect users from attacks and ensure the confidentiality of data sent4. However, existing CH selection approaches are not scalable or energy-efficient, and existing security approaches do not account for emerging threats5. Numerous complicated algorithms have been developed to address these determinants; however, few have considered a combination of integrity and efficiency in their communication protocols6. In addition, many algorithms do not account for the need for real-time adjustment, which is not surprising given the variability of WSNs7. The latest research has examined lighter yet stronger encryption methods and adaptive, dynamic selection of CHs to achieve more efficient, secure networks8. Based on these recent developments9, this study proposes a way to make the entire WSN process, and parts of it, more effective and more secure simultaneously. Given the increased need for large-scale WSN installations, it is crucial to have both a secure and fast network10. This study is designed to create an all-in-one approach to solve the problems of CH selection and security for multiple WSN applications and face up to emerging threats. Problem statement Current methods and security measures used for selecting Cluster heads cannot simultaneously improve power efficiency, scalability, and attack resistance. Existing methods do not incorporate network security features or have real-time change capabilities. The focus of this research is to improve the performance and security level for WSN by addressing these limitations using a hybrid method called Enhanced Dynamic Cluster Head Selection and Security (ED-CHSS). The proposed framework is a clear contrast to traditional clustering protocols: it is the only protocol that considers energy-aware cluster head selection, dynamic scheduling and trust-based security mechanisms to reduce energy consumption and still provide reliable communication in large-scale WSNs. Contributions of this paper The major contributions are;By clustering nodes based on up-to-date network data to select the best CH, the ED-CHSS approach improves energy efficiency and overall network performance.An encryption mechanism with low overhead is integrated into the CH selection process to ensure secure information transfer without negatively impacting network energy and performance.Experimental validation shows that ED-CHSS is more energy-efficient, lasts longer, and better protects against security threats than traditional methods. By clustering nodes based on up-to-date network data to select the best CH, the ED-CHSS approach improves energy efficiency and overall network performance. By clustering nodes based on up-to-date network data to select the best CH, the ED-CHSS approach improves energy efficiency and overall network performance. An encryption mechanism with low overhead is integrated into the CH selection process to ensure secure information transfer without negatively impacting network energy and performance. An encryption mechanism with low overhead is integrated into the CH selection process to ensure secure information transfer without negatively impacting network energy and performance. Experimental validation shows that ED-CHSS is more energy-efficient, lasts longer, and better protects against security threats than traditional methods. Experimental validation shows that ED-CHSS is more energy-efficient, lasts longer, and better protects against security threats than traditional methods. Research novelty The proposed ED-CHSS system integrates energy efficiency, trust assessment, and dynamic scheduling into a single cluster-head selection system. The most important novelty is that trust-aware decision-making is dynamically coupled with real-time scheduling and adaptive weighting, enabling the system to respond efficiently to network changes and malicious activity. Related works This section provides a short overview of research exploring possible solutions to more sustainable and safer WSNs. This section covers many topics, including clustering the CH selection, implementing security in networks, providing adaptable networks in real time, a more extensive network, and more. Currently, researchers in this field have addressed issues such as large-scale, dynamic WSNs, longer networks, secure data transfer, and energy consumption. Cluster head selection algorithms in wireless sensor networks The authors in this article evaluate various methods for generating clusters and cluster heads in wireless sensor networks. Researchers are looking for less energy-intensive clustering methods, such as LEACH11. The issues of scalability in WSNs and energy consumption in different scenarios have to be solved. The authors highlight the need to use flexible approaches to boost network capacity. This study proposes an Improved Energy-efficient Sparrow Search Algorithm (IESSA) to select CHs. The goal of the algorithm is to extend the network’s longevity while using less energy than alternatives12. The study demonstrated that current cluster head (CH) selection methods are ineffective in large networks. The IESSA performs better at energy savings and network performance than traditional approaches. Energy efficiency and network lifetime in WSNs The current study advocates a new method to prolong the life of WSNs. The more energy a network consumes, the longer the network can endure. The proposed model employs energy-specific algorithms to reduce energy consumption13. The findings indicate that the proposed model extends the life of the WSN and provides a more dependable means of transferring information across networks. The findings also discuss the importance of conserving energy and present examples of how real-world implementations have enhanced network stability. The study has discussed a clustering method that requires less energy to prolong the life of the WSN and related topics. The protocol employs an alternative method for forming clusters that changes based on the network state and operates in an adaptive clustering environment14. The study has presented empirical evidence indicating that the methods discussed are appropriate for large-scale deployments and real-time requirements, resulting in lower energy consumption and better overall WSN performance. Security mechanisms in wireless sensor networks Regarding to a trust management-based dynamic security for WSNs, this paper elaborates the shortcomings of the classical solutions. It provides a trustful solution to detect attack opportunities in real time15. The authors think that this is a suitable method for large WSNs with limited capabilities, as it makes it more challenging for attackers to hijack the WSN without consuming too much energy. The authors of the review are mainly interested in the protection of WSNs by ML.. Their review outlines significant limitations due to processing costs and difficulties with live utilization applications16. The speaker’s general conclusion is that machine learning models can be used to detect suspicious behavior and minimize invasions. The authors point out that ML-based solutions can improve security with minimal impact on performance and energy requirements. Real-time adaptability in WSNs The Energy-efficient Dynamic Adaptive State-based Scheduling (EDASS) method for WSNs17 is shown, and it uniquely adapts the scheduling of autonomous nodes to the changing characteristics of the network. The authors demonstrate that EDASS is a network system that requires less energy while improving network performance, especially when the network is dynamic. Since sensor networks will continue to grow in scale without adequate awareness in real applications, this system helps facilitate both efficiency and reliability while also adapting in real time. This paper discusses considerations for media access control (MAC) protocols, with recommendations focused on AdaptiveHART, an IIoT standard. This protocol is designed to adapt to current traffic conditions, leveraging WSNs for industrial use and focusing on resource use18. They found that AdaptiveHART is ideally suited for industrial IoT applications where network and energy stability, and time, are extremely important, as it offers lower latency and greater reliability for information transmission. Scalability and performance challenges in large-scale WSNs The hierarchical routing and clustering approach, a method used to manage large WSNs, has been discussed in this paper. Through the data routing and cluster head creation processes within a protocol, scalability and reliability issues are addressed19. The method has been proven to provide energy savings, as well as data routing accuracy. This can be useful for large-scale networks of sensors that cover a wide range of different operational contexts and have a higher communication burden. The purpose of the paper is based on the fact that Time-Slotted Channel Hopping (TSCH) is a way to make software defined industrial WSNs more scalable. It addresses network congestion and scalability, offers a scalable protocol and describes how it operates in large industrial settings20. Their results conclude that TSCH and SDNs effectively improve network performance, enhance communication reliability, and balance load in challenging industrial environments. The studies examine issues with wireless sensor networks, such as energy consumption, scalability and security. They offered recommendations for improving networks to achieve better longevity and reliability. The studies report using adaptive clustering, dynamic scheduling, and machine learning-based security to improve network performance. These recommendations can help this proposed system, ED-CHSS, achieve a robust and reliable framework, as shown in Table 1.Table 1Related works summary.ReferencesEnvironmentAlgorithmPower schedulingSecurityEnergy efficiencyRaj et al. (2022)WSNCluster head selection methods✓XMediumKathiroli and Selvadurai (2022)WSNImproved sparrow search algorithm✓XHighVijayalakshmi et al. (2025)WSNNetwork lifetime maximizationXXVery highNanthini et al. (2021)WSNEnergy efficient clustering protocol✓XMediumZheng et al. (2021)WSNTrust management security mechanismX✓MediumAhmad et al. (2022)WSNMachine learning for securityX✓LowKhan et al. (2022)WSNEDASS (energy-efficient dynamic and adaptive state-based scheduling)✓XMediumMoon et al. (2022)Industrial IoTAdaptiveHART real-time MAC protocolXXLowSingh and Singh (2021)WSNHierarchical clustering and routing✓XHighOrozco-Santos et al. (2022)Industrial WSNScalability enhancement over TSCHXXMedium Table 1Related works summary.ReferencesEnvironmentAlgorithmPower schedulingSecurityEnergy efficiencyRaj et al. (2022)WSNCluster head selection methods✓XMediumKathiroli and Selvadurai (2022)WSNImproved sparrow search algorithm✓XHighVijayalakshmi et al. (2025)WSNNetwork lifetime maximizationXXVery highNanthini et al. (2021)WSNEnergy efficient clustering protocol✓XMediumZheng et al. (2021)WSNTrust management security mechanismX✓MediumAhmad et al. (2022)WSNMachine learning for securityX✓LowKhan et al. (2022)WSNEDASS (energy-efficient dynamic and adaptive state-based scheduling)✓XMediumMoon et al. (2022)Industrial IoTAdaptiveHART real-time MAC protocolXXLowSingh and Singh (2021)WSNHierarchical clustering and routing✓XHighOrozco-Santos et al. (2022)Industrial WSNScalability enhancement over TSCHXXMedium Related works summary. Even though current protocols like LEACH, HEED, DEEC and Secure-CH focus on different components of clustering, they lack a single protocol that combines adaptive weighting, dynamic trust assessment and schedule conscious cluster head (CH) selection in one decision making model. In particular, HEED and DEEC concentrate on energy-conscious CH selection, but do not include security-conscious trust adaptation, whereas Secure-CH adds trust filtering, but does not optimize multi-criteria weights and scheduling on a dynamic basis. Conversely, ED-CHSS proposes an architecture of strong coupling where (i) energy, communication cost, and trust are jointly determined by an adaptive weighting scheme, (ii) trust assessment is dynamically maintained and it directly affects CH choice, and (iii) transmission timing is incorporated in the clustering decision making. The integrated interplay between layers enables ED-CHSS to realize energy efficiency, resistance against malicious nodes, and stability in communication, which three aspects are not achievable by the current protocols. Proposed ED-CHSS methodology The goal of ED-CHSS is to enhance the performance of WSN to solve key issues like capacity, secure messaging, dynamic scheduling, and energy efficiency. The system utilizes intelligent algorithms to guarantee protected communication, energy effective information exchange and most effective cluster heads choosing, all of which will add to network durability and overall performance. The symbols used in the equations are given in the symbol Table 2.Table 2Symbol.SymbolDescription\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( \Omega \right)$$\end{document}Set of all sensor nodes\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {w\_i} \right)$$\end{document}Sensor node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( \Gamma \right)$$\end{document}Set of selected cluster heads\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( N \right)$$\end{document}Total number of nodes\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( K \right)$$\end{document}Number of clusters\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( k \right)$$\end{document}Packet size\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{i} } \right)$$\end{document}Residual energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {max} \right\}}} } \right)$$\end{document}Maximum initial energy\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {tx,i} \right\}}} } \right)$$\end{document}Transmission energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {rx,i} \right\}}} } \right)$$\end{document}Reception energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {idle,i} \right\}}} } \right)$$\end{document}Idle energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {sleep,i} \right\}}} } \right)$$\end{document}Sleep mode energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {aggr,i} \right\}}} } \right)$$\end{document}Data aggregation energy\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {round} \right\}}} } \right)$$\end{document}Total energy consumed per round\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {consumed,i} \right\}}} } \right)$$\end{document}Energy consumed by node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {elec} \right\}}} } \right)$$\end{document}Electronic energy per bit\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\varepsilon_{{\left\{ {fs} \right\}}} } \right)$$\end{document}Free-space amplifier constant\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\varepsilon_{{\left\{ {mp} \right\}}} } \right)$$\end{document}Multi-path amplifier constant\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {d_{0} } \right)$$\end{document}Threshold transmission distance\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {d_{i} } \right)$$\end{document}Average communication distance of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {d_{{\left\{ {ij} \right\}}} } \right)$$\end{document}Distance between node (i) and node (j)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {x_{i} , y_{i} } \right)$$\end{document}Coordinates of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{i} } \right)$$\end{document}Trust value of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {DT_{i} } \right)$$\end{document}Direct trust\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {IT_{i} } \right)$$\end{document}Indirect trust\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {TD_{i} } \right)$$\end{document}Temporal decay factor\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{{\left\{ {th} \right\}}} } \right)$$\end{document}Trust threshold\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {PF_{i} } \right)$$\end{document}Packet forwarding ratio\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {DR_{i} } \right)$$\end{document}Packet delivery ratio\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {CR_{i} } \right)$$\end{document}Cooperation rate\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {w_{1} , w_{2} , w_{3} } \right)$$\end{document}Weight coefficients\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Score\left( i \right)} \right)$$\end{document}Cluster head selection score\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\alpha , \beta , \gamma } \right)$$\end{document}Weighting parameters\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {S_{i} } \right)$$\end{document}Scheduling probability\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{i}^{{\left\{ {slot} \right\}}} } \right)$$\end{document}Time slot assigned to node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Traffic_{i} } \right)$$\end{document}Incoming traffic of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Capacity_{i} } \right)$$\end{document}Processing capacity of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {L_{i} } \right)$$\end{document}Load factor of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {SF_{i} } \right)$$\end{document}Scheduling factor\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {LBE} \right)$$\end{document}Load balancing efficiency\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( \rho \right)$$\end{document}Node density\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\rho_{{\left\{ {max} \right\}}} } \right)$$\end{document}Maximum node density\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {avg} \right\}}} } \right)$$\end{document}Average residual energy\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Q_{l} } \right)$$\end{document}Link quality metric\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{{\left\{ {dyn} \right\}}} } \right)$$\end{document}Dynamic threshold\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{0} } \right)$$\end{document}Base threshold\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( t \right)$$\end{document}Current time\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {t_{{\left\{ {last} \right\}}} } \right)$$\end{document}Last interaction time(λ)Decay constant\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {H\left( \cdot \right)} \right)$$\end{document}Hash function(Throughput)Data transmission rate(P)Power consumption\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Risk_{i} } \right)$$\end{document}Risk level of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {C_{{\left\{ {enc} \right\}}} } \right)$$\end{document}Encryption computational cost\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {C_{{\left\{ {dec} \right\}}} } \right)$$\end{document}Decryption computational cost\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {C_{{\left\{ {comp} \right\}}} } \right)$$\end{document}Computational complexity\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {M_{{\left\{ {total} \right\}}} } \right)$$\end{document}Total memory usage\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {crypto} \right\}}} } \right)$$\end{document}Cryptographic energy consumption\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( A \right)$$\end{document}Adaptability metric\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Successful_{{\left\{ {Adaptations} \right\}}} } \right)$$\end{document}Number of successful adaptations\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Total_{{\left\{ {Changes} \right\}}} } \right)$$\end{document}Total number of changes Table 2Symbol.SymbolDescription\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( \Omega \right)$$\end{document}Set of all sensor nodes\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {w\_i} \right)$$\end{document}Sensor node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( \Gamma \right)$$\end{document}Set of selected cluster heads\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( N \right)$$\end{document}Total number of nodes\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( K \right)$$\end{document}Number of clusters\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( k \right)$$\end{document}Packet size\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{i} } \right)$$\end{document}Residual energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {max} \right\}}} } \right)$$\end{document}Maximum initial energy\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {tx,i} \right\}}} } \right)$$\end{document}Transmission energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {rx,i} \right\}}} } \right)$$\end{document}Reception energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {idle,i} \right\}}} } \right)$$\end{document}Idle energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {sleep,i} \right\}}} } \right)$$\end{document}Sleep mode energy of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {aggr,i} \right\}}} } \right)$$\end{document}Data aggregation energy\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {round} \right\}}} } \right)$$\end{document}Total energy consumed per round\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {consumed,i} \right\}}} } \right)$$\end{document}Energy consumed by node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {elec} \right\}}} } \right)$$\end{document}Electronic energy per bit\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\varepsilon_{{\left\{ {fs} \right\}}} } \right)$$\end{document}Free-space amplifier constant\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\varepsilon_{{\left\{ {mp} \right\}}} } \right)$$\end{document}Multi-path amplifier constant\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {d_{0} } \right)$$\end{document}Threshold transmission distance\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {d_{i} } \right)$$\end{document}Average communication distance of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {d_{{\left\{ {ij} \right\}}} } \right)$$\end{document}Distance between node (i) and node (j)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {x_{i} , y_{i} } \right)$$\end{document}Coordinates of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{i} } \right)$$\end{document}Trust value of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {DT_{i} } \right)$$\end{document}Direct trust\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {IT_{i} } \right)$$\end{document}Indirect trust\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {TD_{i} } \right)$$\end{document}Temporal decay factor\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{{\left\{ {th} \right\}}} } \right)$$\end{document}Trust threshold\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {PF_{i} } \right)$$\end{document}Packet forwarding ratio\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {DR_{i} } \right)$$\end{document}Packet delivery ratio\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {CR_{i} } \right)$$\end{document}Cooperation rate\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {w_{1} , w_{2} , w_{3} } \right)$$\end{document}Weight coefficients\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Score\left( i \right)} \right)$$\end{document}Cluster head selection score\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\alpha , \beta , \gamma } \right)$$\end{document}Weighting parameters\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {S_{i} } \right)$$\end{document}Scheduling probability\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{i}^{{\left\{ {slot} \right\}}} } \right)$$\end{document}Time slot assigned to node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Traffic_{i} } \right)$$\end{document}Incoming traffic of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Capacity_{i} } \right)$$\end{document}Processing capacity of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {L_{i} } \right)$$\end{document}Load factor of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {SF_{i} } \right)$$\end{document}Scheduling factor\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {LBE} \right)$$\end{document}Load balancing efficiency\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( \rho \right)$$\end{document}Node density\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\rho_{{\left\{ {max} \right\}}} } \right)$$\end{document}Maximum node density\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {avg} \right\}}} } \right)$$\end{document}Average residual energy\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Q_{l} } \right)$$\end{document}Link quality metric\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{{\left\{ {dyn} \right\}}} } \right)$$\end{document}Dynamic threshold\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {T_{0} } \right)$$\end{document}Base threshold\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( t \right)$$\end{document}Current time\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {t_{{\left\{ {last} \right\}}} } \right)$$\end{document}Last interaction time(λ)Decay constant\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {H\left( \cdot \right)} \right)$$\end{document}Hash function(Throughput)Data transmission rate(P)Power consumption\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Risk_{i} } \right)$$\end{document}Risk level of node (i)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {C_{{\left\{ {enc} \right\}}} } \right)$$\end{document}Encryption computational cost\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {C_{{\left\{ {dec} \right\}}} } \right)$$\end{document}Decryption computational cost\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {C_{{\left\{ {comp} \right\}}} } \right)$$\end{document}Computational complexity\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {M_{{\left\{ {total} \right\}}} } \right)$$\end{document}Total memory usage\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {E_{{\left\{ {crypto} \right\}}} } \right)$$\end{document}Cryptographic energy consumption\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( A \right)$$\end{document}Adaptability metric\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Successful_{{\left\{ {Adaptations} \right\}}} } \right)$$\end{document}Number of successful adaptations\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Total_{{\left\{ {Changes} \right\}}} } \right)$$\end{document}Total number of changes Symbol. Step 1: overview of the proposed system ED-CHSS The suggested ED-CHSS could improve the performance of wireless sensor networks, and this system takes into account aspects such as security, scalability, data transfer, and energy use. The system has four main parts: trust management, dynamic scheduling, picking the cluster head that uses the least energy, and making it easier for huge networks to grow. Figure 1 shows that the system’s smart management of network resources, which shifts in real time, extends network lifespan and improves communication efficiency. The ED-CHSS design modifies the node schedule based on changes in network traffic and popularity, and it picks cluster commanders on the fly based on how much energy they use21. It uses a trust-based security system to further protect against malicious behavior and data manipulation. Breaking a large WSN into smaller, easier-to-manage sub-networks can improve scalability and make the network faster and more reliable.1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{total} = \mathop \sum \limits_{x = i}^{1} \left[ {E_{tx} ,i + E_{rx} ,i + E_{idle} ,i} \right]$$\end{document}Fig. 1Overview of the proposed system. Overview of the proposed system. Equation (1) represents the total energy consumption of the wireless sensor network. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{tx} ,i, E_{rx} ,i , E_{idle} ,i$$\end{document} represent transmission, reception, and idle energy of node i, respectively. N is the number of nodes. It records total energy consumption across all operational states and helps measure the network’s lifetime and efficiency. Dynamic threshold adaptation function 2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{dyn} = T_{0} \times \left[ {1 + \alpha \left( {\frac{\rho }{{\rho_{max} }}} \right) + \beta \left( {\frac{{E_{avg} }}{{E_{max} }}} \right) + \gamma \left( {\frac{{Q_{l} }}{{Q_{max} }}} \right)} \right]$$\end{document} The dynamic threshold \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{dyn}$$\end{document} is adapted based on network conditions. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{0}$$\end{document} is the base threshold, ρ is the density of nodes, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{avg}$$\end{document} is the average residual energy of the nodes, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Q_{l}$$\end{document} is the quality of links (e.g., packet success rate). \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho_{max}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{max}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Q_{max}$$\end{document} are normalization factors, and α, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta , \gamma$$\end{document} are weighting factors, so that α, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta , \gamma = 1.$$\end{document} The threshold increases with node density to limit the selection of cluster heads and prevent high overheads, as shown in Eq. (2). The greater the residual energy, the higher the threshold, in preference of stronger nodes. The threshold is also increased by the higher quality of links, which ensures good communication. This dynamic formulation enables the system to adapt dynamically to changing network conditions, thus ensuring greater stability, efficiency and robustness.Algorithm 1Dynamic cluster head selection and scheduling (D-CHSS) algorithm Dynamic cluster head selection and scheduling (D-CHSS) algorithm The method determines how much work each network node responsible for sensing must perform. Selects cluster heads according to the load and traffic status. When the load of a node drops below a certain threshold, the node is elected as cluster head22. If its load is equal to the best load already in use, it will be added to the current group of heads. With algorithm 1, it ensures network management can grow and energy is sent out effectively. Extended energy consumption model 3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{total} = E_{tx} + E_{rx} + E_{aggr} + E_{idle} + E_{sleep}$$\end{document} The overall energy usage of a sensor node is defined as the sum of transmission \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{tx}$$\end{document}, reception \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{rx}$$\end{document}, data aggregation \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{aggr}$$\end{document}, idle listening \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{idle}$$\end{document} and sleep mode \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{sleep}$$\end{document} energies. Transmission and reception are considered communication costs, while aggregation reduces the data processing that is redundant as indicated in Eq. (3). Idle energy is the power used when the node is not transmitting, but is on; sleep energy is the low-power state. This model gives a practical description on the behavior of the nodes and improves the accuracy of the energy-efficiency analysis in wireless sensor networks. Step 2: energy–trust aware cluster head selection in ED-CHSS In WSNs, the energy-efficient cluster-head selection technique is very crucial to enhance the overall performance and saving energy. As it considers network conditions at runtime, and selects cluster chiefs based on energy, it could lead to premature node failure and a shorter network lifetime. The basic concept of this scheme is to select cluster heads according to their remaining energy, distance from other nodes and workload of the network23. Dynamic mechanism is implemented to select cluster heads to avoid consuming too much energy in the network. Nodes that have high energy and are close to each other are shown in Fig. 2, and are weighted more. The selection of cluster heads carefully ensures that the distance between cluster head and nodes is minimum. This uses less energy for communication. This uses more energy. This approach significantly boosts the overall performance of the network by extending its lifespan, minimizing communication overhead, and optimizing data aggregation efficiency. The ED-CHSS algorithm improves the energy efficiency by utilizing the residual node energy and adds dynamic scheduling and trust-based cluster head selection. Unlike the traditional LEACH protocol, which includes a probabilistic selection mechanism, the ED-CHSS adopts the energy, distance and trust measurement based weighted decision mechanism. Also, adaptive scheduling eliminates redundant transmissions, and load-conscious clustering avoids energy imbalance, thereby greatly increasing the network lifetime.Fig. 2Representation of energy-efficient cluster head selection. Representation of energy-efficient cluster head selection. Adaptive weighting mechanism Residual energy, node density and the trust levels are used in a heuristic rule-based mechanism inspired by fuzzy logic to adaptively adjust the weighting coefficients. The input parameters, including residual energy, node density and level of trust, are translated into linguistic variables (low, medium, high). The energy efficiency, communication costs and trust are dynamically reweighted as per pre-defined fuzzy rules. The defuzzified output yields in real time the best weights, which results in a more balanced and reliable cluster head selection in case of different traffic loads, node failures, and even possible attacks. The fixed weighting is replaced by the dynamically calculated weighting values by the adaptive system according to the current network conditions. Modulation of energy, distance and trust parameters weights is performed by applying fuzzy inference to the system. This helps to improve flexibility, decision accuracy and performance stability, even in highly dynamic or hostile wireless sensor network environment. Multi-objective formulation for CH selection The cluster head selection process becomes a multi-objective problem where the energy efficiency, communication cost and reliability are taken into account. These factors ensure optimal utilization of the resources and communication between nodes that helps in selecting the optimal ones, improving the overall performance and reliability of the network in wireless sensor networks.4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f1\left( i \right) = \frac{{E_{i} }}{{E_{max} }}$$\end{document} Equation 4 is the residual energy ratio of node i, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{i}$$\end{document} is the energy left of node i and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{max}$$\end{document} is the maximum initial energy of any node. The larger the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f1\left( i \right)$$\end{document} is the more energy resources the node has, the more likely it is to be selected as the cluster head.5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f2\left( i \right) = \frac{1}{{\left( {d_{i} + \varepsilon } \right)}}$$\end{document} Equation 5 is a model of the communication cost of the node i, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$d_{i}$$\end{document} is the average distance between node i and the surrounding nodes or the base station. \varepsilon is a small constant to prevent division by zero. When distance is low \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f2\left( i \right)$$\end{document} is large, which implies low transmission energy and enhanced communication efficiency.6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f3\left( i \right) = w1 \cdot PF_{i} + w2 \cdot DR_{i} + w3 \cdot CR_{i}$$\end{document} Equation 6 is used to compute the trust index of node i, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$PF_{i}$$\end{document} denotes packet forwarding ratio, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$DR_{i}$$\end{document} denotes packet delivery ratio and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$CR_{i}$$\end{document} denotes cooperation rate and w1, w2 and w3 are weighting coefficients, w1 + w2 + w3 = 1. A stronger trust value ensures a safe and trustworthy involvement in cluster head selection. The proposed formulation combines the residual energy, communication cost and trust index in one decision model. The model is able to evaluate these parameters simultaneously, ensuring the efficient selection of cluster heads, high security and even load distribution, thus achieving maximum network life, minimum overhead and high resilience against bad node behavior.7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$CH_{i} = \alpha \left( {\frac{{E_{i} }}{{E_{max} }}} \right) + \beta \left( {\frac{1}{{d_{i} }}} \right) + \gamma T_{i}$$\end{document} Equation (7) calculates the score of the cluster head selection \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$CH_{i}$$\end{document} of node i. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{i}$$\end{document} is leftover energy, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$d_{i}$$\end{document} is the distance to adjacent nodes, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i}$$\end{document} is a trust value. α, β and γ are weighting factors. The nodes with higher scores are used as cluster heads to ensure that energy is used evenly and that communication is secure.8\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$d_{ij} = \sqrt {\left[ {\left( {x_{i} - x_{j} } \right)^{2} + \left( {y_{i} - y_{j} } \right)^{2} } \right]}$$\end{document} Euclidean distance \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$d_{ij}$$\end{document} can be obtained between nodes i and j using coordinates of nodes as shown in Eq. (8). Distance estimation plays a vital role in wireless sensor networks because transmission energy depends on distance. The distance is minimized to reduce energy use and enhance communication efficiency.9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{{tx\left( {k,d} \right)}} = \begin{array}{*{20}l} {kE_{elec} + k\varepsilon_{fs} d^{2} ,\;if\;d < d_{0} } \hfill \\ {kE_{elec} + k\varepsilon_{mp} d^{4} ,\;if\;d \ge d_{0} } \hfill \\ \end{array}$$\end{document} The Eq. (9) represents the energy needed to transmit k-bit data at a distance d as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{{tx\left( {k,d} \right)}}$$\end{document}. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{elec}$$\end{document} is electronic energy and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon_{fs}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon_{mp}$$\end{document} are amplifier constants. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$d_{0}$$\end{document} is the threshold distance. It guarantees proper modeling of both short- and long-range communication of energy consumption24. Time and communication complexity analysis The ED-CHSS algorithm is discussed to evaluate the scalability of the algorithm in large scale wireless sensor network. Complexity of time is \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O\left( {n k} \right)$$\end{document} where n is number of nodes and k is number of decision parameters (energy, distance and trust). Here, k is a constant and the complexity is \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O\left( n \right)$$\end{document}, which is linear. The complexity of communication is \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O\left( {n + m} \right)$$\end{document} where n is the number of nodes in the system and m is the number of communication links. The algorithm is based on local interaction to cluster and evaluate trust, reducing global interactions. The linear time complexity and bounded communication overhead, in general, demonstrate that ED-CHSS is efficient and can be used in large-scale deployments. Objective function The CH selection problem is formulated as an optimization problem with constraints and designed to optimize energy efficiency, communication quality and node trustworthiness. The objective function is given as:10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Maximize: F = w1 \times \left( {\frac{{E_{i} }}{{E_{max} }}} \right) - w2 \times \left( {\frac{{C_{i} }}{{C_{max} }}} \right) + w3 \times \left( {\frac{{T_{i} }}{{T_{max} }}} \right)$$\end{document} The goal is to maximize the total weighted sum of the normalized residual energy \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{i}$$\end{document}, the communication cost \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{i}$$\end{document} and the trust index \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i}$$\end{document} of each node. The higher the energy and trust, the higher the objective value, following Eq. (10). The weights \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {w1,w2,w3} \right)$$\end{document} are used to control the importance of the factors. Constraints ensure the validity of the parameters and that at least one cluster head is chosen. This formulation helps in making balanced decisions, which in turn increases the efficiency of energy, reliability and security of wireless sensor networks. Such constraints guarantee that the objective function will be contained within the normalized values, that feasible weight allocations will be available and that there will be sufficient choices of cluster heads to have a stable network that operates. Step 3: adaptive energy-aware scheduling mechanism in ED-CHSS Dynamic scheduling is an important technique to enhance the performance of WSN and energy saving. The ED-CHSS system always knows what is happening in the network, such as traffic, node power, network loading and so on. It can use this to optimize the use of bandwidth and prevent congestion by changing the node scheduling at runtime. To save energy and avoid the network being too congested, nodes send data at certain times as shown in Fig. 3. For example, nodes with more energy are given shorter transmission times or even put into sleep mode to save energy. Conversely, nodes having less energy will be scheduled to communicate more frequently. This adaptive method makes sure that resources are allocated fairly, optimizes bandwidth usage and decreases the delay in transmission. Dynamic scheduling technique dynamically shifts network traffic to optimize throughput, reduce latency, and save energy25.11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_{i} = \frac{{E_{i} }}{{\mathop \sum \nolimits_{j = i}^{1} E_{j} }}$$\end{document}Fig. 3Dynamic scheduling based on network state overview. Dynamic scheduling based on network state overview. The probability of the scheduling \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_{i}$$\end{document} of node i is given by the Eq. (11). Nodes that have greater residual energy \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{i}$$\end{document} are more likely to get transmitted. This approach balances the load sharing and avoids quick exhaustion of the low energy nodes, which improves the network lifetime.12\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{i} = \frac{{Traffic_{i} }}{{Capacity_{i} }}$$\end{document} Equation (12) measures the load \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{i}$$\end{document} on node i, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Traffic_{i}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Capacity_{i}$$\end{document} are the incoming data and the node’s processing capability, respectively. It helps balance the network by preventing overloading of specific nodes, enabling efficient data processing and minimizing congestion.13\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{round} = \Sigma \left( {i = 1 to N} \right)E_{consumed} ,i$$\end{document} The amount of energy used during a single round \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{round}$$\end{document} of communication is calculated in Eq. (13). \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{consumed} ,i$$\end{document} is the power consumed by node i. It can be effective for analyzing the rate at which energy is exhausted and for estimating the lifetime of the networks.14a\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i} = \frac{{E_{i} }}{\max \left( E \right)}$$\end{document}14b\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau_{i} = f\left( {E_{i} } \right)$$\end{document} Equation (14) assigns transmission time slots \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i}$$\end{document} depending on the node energies E. The nodes with more residual energy \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{i}$$\end{document} are given a longer transmission time. This provides effective scheduling and a balanced energy use across the network. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i}$$\end{document} is the trust value of node i in this work, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau_{i}$$\end{document} is the transmission time slot of node i. Step 4: scalable clustering strategy for large-scale WSNs in ED-CHSS The main purpose of ED-CHSS’s secure data transmission method is to ensure that all data sent over the wireless sensor network is authentic, complete, and transmitted privately. In a trust management system, nodes are assigned trust values based on their past behavior and interactions with other nodes26. Trust levels are used to prevent bad or untrustworthy nodes from participating in the data transmission process shown in Fig. 4. This protects against man-in-the-middle attacks, data tampering, and denial-of-service attacks. A trust-based strategy is employed to keep data safe. Nodes with high trust levels are allowed to communicate sensitive data. This strategy reduces the likelihood that infected nodes will disrupt the network. The system has trust-evaluation algorithms that adjust trust levels based on node behavior. This makes communication in the WSN more reliable and secure by ensuring that data remains protected. Lightweight cryptography uses symmetric-key encryption to reduce computational overhead on resource-constrained WSN nodes. The Cluster Head (CH) generates and distributes session keys to member nodes during cluster formation. Periodically, keys are changed to improve security. Safe key communication is carried out using pre-shared keys and trust-based authentication, providing confidentiality, integrity and efficient communication within the cluster.Fig. 4Path diagram for secure data transmission with trust management method. Path diagram for secure data transmission with trust management method. A weighted-average model is used to calculate the Trust Index, which is based on the packet-forwarding success rate, the delivery rate, and the level of cooperation. The parameters have weights based on their significance, and the sum of these scores determines the node’s trustworthiness. Trust values are dynamically updated after each communication round, allowing precise identification of unreliable or malicious nodes at minimal computational cost. Threat model The ED-CHSS system assumes a semi-trusted wireless sensor network, in which nodes can be compromised to become malicious. Attackers can drop packets, modify routing information, and spoof identities, but are constrained by energy and computational resources. The base station is considered secure, and nodes communicate using pre-shared keys. The model accounts for selective forwarding, sinkhole and Sybil attacks as the main threats. These attacks interfere with packet transmission, routing, and network topology. To alleviate them, the suggested trust mechanism considers node actions in terms of forwarding and cooperation. Bad nodes are identified and assigned very low trust scores, and they are not considered when selecting cluster heads or communicating. Trust assessment model To achieve reliable and secure communication, the ED-CHSS model uses a multi-dimensional trust assessment model that incorporates both direct trust, indirect trust and temporal decay to reflect the dynamic behavior of nodes. Direct trust (DT) Direct trust \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$DT_{i}$$\end{document} is calculated based on the behavior node i exhibited when forwarding packets. It can be described as the number of packets sent that were forwarded successfully, and to total number of sent packets:15\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$DT_{i} = \frac{{P_{forward} ,i}}{{P_{total} ,i}}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{forward} ,i$$\end{document} is the correct number of packets forwarded, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{total} ,i$$\end{document} is the number of packets the node i handled. This is an indication of the node’s trustworthiness in data transmission, as defined by Eq. (15). Indirect trust (IT) The indirect trust \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$IT_{i}$$\end{document} is obtained based on recommendations from adjacent nodes. It summarizes the trust opinion of a group of neighboring nodes N.16\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$IT_{i} = \left( {\frac{1}{{\left| {N_{i} } \right|}}} \right)* \Sigma \left( {j \in N_{i} } \right)T_{j\left( i \right)}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{j\left( i \right)}$$\end{document} is the trust in node i as per node j and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left| {N_{i} } \right|$$\end{document} is the number of neighbors. This is used to identify bad nodes that behave inconsistently across the network, as defined by Eq. (16). Temporal decay factor (TD) A time decay factor \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$TD_{i}$$\end{document} is added to capture nodes’ dynamic behavior.17\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$TD_{i} = e^{{ - \lambda \left( {t - t_{last} } \right)}}$$\end{document}where λ is the decay constant, t is the present time, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$t_{last}$$\end{document} is the last interaction time. This ensures that the information from the old trusts is phased out by Eq. (17). Overall trust score The last trust value calculated is:18\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i} = \alpha DT_{i} + \beta \left( {IT_{i} \times TD_{i} } \right)$$\end{document}where the weighting factors, α and β, have values 1 and 2, respectively. This expression is a reaction to local observations and network-wide suggestions and accounts for changes in behavior over time using Eq. (18). Cryptographic overhead analysis The cryptographic cost of ED-CHSS is assessed based on its computational complexity, memory usage, and the energy overhead associated with encryption and decryption.19\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{enc\left( n \right)} = O\left( n \right), C_{dec\left( n \right)} = O\left( n \right)$$\end{document} Encryption \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{enc\left( n \right)}$$\end{document} and decryption \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{dec\left( n \right)}$$\end{document} are linear in the computational complexity with respect to the packet size n. This guarantees scalability and applicability to real-time communication in resource-constrained wireless sensor networks, as shown in Eq. (19).20\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{total} = M_{key} + M_{algo} \approx 1 - 2 KB$$\end{document} The nodes need memory \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{total}$$\end{document} to store cryptography keys \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{key}$$\end{document} and encryption/decryption algorithm \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{algo}$$\end{document}. The memory overhead is also small, so that it can be used in low-memory sensor nodes using Eq. (20).21\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{crypto} = E_{enc} + E_{dec} \approx \left( {0.05 - 0.1} \right) \times E_{tx}$$\end{document} Cryptographic operations \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{crypto}$$\end{document} require a total amount of encryption energy \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{enc}$$\end{document} and decryption energy \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{dec}$$\end{document}. Equation 21 is estimated to be 5–10% of the transmission energy \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{tx}$$\end{document}, meaning that security enhancements do not add a significant amount of energy overhead. Equation 9 is how to figure out how trustworthy nodes tr in the network are \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$nt_{w}$$\end{document} are based on what they have done and how they have interacted N with each other in the past.22\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$tr = \left\langle {\left( {nt_{w} - N} \right) \left( {Ts} \right),t_{a} } \right\rangle + pf\left\langle {\left( {T_{s} - if} \right)\left( {tu} \right),u_{n} } \right\rangle$$\end{document} The trust score \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {Ts} \right)$$\end{document} is based on three primary things: data aggregation tasks \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$t_{a}$$\end{document}, cooperation, and packet forwarding pf. By giving trust scores \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{s}$$\end{document} to nodes, this Eq. (22) makes data safer and keeps bad actors from interfering if with the network by letting trustworthy nodes \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {tu} \right)$$\end{document} communicate27.23\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i} \ge T_{th}$$\end{document} The participation in the network is secure due to Eq. (23). They can only communicate with nodes whose trust value \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i}$$\end{document} exceeds threshold \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{th}$$\end{document}. This eliminates the risk of malicious or untrustworthy nodes influencing the system.24\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H\left( {data_{send} } \right) = H\left( {data_{receive} } \right)$$\end{document} The integrity of data is checked with an Eq. (24) that uses a hash function H. When the transmitted \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {data_{send} } \right)$$\end{document} and received \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {data_{receive} } \right)$$\end{document} When hash values are equal, the data is considered secure and uncorrupted. It prevents data tampering during communication. Step 5: scalability enhancement in large-scale networks In a large-scale WSN, hundreds or thousands of sensor nodes are deployed, which can create several issues, including high communication overhead, network congestion, and resource depletion, highlighting the need for scalability. The ED-CHSS system overcomes this challenge by dividing the network into smaller, more manageable clusters. By having communication concentrated within each cluster, as shown in Fig. 5, the cluster heads make it easier for the network to communicate. This hierarchical system allows for the transmission of less data across the entire network, for more effective routing of data and for load balancing. Based on the traffic load and size of the network, the cluster heads and the formation of clusters changes dynamically in real time. This architecture will ensure low latency, energy balance and performance while maintaining the scalability in large scale deployments of WSNs28.25\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C = \frac{N}{K}$$\end{document}Fig. 5Scalability enhancement in large-scale networks. Scalability enhancement in large-scale networks. Network scalability is defined in Eq. (25). N is the total number of nodes, and K is the number of clusters C. It guarantees equal cluster formation, enhancing efficiency and communications in large-scale networks. Trust management implementation \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i} = w_{d} \times T_{direct} + w_{i} \times T_{indirect}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{direct} = \frac{{S_{i} }}{{\left( {S_{i} + F_{i} } \right)}}$$\end{document} 26\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{indirect} = \left( {\frac{1}{{N_{i} }}} \right) \times \mathop \sum \limits_{j} \in N_{i} T_{j}$$\end{document} A node i has a trust score \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i}$$\end{document} which is a weighted sum of direct trust \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{direct}$$\end{document} and indirect trust \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{indirect}$$\end{document}. The computation of direct trust is based on successful \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_{i}$$\end{document} and failed \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$F_{i}$$\end{document} packet forwarding events using Eq. (26). The average recommendation of neighbors \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$N_{i}$$\end{document} is known as the indirect trust. Local observations and neighbor opinions are balanced with weights \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$w_{d}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$w_{i}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$where (w_{d} + w_{i} = 1$$\end{document}). Nodes periodically update \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i}$$\end{document}, nodes with \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i}$$\end{document} less than 0.7 (e.g., 0.7) are not considered in CH selection and routing. For final normalized trust score in CH selection and routing with Eq. (26). Trust values are suitable for cluster head selection and routing decisions. Flow diagram The initial stage in the ED-CHSS system is to put sensor nodes in a network. Then, based on energy and network needs, cluster heads are chosen, as illustrated in Fig. 6. Dynamic scheduling adjusts node transmission based on current conditions to ensure energy is used efficiently. Trust-based security methods protect data in interpersonal interactions. Finally, the equipment enhances versatility by enabling large networks to run faster and breaking them into smaller, easier-to-manage groups.Fig. 6Overall flow diagram of this ED-CHSS system. Overall flow diagram of this ED-CHSS system. The ED-CHSS system improves WSN by selecting cluster heads that consume less energy, making it easier to add new nodes, employing trust-based security, scheduling jobs in real time, and more. Adaptive resource allocation and adaptation to network conditions use less energy, keep data safe, and improve communication, even in large networks. This plan will make the network more stable and better prepared to handle varied conditions. Implementation details Dataset description The Wireless Sensor Networks Dataset from the Kaggle project provides details on the network’s layout, the energy utilization of nodes that collect data, and how they communicate with one another. In this study secure data advancement, dynamic scheduling and picking cluster heads with lesser energy is tested in the dataset. It provides an opportunity to experiment and adapt the proposed system in a real situation, and studies its performance in terms of data handling and eco-friendliness29. Kaggle-based WSN data are used as a starting point to establish node characteristics, including energy states, communication strategies and network structure. Malicious behavior is synthetically added to the NS-3/OMNeT + + environment to simulate dynamic attack scenarios, including sinkhole and selective forwarding. In particular, a set of nodes is dynamically configured to change packet forwarding behavior, selectively drop packets, or misdirect traffic across a series of simulation rounds, simulating a real-time attack scenario. Simulation tools The simulations were run on an ordinary computing system with an Intel i7/i5 processor (or a similar one), 16–32 GB of RAM, and a Linux-based operating system. It was implemented using the NS-3 and OMNeT + + simulation tools. Because of the event-driven, computationally sparse nature of simulations of wireless sensor networks, there was no need to use specialized GPU hardware to implement the proposed ED-CHSS algorithm. A portion of the nodes are set to be malicious, 5% to 20% of the total network, and the main results are given when the attack intensity is at the moderate level of 10%. Malicious behavior is introduced by modifying the packet handling at the node level during run-time: selective forwarding nodes drop 30–70% of packets, blackhole nodes drop all packets, sinkhole nodes advertise falsified high quality routes that attract traffic, the Sybil attackers duplicate multiple identities using duplicated node IDs and the wormhole attack is emulated by a low latency tunnel between colluding nodes. The simulation runs for 3000 rounds—attacks are turned on after a period of stabilization (round 500) and continue until the end of the simulation. Static and dynamic attack scenarios are used, where in dynamic attacks some nodes will switch from normal to unsafe function at regular times to mimic stealthy attackers. The proposed trust-based mechanism continuously observes the forwarding behavior, delivery success and cooperation rate to find the anomalies, and the nodes that are identified malicious are given low trust value and excluded from the cluster head selection and routing. Because this parameterized configuration guarantees that the results for the packet delivery ratio and detection rate are reproducible and represent realistic adversarial conditions. Hardware configuration The proposed ED-CHSS framework lacks any machine learning components for trust assessment. The index of trust is measured by a lightweight model that uses packet delivery, forwarding behavior, and cooperation metrics. Consequently, GPU acceleration was not necessary, and all simulations were run on the CPU in the NS-3 and OMNeT + + environments. Table 3 presents an overview of the network setup, energy parameters, algorithm parameters, and tools, which guarantee easy reproducibility. It offers a systematic perspective of the conditions of simulation to test the ED-CHSS approach.Table 3Implementation and simulation configuration.CategoryParameterValue/descriptionNetwork setupNumber of nodes50–500Deployment area100 × 100–500 × 500 m2Node distributionRandomBase station positionOutside sensing areaEnergy modelInitial energy1.0 JE_elec50 nJ/bitAmplifier energy100 pJ/bit/m2Data aggregation energy5 nJ/bitCommunicationPacket size4000 bitsTransmission range50–100 mAlgorithm settingsTrust threshold0.7CH selection criteriaEnergy, distance, trustScheduling mechanismDynamicSimulation controlSimulation rounds3000Number of runs10–20Tools usedProgramming languagePythonLibrariesNumPy, Matplotlib, PandasTraffic modelPeriodic data transmission(CBR—Constant Bit Rate)Data generation rate1 packet per node per roundPacket intervalFixed interval per round Table 3Implementation and simulation configuration.CategoryParameterValue/descriptionNetwork setupNumber of nodes50–500Deployment area100 × 100–500 × 500 m2Node distributionRandomBase station positionOutside sensing areaEnergy modelInitial energy1.0 JE_elec50 nJ/bitAmplifier energy100 pJ/bit/m2Data aggregation energy5 nJ/bitCommunicationPacket size4000 bitsTransmission range50–100 mAlgorithm settingsTrust threshold0.7CH selection criteriaEnergy, distance, trustScheduling mechanismDynamicSimulation controlSimulation rounds3000Number of runs10–20Tools usedProgramming languagePythonLibrariesNumPy, Matplotlib, PandasTraffic modelPeriodic data transmission(CBR—Constant Bit Rate)Data generation rate1 packet per node per roundPacket intervalFixed interval per round Implementation and simulation configuration. Tuning of baseline protocol and parameters To be fair in comparison, the baseline protocols such as LEACH, DEEC, HEED and Secure-CH are executed in the same simulation environment and parameter settings. Each protocol uses the same network parameters, such as node distribution, initial energy, packet size, and transmission range. In LEACH, the probabilistic thresholds for cluster head selection determine the percentage of cluster heads. The DEEC selection of cluster heads varies depending on residual and average energy. HEED selects cluster heads based on residual energy and communication cost. Simple trust-based filtering is implemented in Secure-CH. Every protocol is optimized using generally recommended parameter values from the literature, and several simulation runs are carried out to obtain stable, unbiased results. A paired t-test is used to assess the statistical significance of the difference between the ED-CHSS and baseline protocols based on the outcomes of 10–20 independent simulation runs. The test is used to evaluate key performance indicators such as energy consumption, network lifetime and packet delivery ratio. The acquired p-values (< 0.05) indicate that the changes ED-CHSS has received are statistically significant and not due to random variation. Component-wise contribution analysis Three important elements affect the performance improvement of ED-CHSS. The most significant contribution is the trust mechanism, which excludes malicious or unreliable nodes, thereby enhancing the packet-delivery ratio and the detection rate. The energy model ensures balanced energy consumption across nodes, prolonging the network’s lifespan. The adaptive scheduling mechanism will increase throughput and reduce congestion by optimally varying transmission in response to network conditions. Trade-off analysis The incorporation of trust assessment enhances network security and reliability, but it adds some computational overhead due to the constant updates of trust. Adaptive scheduling, on the other hand, enhances responsiveness to fluctuating network traffic, efficiency, and scalability while potentially leading to slight fluctuations in the system in the face of highly dynamic traffic. Moreover, the proposed ED-CHSS achieves a desirable security, energy efficiency, and stability. Experimental analysis and parameters The experimental analysis in this section investigates the performance of the ED-CHSS system in terms of scalability, energy efficiency, secure communication and dynamic scheduling. The research analyzes various factors such as the length of a network, security of the data and the amount of data to be transmitted at a time and compares ED-CHSS with other solutions for various network conditions. Analysis of network lifetime This indicator discusses how long the network remained online before a node went down due to a power interruption, using Eq. (27). ED-CHSS has the longest network lifetime at 280 h, which is longer than LEACH (200 h) and other methods in Fig. 7. The network’s operational lifespan is longer since it can change its schedule and choose cluster heads that consume less energy. This means that nodes do not run out of energy too soon.27\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P = \frac{E}{t}$$\end{document}Fig. 7Network lifetime analysis. Network lifetime analysis. Power consumption P is determined by Eq. (27), where E is the energy consumed, and t is the time consumed. It aids in assessing the efficiency of energy use in the network operations. Detailed network lifetime analysis The findings indicate that ED-CHSS is far superior to current methods at enhancing network lifetime at all levels, as shown in Table 4. Higher FND indicates that node failures will be delayed, a sign of improved energy balancing. The enhanced HND indicates that half of the nodes remain active for a prolonged period, ensuring the network’s endurance. The LND metric validates the network’s lifetime, with ED-CHSS having the longest lifetime due to its efficient cluster-head selection, adaptive scheduling, and even energy distribution. All these metrics demonstrate that the suggested approach would increase the stability, reliability, and long-term performance of wireless sensor networks.Table 4Network lifetime metrics comparison.MethodFND (rounds)HND (rounds)LND (rounds)LEACH80015002000DEEC100019002600HEED95018002400Secure-CH110021002700ED-CHSS130024003000 Table 4Network lifetime metrics comparison.MethodFND (rounds)HND (rounds)LND (rounds)LEACH80015002000DEEC100019002600HEED95018002400Secure-CH110021002700ED-CHSS130024003000 Network lifetime metrics comparison. Analysis of energy consumption ED-CHSS consumes the least energy (85 J) among other methods, as shown in Eq. (28). It reduces overhead costs while optimizing energy use through dynamic scheduling and data transfer (Fig. 8). This is better than LEACH (120 J) and ISSA (110 J). The technology prevents power loss and maintains a stable, efficient network by selecting cluster heads based on available energy.28\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{{i\left( {new} \right)}} = E_{{i\left( {old} \right)}} - E_{consumed}$$\end{document}Fig. 8Energy consumption analysis. Energy consumption analysis. The residual energy \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{{i\left( {new} \right)}}$$\end{document} is the value of node i is updated with Eq. (28) after every transmission as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{{i\left( {old} \right)}}$$\end{document} or operation as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{consumed}$$\end{document}. It provides proper monitoring of node energy states to inform future decisions on cluster head selection and scheduling. Analysis of data throughput The ED-CHSS system achieves the highest data throughput of 600 kbps, even compared to TSCH (550 kbps) and other technologies shown in Table 5. The system’s adaptive scheduling and effective cluster head management ensure no congestion and a high data transfer rate. ED-CHSS can handle more traffic than traditional methods while still allowing high-speed communication in dense networks. Network throughput is measured using an equation and is considered the quantity of data received successfully over a period of time. Increased throughput means effective communication and improved network performance.Table 5Data throughput analysis.AlgorithmDescriptionData throughput (kbps)TrendLEACHBasic clustering protocol with limited throughput due to static configuration500–ISSAOptimized clustering improves throughput slightly520↑ + 20EDASSEfficient adaptive scheduling allows better data delivery rate540↑ + 20AdaptiveHARTReal-time adaptation slightly lowers average throughput530↓ − 10TSCHTime-slot scheduling and hopping ensure high throughput and stability550↑ + 20ED-CHSSHighest throughput with efficient cluster head selection and energy balance600↑ + 50 Table 5Data throughput analysis.AlgorithmDescriptionData throughput (kbps)TrendLEACHBasic clustering protocol with limited throughput due to static configuration500–ISSAOptimized clustering improves throughput slightly520↑ + 20EDASSEfficient adaptive scheduling allows better data delivery rate540↑ + 20AdaptiveHARTReal-time adaptation slightly lowers average throughput530↓ − 10TSCHTime-slot scheduling and hopping ensure high throughput and stability550↑ + 20ED-CHSSHighest throughput with efficient cluster head selection and energy balance600↑ + 50 Data throughput analysis. Analysis of security (trust management) ED-CHSS has strong trust management, ensuring that the data is protected and examined using Eq. (29). Trust-based security measures make sure that nodes are reliable and that data is safe. ED-CHSS ensures data is transferred safely by removing risky nodes and minimizing the risk of data tampering, as shown in Fig. 9. This makes it well-suited for secure applications, unlike LEACH (poor security), which lacks advanced security.29\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_{i} = \frac{{\left( {T_{i} \times E_{i} } \right)}}{{Risk_{i} }}$$\end{document}Fig. 9Security (trust management) analysis. Security (trust management) analysis. Equation 29 is a method of assessing node security \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_{i}$$\end{document} that utilizes trust \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i}$$\end{document}, energy \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{i}$$\end{document}, and risk factors \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Risk_{i}$$\end{document}. Communication should be in high trust/high energy/low risk nodes to support secure and reliable network operations. Attack-resilient communication analysis Table 6 provides a comparison of the Packet Delivery Ratio (PDR) with normal and attack cases. Selective forwarding and sinkhole attacks are well mentioned to be serious issues in conventional methods like LEACH and ISSA. On the other hand, the proposed ED-CHSS achieves a higher PDR value due to the filtering mechanism based on trust. Added to this, ED-CHSS has the highest detection rate, indicating its efficiency in detecting and isolating malicious nodes in the network.Table 6Attack-resilient communication analysis.MethodNormal PDR (%)PDR under selective forwarding (%)PDR under sinkhole attack (%)Detection rate (%)LEACH92686560ISSA94757270TSCH95807875AdaptiveHART93777473ED-CHSS97908892 Table 6Attack-resilient communication analysis.MethodNormal PDR (%)PDR under selective forwarding (%)PDR under sinkhole attack (%)Detection rate (%)LEACH92686560ISSA94757270TSCH95807875AdaptiveHART93777473ED-CHSS97908892 Attack-resilient communication analysis. Attack-resilient communication analysis The proposed ED-CHSS method was compared to the K-NN-PEGASIS method proposed by Juwaied and Romanowski30 with special focus on Black-hole attack, Selective Forwarding attack, and False Energy Reporting attack. As it can be seen from the comparison, the cluster head selection process and adaptive scheduling in ED-CHSS provide good resilience. The forwarding behavior, delivery success rate, and cooperation metrics are dynamically utilized for malicious nodes detection to detect nodes that are malicious to drop packets, advertise false routing information, and report false energy consumption. As a result, ED-CHSS results in the higher packet delivery ratio, improved network stability and more reliable communication in adverse network conditions. Analysis of load balancing efficiency The ED-CHSS system is the best, as it achieves 90% load balancing efficacy using Eq. (30). It ensures that no node is overwhelmed by adjusting its transmission schedule based on how much energy it has and how much traffic there is, as mentioned in Table 7. ED-CHSS ensures that resources are used fairly by minimizing congestion and improving performance across all network nodes. This is different from LEACH (60%), which has problems with load distribution.30\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$LBE = \frac{{ \min \left( {L_{i} } \right)}}{{\max \left( {L_{i} } \right)}}$$\end{document}Table 7Load balancing efficiency analysis.MethodologyDescriptionLoad balancing efficiency (%)TrendLEACHLow-energy adaptive clustering hierarchy: a basic clustering protocol60–ISSAImproved sparrow search algorithm for optimized resource distribution70↑ + 10%EDASSEnergy-driven adaptive scheduling Scheme ensuring balanced energy usage75↑ + 5%AdaptiveHARTHybrid adaptive real-time algorithm enhancing dynamic network performance80↑ + 5%TSCHTime-slotted channel hopping enables efficient, collision-free scheduling85↑ + 5%ED-CHSSEnergy-driven cluster head selection scheme with superior load distribution90↑ + 5% Table 7Load balancing efficiency analysis.MethodologyDescriptionLoad balancing efficiency (%)TrendLEACHLow-energy adaptive clustering hierarchy: a basic clustering protocol60–ISSAImproved sparrow search algorithm for optimized resource distribution70↑ + 10%EDASSEnergy-driven adaptive scheduling Scheme ensuring balanced energy usage75↑ + 5%AdaptiveHARTHybrid adaptive real-time algorithm enhancing dynamic network performance80↑ + 5%TSCHTime-slotted channel hopping enables efficient, collision-free scheduling85↑ + 5%ED-CHSSEnergy-driven cluster head selection scheme with superior load distribution90↑ + 5% Load balancing efficiency analysis. Equation (30) indicates the evenness of load distribution among nodes LBE. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{i}$$\end{document} is node load. Values near 1 indicate a balanced distribution that enhances network stability and averts overloading. Analysis of scalability ED-CHSS is good at handling large networks and can grow easily, as shown in Fig. 10. It changes by splitting networks into smaller groups to reduce communication overhead and congestion, using Eq. (31). ED-CHSS is far better than other technologies like ISSA and LEACH for building networks or Internet of Things systems with thousands of nodes, making it quick and easy.31\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$SF_{i} = \frac{{E_{i} }}{{L_{i} }}$$\end{document}Fig. 10Scalability analysis. Scalability analysis. Equation (31) gives priorities to nodes to be scheduled \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$SF_{i}$$\end{document} according to their energy and load. The high energy \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$E_{i}$$\end{document} and low load \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{i}$$\end{document} nodes are chosen initially, resulting in effective resource use and high performance. Scalability analysis under varying conditions The scalability of the proposed ED-CHSS algorithm is validated by varying the network size, node density, deployment area for the nodes. They run three experiments to investigate how the system reacts to different network conditions.Scalability versus network size (area = 200 × 200 m2) Scalability versus network size (area = 200 × 200 m2) Scalability versus network size (area = 200 × 200 m2) In the initial analysis, the number of nodes is varied between 50 and 500, with the deployment area maintained in Table 8. As the number of nodes grows, the scalability score decreases as communication overhead increases and resources become more contentious. Further, it’s not extreme drop, suggesting algorithm does not negatively impact heavily on performances even in large networks.(b)Scalability versus node density (nodes = 200)Table 8Scalability versus network size analysis.Number of nodesNode density (nodes/m2)Scalability score500.0012598.941000.0025097.882000.0050095.753000.0075093.634000.0100091.505000.0125089.38 (b)Scalability versus node density (nodes = 200) Scalability versus node density (nodes = 200) Table 8Scalability versus network size analysis.Number of nodesNode density (nodes/m2)Scalability score500.0012598.941000.0025097.882000.0050095.753000.0075093.634000.0100091.505000.0125089.38 Scalability versus network size analysis. The second analysis in Table 9 manipulates node density by varying the deployment area with a constant node population. More interference, collisions, and redundant communication due to increased density leads to a slight reduction in scalability. However, adaptive clustering mechanism is helpful in managing interactions between nodes and minimising the performance degradation.(c)Scalability versus deployment area (nodes = 200)Table 9Scalability versus node density.Deployment area (m × m)Node density (nodes/m2)Scalability score100 × 1000.0200095.00200 × 2000.0050095.75300 × 3000.0022295.89400 × 4000.0012595.94500 × 5000.0008095.96 (c)Scalability versus deployment area (nodes = 200) Scalability versus deployment area (nodes = 200) Table 9Scalability versus node density.Deployment area (m × m)Node density (nodes/m2)Scalability score100 × 1000.0200095.00200 × 2000.0050095.75300 × 3000.0022295.89400 × 4000.0012595.94500 × 5000.0008095.96 Scalability versus node density. The next analysis in Table 10 shows that the deployment area is expanded while the number of nodes remains the same. The higher the area, the fewer the nodes, the less the communication interference, the more scalable. This is an example to show its effectiveness in dense network and sparse network.Table 10Scalability versus deployment area.Deployment area (m × m)Node density (nodes/m2)Scalability score100 × 1000.0200095.00200 × 2000.0050095.75300 × 3000.0022295.89400 × 4000.0012595.94500 × 5000.0008095.96 Table 10Scalability versus deployment area.Deployment area (m × m)Node density (nodes/m2)Scalability score100 × 1000.0200095.00200 × 2000.0050095.75300 × 3000.0022295.89400 × 4000.0012595.94500 × 5000.0008095.96 Scalability versus deployment area. The results show that ED-CHSS is very scalable and its performance degrades slightly as the network size and density increase. The algorithm is distributed, adapts the cluster formation, and uses resources efficiently, so that it has similar performance in different large-scale wireless sensor network environments. Analysis of computational overhead In Fig. 11, ED-CHSS’s scheduling algorithm and trust management algorithm are very simple in nature, which the computation burden is light. If wireless network resources are limited, then ED-CHSS is more desirable than LEACH (as LEACH has more overhead), which requires strict iterative operations, as outlined in Eq. (32). ED-CHSS is an efficient and lightweight approach to minimize data and selection costs and energy consumption.32\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{comp} = O\left( {N\log N} \right)$$\end{document}Fig. 11Computational overhead analysis. Computational overhead analysis. The complexity of the proposed algorithm is presented in Eq. (32) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{comp}$$\end{document} is the number of nodes N. It shows that this system can be scalable and work for large-scale wireless sensor networks. The computational cost is measured by the number of computational operations per node such as in evaluating energy parameters, computing trust and estimating communication costs. Operations are distributed to each normalized computation unit (CU) and the total overhead is compared with a baseline approach (LEACH). Although the extra trust evaluation, ED-CHSS’s distributed design requires performing only minimal numbers of re-computations, and thus incurs only a small overhead. To some extent, the computational cost of ED-CHSS is less than that of other secure protocols, and its performance is more than that of other secure protocols, so it is suitable for resource-constrained wireless sensor networks. Analysis of adaptability to network changes Many different network changes are possible with ED-CHSS, including node outages, mobility, and traffic changes, as analyzed in Table 11. It utilizes real-time data for scheduling, and dynamically changes the cluster head, more adaptive than LEACH, according to Eq. (33). ED-CHSS is the best choice for handling changing scenarios and maintaining network performance because it is highly flexible.33\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$A = \frac{{Successful_{Adaptations} }}{{Total_{Changes} }}$$\end{document}Table 11Adaptability to network changes analysis.MethodologyAdaptability (score)TrendLEACH2.0–ISSA3.0↑ + 1.0EDASS4.0↑ + 1.0AdaptiveHART4.5↑ + 0.5TSCH3.5↓ − 1.0ED-CHSS5.0↑ + 1.5 Table 11Adaptability to network changes analysis.MethodologyAdaptability (score)TrendLEACH2.0–ISSA3.0↑ + 1.0EDASS4.0↑ + 1.0AdaptiveHART4.5↑ + 0.5TSCH3.5↓ − 1.0ED-CHSS5.0↑ + 1.5 Adaptability to network changes analysis. Equation (33) quantifies the ability of the system to adjust to network changes, including node failures or traffic changes as A. The higher value means that it is adaptable and robust in dynamic conditions. Packet delivery ratio (PDR) analysis The Packet Delivery Ratio is used to assess the dependability of the data delivery by determining the percentage of packets that reach the base station shown in Fig. 12. Congestion and packet loss typically decrease PDR as the number of nodes increases in traditional approaches like LEACH and ISSA. EMO-PEGASIS is a multi-objective optimization-based routing protocol which is used to enhance the delivery of packets and the reduction of delay caused by a chain-based communication. It however, does not include trust-based security or adaptive scheduling, unlike ED-CHSS, which restricts its ability to withstand adversarial conditions. EMO-PEGASIS31 enhances performance by making optimal routing choices. In addition, the offered ED-CHSS is the most successful in terms of PDR because it employs a trust-based communication approach and a dynamic scheduling algorithm that reduce packet loss and enhance data delivery reliability. This indicates the strength of ED-CHSS in handling dense, dynamic network conditions.Fig. 12Packet delivery ratio analysis. Packet delivery ratio analysis. End-to-end delay analysis End-to-End Delay is the time it takes data packets to travel from source nodes to the base station. With a larger network, delays in traditional protocols are increased by ineffective routing and congestion, as shown in Fig. 13. The EMO-PEGASIS minimizes delay through route optimization. Nonetheless, ED-CHSS has the least delay due to its favorable scheduling and energy-conscious cluster-head selection algorithm. The system reduces retransmissions and delays during transmission by giving high-energy, trusted nodes priority. This leads to quicker, more effective communication, and ED-CHSS is appropriate for real-time, latency-conscious wireless sensor networks.Fig. 13End-to-end delay analysis. End-to-end delay analysis. Comparison with ACLB-CH approach Table 12 provides comparisons between the ED-CHSS and ACLB-CH approach and other techniques based on numerical measures. The efficient clustering and dynamic scheduling enable ED-CHSS to consume the least energy and achieve the maximum network lifetime. Although ACLB-CH is effective at energy optimization, it lacks effective attack detection. In a wireless sensor network setting, ED-CHSS enhances the packet delivery rate and detection rate by a significant margin through its trust-based mechanism, which guarantees the dependability and safety of the communication process. The findings indicate that ED-CHSS has lower latency rates because it is efficient in selecting the cluster head and in adaptive scheduling, which minimize transmission delays. Localized decision-making and minimizing control message exchange also reduce control overhead.Table 12Comparison with ACLB-CH.MethodEnergy (J)Lifetime (rounds)PDR (%)Detection rate (%)Latency (ms)Control overhead (%)ACLB-CH95260886511016LEACH120200856012018ISSA110230897010515DEEC95–105240–26090–9165–70100–11014–16ED-CHSS8528092928512 Table 12Comparison with ACLB-CH.MethodEnergy (J)Lifetime (rounds)PDR (%)Detection rate (%)Latency (ms)Control overhead (%)ACLB-CH95260886511016LEACH120200856012018ISSA110230897010515DEEC95–105240–26090–9165–70100–11014–16ED-CHSS8528092928512 Comparison with ACLB-CH. Comparative analysis with existing protocols To prove the efficiency of the proposed ED-CHSS algorithm, a detailed comparative study is performed with the existing clustering protocols like SEP, DEEC, HEED and secure clustering schemes as listed in Table 13. The standards make it easier to compare the efficiency of energy use, scaling and security of wireless sensor networks.Table 13Comparative analysis with existing protocols.ProtocolEnergy efficiency (J)Network lifetime (rounds)PDR (%)Security levelScalabilitySEP110210088LowModerateDEEC95260091LowHighHEED100240089MediumModerateSecure-CH90270093HighModerateED-CHSS85300096Very highHigh Table 13Comparative analysis with existing protocols.ProtocolEnergy efficiency (J)Network lifetime (rounds)PDR (%)Security levelScalabilitySEP110210088LowModerateDEEC95260091LowHighHEED100240089MediumModerateSecure-CH90270093HighModerateED-CHSS85300096Very highHigh Comparative analysis with existing protocols. Statistical validation of results To provide statistical strength, the all-essential performance indicators: energy consumption, network lifetime, packet delivery ratio (PDR), detection rate and end-to-end delay have 95% confidence intervals (CIs). Each experiment was conducted multiple times with random seeds and mean and standard deviation was calculated. Table 14 shows the statistical validation of the ED-CHSS with known protocols such as mean, standard deviation and confidence interval. The proposed technique is less energy consuming and a long lifetime is achieved by the proposed network as the standard deviation values are low for the proposed method with little variation. The small confidence interval for the packet delivery ratio means ED-CHSS is fairly consistent from simulation to simulation. Other methods, on the contrary, are more variable and have broader confidence intervals. These results confirm the feasibility of ED-CHSS for large-scale wireless sensor networks, and its high consistency, reliability and repeatability in various network environments.Table 14Statistical validation of results.MethodEnergy (J) mean ± StdLifetime (rounds) mean ± StdPDR (%) mean ± Std95% CI (PDR)LEACH120 ± 52000 ± 8085 ± 2[83, 87]DEEC95 ± 42600 ± 7091 ± 1.8[89.5, 92.5]HEED100 ± 4.52400 ± 7589 ± 2[87, 91]Secure-CH90 ± 32700 ± 6093 ± 1.5[91.5, 94.5]ED-CHSS85 ± 2.53000 ± 5096 ± 1[95, 97] Table 14Statistical validation of results.MethodEnergy (J) mean ± StdLifetime (rounds) mean ± StdPDR (%) mean ± Std95% CI (PDR)LEACH120 ± 52000 ± 8085 ± 2[83, 87]DEEC95 ± 42600 ± 7091 ± 1.8[89.5, 92.5]HEED100 ± 4.52400 ± 7589 ± 2[87, 91]Secure-CH90 ± 32700 ± 6093 ± 1.5[91.5, 94.5]ED-CHSS85 ± 2.53000 ± 5096 ± 1[95, 97] Statistical validation of results. Sensitivity analysis of key parameters The sensitivity analysis shows the effect of variation in the parameters on the performance of ED-CHSS as presented in Table 15. The difference of the weight coefficients shows that the best performance is obtained when energy, distance and trust are all optimized equally. The trust threshold significantly affects the security and reliability, and moderate threshold (around 0.7) has the best security/efficiency to detection/communication efficiency trade-off. With increasing initial energy, the network lifetime and PDR are expected to increase. Furthermore, the more special values are assigned to the parameter, the worse the performance will be. The results of this study are in line with the notion that ED-CHSS is a stable and flexible tool that can yield reproducible results across a broad range of parameters.Table 15Sensitivity analysis results.Parameter variationNetwork lifetime (rounds)PDR (%)w1 = 0.3, w2 = 0.3, w3 = 0.4285094w1 = 0.4, w2 = 0.3, w3 = 0.3300096w1 = 0.5, w2 = 0.2, w3 = 0.3290095Trust threshold = 0.5275092Trust threshold = 0.7300096Trust threshold = 0.9295095Initial energy = 0.5 J220090Initial energy = 1.0 J300096Initial energy = 1.5 J340097 Table 15Sensitivity analysis results.Parameter variationNetwork lifetime (rounds)PDR (%)w1 = 0.3, w2 = 0.3, w3 = 0.4285094w1 = 0.4, w2 = 0.3, w3 = 0.3300096w1 = 0.5, w2 = 0.2, w3 = 0.3290095Trust threshold = 0.5275092Trust threshold = 0.7300096Trust threshold = 0.9295095Initial energy = 0.5 J220090Initial energy = 1.0 J300096Initial energy = 1.5 J340097 Sensitivity analysis results. Extended security evaluation under multiple attack scenarios As Table 16 illustrates, ED-CHSS can perform at a high level even under different attack conditions. In Sybil attacks, the trust mechanism is able to detect fake identities and ensures reliable communication. Although packet delivery is compromised slightly in wormhole attacks, the impact on packet delivery is minimized with distributed trust evaluation. Abnormal packet dropping behavior can be used to accurately detect black hole attacks. The degradation of PDR is almost negligible under normal conditions which shows that it is highly resistant. In all, ED-CHSS is able to prevent various sorts of assaults, so that wireless sensor system would more secure and reliable in communication.Table 16Performance under different attack scenarios.Attack typePDR (%)Detection rate (%)No attack96–Sybil9190Wormhole8988Blackhole8792Selective forwarding9091 Table 16Performance under different attack scenarios.Attack typePDR (%)Detection rate (%)No attack96–Sybil9190Wormhole8988Blackhole8792Selective forwarding9091 Performance under different attack scenarios. Table 17 and Fig. 14 show the comparative analysis of proposed ED-CHSS method with K-NN-PEGASIS method for various attack scenarios in WSNs. K-NN-PEGASIS enhances secure leader selection via machine learning, while ED-CHSS achieves higher resilience against Black-hole attacks, Selective Forwarding attacks and False Energy Reporting attacks by leveraging adaptive trust management, dynamic scheduling, and continuously monitoring node-behaviour. Trust-aware energy verification, in combination with malicious-node isolation capabilities, helps to preserve the reliability of packet delivery, the security of communications, and network stability in adversarial environments.Table 17Attack-resilient communication analysis.Feature/attack scenarioK-NN-PEGASISED-CHSSBlack-hole attack detectionModerateHighSelective forwarding resistanceHighHighFalse energy reporting protectionLimitedHighTrust-based cluster head selectionYesYesAdaptive scheduling supportNoYesDynamic trust updatingPartialFullPacket delivery reliability under attackModerateHighEnergy-aware secure routingModerateHighMalicious node isolationPartialEffectiveScalability in large-scale WSNModerateHighFig. 14Security performance comparison between K-NN-PEGASIS and ED-CHSS under different WSN attack scenarios. Table 17Attack-resilient communication analysis.Feature/attack scenarioK-NN-PEGASISED-CHSSBlack-hole attack detectionModerateHighSelective forwarding resistanceHighHighFalse energy reporting protectionLimitedHighTrust-based cluster head selectionYesYesAdaptive scheduling supportNoYesDynamic trust updatingPartialFullPacket delivery reliability under attackModerateHighEnergy-aware secure routingModerateHighMalicious node isolationPartialEffectiveScalability in large-scale WSNModerateHigh Attack-resilient communication analysis. Security performance comparison between K-NN-PEGASIS and ED-CHSS under different WSN attack scenarios. The comparison (Table 18) is unified and removes inconsistencies in previous sections by comparing all the baseline protocols under the same conditions of simulation and metrics. This unified analysis allows making the direct cross-protocol comparisons and proves that ED-CHSS is always superior to the current methods in various aspects of performance.Table 18Comparative analysis.ProtocolEnergy (J)Lifetime (rounds)PDR (%)Detection rate (%)Delay (ms)LEACH1202008560120HEED1102208765110DEEC1002509070105SEP1052408868108Secure-CH1152308980112EMO-PEGASIS952609175100ACLB-CH952608865110ISSA1102308970105TSCH105240907295ED-CHSS85280929285 Table 18Comparative analysis.ProtocolEnergy (J)Lifetime (rounds)PDR (%)Detection rate (%)Delay (ms)LEACH1202008560120HEED1102208765110DEEC1002509070105SEP1052408868108Secure-CH1152308980112EMO-PEGASIS952609175100ACLB-CH952608865110ISSA1102308970105TSCH105240907295ED-CHSS85280929285 Comparative analysis. To confirm that ED-CHSS outperforms the baseline protocols, we used paired t-tests to compare the baseline protocols with ED-CHSS on the main performance measures of energy consumption, network lifetime, packet delivery ratio (PDR), detection rate, and end-to-end delay. The experiments were repeated in a series of runs with varying random seeds and comparisons were made using mean values. Table 18 documents the precise p-values of the pairwise comparisons between ED-CHSS and each of the baseline protocols. In both instances, the p-values are less than the significance level of 0.05, which proves that the improvements observed are statistically significant and not because of random variation. The computational complexity of ED-CHSS algorithm is analyzed by dividing it into operational phases. The selection of Cluster Head and trust evaluation in the Primary Cluster has linear time complexity of O(n) because all nodes are scanned and local metrics like residual energy, communication cost, trust value etc. are calculated. This is done by adding another sorting or ranking operation which has an O(nlogn) complexity to select optimal cluster head candidates based on their computed scores. Thus, the total running time of ED-CHSS is O(nlogn), and the main operational cost per round is roughly linear in practice. This difference is key to the observed scalability in large-scale wireless sensor networks. The findings show that every element plays an important role towards the overall performance. Losing the trust mechanism results in an area of significant reduction in detection rate and percentage of packet delivery because unreliable nodes are added. Adaptive weighting is also more energy efficient because the model is no longer able to dynamically focus on the important parameters in different network conditions. Similarly, the lack of awareness of the scheduling contributes to the latency of communications and the overhead of control due to lack of coordination in transmitting data. The testing results indicate that the ED-CHSS system is superior to the existing systems, as it is capable of adapting to network changes, reducing energy consumption, and providing optimal protection of data. The system does not reveal that the proposed mechanism for selecting and organizing cluster heads is effective, and it guarantees that throughput and stability of clusters do not decrease with the increase of the number of cluster nodes. Conclusion and future work Overall, the ED-CHSS system achieves superior performance in all metrics compared to other systems like AdaptiveHART, TSCH, LEACH, ISSA and EDASS. The advantages of this technology are of great relevance for large Wireless Sensor Networks, as it demands very low power consumption, generates a high data volume, has a long lifetime, can be extended, is safe and flexible. By carefully controlling trust and selecting cluster heads, ED-CHSS uses less energy, extends network lifetime, and enhances data security. In the real world, there are lower operating costs, more robust networks, and better resource use. Future work In the future, ED-CHSS will integrate with Internet of Things (IoT) systems to better manage devices, employ edge computing to reduce latency, and make it easier to adapt to network changes, including when people move about. To make the system even better for a wider range of large-scale applications, researchers can explore hybrid energy management for long-term use in remote areas and improve security through stronger encryption.