A Novel Zero-Touch, Zero-Trust, AI/ML Enablement Framework for IoT Network Security

Sushil Shakya, Robert Abbas, Sasa Maric · 2025

Read A Novel Zero-Touch, Zero-Trust, AI/ML Enablement Framework for IoT Network Security by Sushil Shakya, Robert Abbas, Sasa Maric (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-03614.txt.

Introduction

Sushil Shakya, Robert Abbas, and Sasa Maric wrote a 2025 paper on a security framework for Internet of Things networks. Billions of connected devices will ride on 5G and 6G networks, and each device is a potential entry point for attack, especially distributed denial-of-service floods.

Their framework fuses three ideas. Zero trust means no device or user is trusted by default. Zero touch means devices are configured and secured automatically, without manual setup. Artificial intelligence and machine learning supply the detection and response. Together they aim at autonomous detection, mitigation, and prevention of DDoS attacks across the IoT ecosystem. The paper presents the architecture as a step toward self-defending device networks.

Catalog record

CREATORSushil Shakya, Robert Abbas, Sasa Maric
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE25 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-03614

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A Novel Zero-Touch, Zero-Trust, AI/ML Enablement Framework for IoT Network Security Sushil Shakya, Robert Abbas, Sasa Maric Abstract The IoT facilitates a connected, intelligent, and sustainable society; therefore, it is imperative to protect the IoT ecosystem. The IoT-based 5G and 6G will leverage the use of machine learning and artificial intelligence (ML/AI) more to pave the way for autonomous and collaborative secure IoT networks. Zero-touch, zero-trust IoT security with AI and machine learning (ML) enablement frameworks offers a powerful approach to securing the expanding landscape of Internet of Things (IoT) devices. This paper presents a novel framework based on the integration of Zero Trust, Zero Touch, and AI/ML powered for the detection, mitigation, and prevention of DDoS attacks in modern IoT ecosystems. The focus will be on the new integrated framework by establishing zero trust for all IoT traffic, fixed and mobile 5G/6G IoT network traffic, and data security (quarantine-zero touch and dynamic policy enforcement). We perform a comparative analysis of five machine learning models, namely, XGBoost, Random Forest, K-Nearest Neighbors, Stochastic Gradient Descent, and Naïve Bayes, by comparing these models based on accuracy, precision, recall, F1-score, and ROC-AUC. Results show that the best performance in detecting and mitigating different DDoS vectors comes from the ensemble-based approaches. By incorporating network slicing, micro-segmentation, continuous authentication, and resilient 5G/6G strategies, the framework offers robust, scalable security against increasingly sophisticated ransomware-based DDoS attacks. Zero-touch, zero-trust IoT security with AI/ML enablement is the paradigm of a robust cybersecurity strategy in the age of the 5G/6G-based Internet of Things and Industry 4.0 and 5.0. By integrating these technologies, organizations can effectively secure their IoT environments, protect sensitive data, and maintain business continuity in the face of evolving cyber threats I Introduction The IoT is revolutionizing industries by connecting tens of billions of devices across healthcare, transport, smart cities, smart industries, smart mining, smart agriculture, and more. The Internet of Things (IoT) is bringing an influx of difficult-to-secure gadgets to enterprise networks. With the rapid increase in the number of IoT devices, efficiency and automation have increased, but so have the risks in IoT systems. One of the most tangible looming threats-attacks is DDoS, which causes a disruption of critical services and resource exhaustion by using botnets made of compromised devices. The combination of DDoS with ransomware tactics increased the risks by encrypting sensitive data and asking for ransom payments. The various features brought in by 5G/6G networks further complicate the security of IoT with network slicing, low latency, and edge computing. While these diverse advances will enable diverse and high-performance applications, they will also enlarge the attack surface, for which traditional perimeter-based defenses are no longer effective. Such complex and highly distributed networks demand scalable, adaptive, and intelligent solutions to secure them. We thereby provide an integrated framework: Zero-Touch provisioning , Zero-Trust security , and AI/ML-based threat detection . While Zero-Touch will automate the onboarding process for a secure device, Zero-Trust will proceed with the process of authentication in order to get rid of implicit trust. AI/ML algorithms allow for real-time anomaly detection and adaptation against evolving threats. Scalable and robust, the proposed solution can mitigate current IoT security risks in 5G/6G environments. II Literature Review II-A IoT Security In the IoT context, the efficiency of traditional security approaches like perimeter defenses and signature-based IDS are relatively falling into inefficiency due to the dynamic nature of IoT ecosystems, which are heterogeneous by nature. Most of them are not even able to mitigate modern threats that include DDoS, which depends on the high number of IoT devices with vulnerabilities to perform service disruption along with resource exhaustion [ 4 , 5 ] . The rapid proliferation of IoT devices has expanded the attack surface, necessitating more robust and adaptive security measures [ 1 , 2 ] . II-B ML-Enabled Security The potential of ML in strengthening IoT security includes conducting real-time anomaly detection, predictive analytics, and adapting responses against ever-evolving threats. Among these models, ensemble models, especially XGBoost and Random Forest, show the greatest efficiency in identifying deviations in network traffic patterns that could hint at malicious activity [ 16 , 18 ] . While effective, ML-based solutions are still plagued by problems of computational overhead, algorithmic bias, and explainability gaps, making them ineffective and very difficult to deploy at scale in IoT environments [ 15 , 31 ] . II-C Zero-Trust Security and Network Sclicing Network slicing is a technique that creates multiple virtual networks on a single physical network. Network slicing enables virtualized networks to operate on the same physical network infrastructure. The basic idea of network slicing is to “slice” the original architecture into multiple logical and independent networks. These sliced networks can then be configured to effectively meet various application needs and service requirements. Network slicing and Micro-segmentation can not only make up for the shortcomings of firewalls but when implemented network-wide, can remove the need for a firewall altogether. Zero-Trust security ensures continuous authentication and authorization of devices and communications to remove implicit trust within the network boundary. It follows the principle ”never trust, always verify,” which drastically reduces the chances of lateral movement on the part of the attacker [ 9 , 10 ] . Coupled with real-time network traffic analysis and dynamic policy enforcement, this makes it highly suitable for the protection of IoT ecosystems-especially in the context of 5G/6G [ 19 , 22 ] . II-D Zero-Touch Provisioning Zero-Touch provisioning automates the supply chain for onboarding IoT devices in a secured manner with minimum human intervention. This reduces the chances of configuration errors, ensuring that each device is authenticated, securely booted, and checked against the specified policies before joining the network [ 2 , 22 ] . This is particularly important for the Zero-Touch scaling of security as IoT networks continue to increase in number and complexity. II-E Methodology and Dataset Details For this work, we used a labeled dataset of IoT network traffic that had been anonymized so as to be able to apply machine learning analysis. The dataset contains typical features of network traffic, such as packet size and rate, session duration, protocol type, and labels indicating whether the traffic was part of a DDoS attack or normal activity. To prepare the data for analysis, we conducted several preprocessing steps: • Data Cleaning : Removed incomplete or corrupt records that could skew results. • Feature Selection : Selected relevant features based on their correlation with DDoS attack detection. • Normalization : Standardized feature values to ensure consistency in scale, using Min-Max scaling. For each feature x , the normalized value x^{\prime} is calculated as: x^{\prime}=\frac{x-\min(x)}{\max(x)-\min(x)} (1) II-F Selected ML Models We evaluated five machine learning models based on their suitability for classification tasks within cybersecurity contexts[21]. Fig. 1: Comparison of Machine Learning Models II-F 1 XGBoost The gradient-boosted decision tree model XGBoost [ 16 ] is meant to be quick and efficient. It works well with complicated datasets because it introduces regularisation to reduce overfitting. The model optimizes an objective function \mathcal{L} , which includes a loss term and a regularization term: \mathcal{L}=\sum_{i=1}^{n}l(y_{i},\hat{y}_{i})+\sum_{k=1}^{K}\Omega(f_{k}) (2) where l(y_{i},\hat{y}_{i}) is the loss function and \Omega(f_{k}) represents the regularization applied to each tree. II-F 2 RandomForest Random Forest is an

Read the complete text