AoI in Context-Aware Hybrid Radio-Optical IoT Networks

Aymen Hamrouni, Sofie Pollin, and Hazem Sallouha · 2024

Read AoI in Context-Aware Hybrid Radio-Optical IoT Networks by Aymen Hamrouni, Sofie Pollin, and Hazem Sallouha (2024) 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-2412-12914.txt.

Introduction

Aymen Hamrouni, Sofie Pollin, and Hazem Sallouha wrote a research paper in two thousand twenty-four about keeping data fresh in networks of connected devices. Their subject is hybrid networks that pair radio links with optical links, meaning links that use light. They track freshness with a measure called Age of Information, which shows how old the newest data is.

The authors set up a math problem that weighs three goals at once. It seeks high data flow, low energy use, and few switches between the two link types. A scaling method keeps one goal from drowning out the others. In tests, adding the optical link cut both the mean and peak age of data.

Catalog record

CREATORAymen Hamrouni, Sofie Pollin, and Hazem Sallouha
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE32 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-12914

Browse the full library · About the Tower of Babel

Full text

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

AoI in Context-Aware Hybrid Radio-Optical IoT Networks Aymen Hamrouni, Sofie Pollin, and Hazem Sallouha Abstract With the surge in IoT devices ranging from wearables to smart homes, prompt transmission is crucial. The Age of Information (AoI) emerges as a critical metric in this context, representing the freshness of the information transmitted across the network. This paper studies hybrid IoT networks that employ Optical Communication (OC) as a reinforcement medium to Radio Frequency (RF). We formulate a non-linear convex optimization that adopts a multi-objective optimization strategy to dynamically schedule the communication between devices and select their corresponding communication technology, aiming to balance the maximization of network throughput with the minimization of energy usage and the frequency of switching between technologies. To mitigate the impact of dominant sub-objectives and their scale disparity, the designed approach employs a regularization method that approximates adequate sub-objective scaling weights. Simulation results show that the OC supplementary integration alongside RF enhances the network’s overall performances and significantly reduces the Mean AoI and Peak AoI, allowing the collection of the freshest possible data using the best available communication technology. I Introduction In recent years, the proliferation of Internet of Things (IoT) devices has revolutionized the way we live and work. From wearable gadgets that monitor our health to smart home systems that enhance our living environment, the ubiquity of IoT technologies is undeniable [ 1 ] . A critical aspect of IoT applications is the collection and transmission of data on time. In a healthcare setting, for example, IoT enables remote monitoring systems to track patients’ vitals in real-time. These systems continuously gather data on heart rate, blood pressure, and glucose levels, which must be sent promptly to enable accurate medical interventions and personalized care plans. One of the paramount Key Performance Indicators (KPIs) in IoT ecosystems is the Age of Information (AoI), which measures the freshness of the data being transmitted to and processed by a receiver. AoI is defined as the difference between the current time and the time at which the last successfully received update was generated at the source [ 2 , 3 ] . This metric is important, especially when the data is time-sensitive and the network is energy-limited. Finding a balance between keeping the information as fresh as possible and minimalist energy consumption is crucial for a smooth IoT network operation. Managing this information freshness versus energy trade-off in a single-technology IoT network presents a significant challenge. There have been some adaptive strategies that adjust the frequency of updates based on the criticality of information and current network conditions [ 4 ] . However, single-technology IoT networks, which predominantly rely on Radio Frequency (RF) communication are unscalable due to the limited RF resources. This scalability problem leads to an unpredictable effect on both AoI and energy efficiency. The integration of a secondary technology into the IoT network, such as Optical Communication (OC), can be a promising solution to such challenges [ 5 ] . OC technology can be designed as a reinforcement communication medium to the existing RF mediums to kick in and accelerate data transmission when critical updates are necessary. This enables IoT networks to achieve more flexible and efficient performance by leveraging: 1) the high-speed data transmission capabilities of OC and 2) the energy efficiency and physical barrier penetration of RF [ 6 ] . Indeed, most IoT devices are currently equipped with RF communication, but an OC extension is inexpensive and easy to set up [ 7 ] . The integration of OC serves as a supplementary rather than a primary communication technology, promoting a harmonious balance that optimizes energy consumption, ensures data freshness, and accommodates the dynamic conditions of IoT environments [ 5 , 8 ] . This being said, minimizing the AoI metric in an optimization problem with such hybrid settings represents challenge in itself. This is mainly due to the AoI’s inherently non-linear [ 9 ] and non-convex nature concerning decision variables such as scheduling, packet generation rates, and resource allocation [ 10 ] . Such characteristics render the AoI tractable only in highly simplified settings. Significant research has been conducted on hybrid RF-OC systems. Fakirah et al. [ 11 ] explored a visible light with an RF system that uses a set of hybrid access points to exchange information between the vehicles traversing the roundabout and roadside units in the vehicle-to-infrastructure mode. Obeed et al. [ 12 ] designed an iterative optimization-based algorithm for load balancing and power allocation schemes for a hybrid light and RF system consisting of one RF access point and multiple OC access points. Xiao et al. [ 13 ] investigated a hybrid downlink system that simultaneously uses visible light communication and RF with a cognitive-based resource allocation policy.On the other hand, studies focusing on the AoI include Lui et al. [ 14 ] , where the average AoI in a wireless-powered MEC system is assessed. Chen et al. [ 2 ] studied the AoI in a multi-channel network-sided information. Xu et al. [ 15 ] analyzed the Peak AoI across two scenarios: one considering a buffer size of one and infinite, and the other examining the infinite buffer size case. To the best of our knowledge, no existing work has investigated the impact of hybrid RF-OC devices on the AoI in IoT networks. In this paper, we formulate a non-linear convex optimization algorithm within a hybrid RF-OC network that seeks to 1) maximize the overall throughput across the network, 2) minimize the energy consumption, and 3) minimize the frequency of technology switch between RF and OC within the network. A unique feature of this approach is the dynamic selection mechanism that enables nodes to choose the most appropriate technology—RF or OC—at any given moment. To address the multifaceted nature of this multi-objective system, we conduct a regularization analysis to select the weights of the individual sub-objectives and avoid dominance or saturation. While the AoI was not minimized directly in the optimization problem, simulation results show that the RF-OC hybrid system achieves lower Mean AoI (M-AoI) and Peak AoI (P-AoI) compared to a single-technology RF system. Moreover, our problem formulation with adequate regularization inherently leads to improved information freshness in the RF-OC setup while leveraging conventional convex optimization. II System Model In this section, we introduce the system model and present the overall network and communication setups. II-A Network Setup We consider an IoT network comprising a set of IoT nodes and a set of Access Points (APs) having OC and RF capabilities. Nodes communicate with APs using either OC or RF, whereas only RF is used for inter-node communication. The system model with an RF-OC hybrid configuration is illustrated in Fig. 1 with different IoT nodes transmitting multiple types of application’s data. Let N_{d} be the number of IoT nodes and N_{APs} be the number of APs in the system. The total number of devices in the network is denoted by N=N_{d}+N_{APs} . Let L be the number of different types of data in the system. For example, in the case of healthcare settings, IoT devices can exchange two types of messages, say heart rates and oxygen levels. In this case, L is set to 2 . Let M be

Read the complete text