arXiv subjects

Hazem Sallouha

Publications and source records attributed to Hazem Sallouha.

At least 19 recordsLinked to original sources

arXiv2026

Joint Mobile User Positioning and Passive Target Sensing using Optimized Sequential Beamforming

Integrated sensing and communication (ISAC) relies on monostatic sensing (MS) and bistatic positioning (BP) to enable comprehensive environmental awareness and user localization. However, existing frameworks predominantly assume static geometries and optimize these modalities independently, neglecting user mobility and sequential information sharing. In this paper, we propose a velocity-aware sequential beamforming framework that dynamically couples MS and BP in time. We derive the Cramer-Rao bounds (CRBs) in the position domain to formulate a non-convex resource allocation problem. Instead of relying on static weighted-sum tradeoffs, we introduce a sequential Bayesian optimization strategy where MS is executed first to construct a reliable structural prior on the UE and passive targets (PTs). This covariance prior is subsequently passed to the UE to regularize the BP estimation stage. We demonstrate that optimizing a single shared beamformer globally across both phases yields superior synergistic gains compared to a two-stage greedy approach. Simulation results validate that the shared sequential design efficiently balances limited symbol resources, achieving centimeter-level positioning accuracy for both the UE and PTs, robust velocity estimation, and a significantly reduced computational runtime.

eess.SP↗

arXiv2026

Disaggregated multi-domain interference classification for O-RAN

Spectrum sharing and dynamic spectrum reuse are becoming increasingly critical in modern wireless networks to address spectrum scarcity. However, these techniques inevitably increase Cross-Technology Interference (CTI). In this context, the Open Radio Access Network (O-RAN), as a modern and disaggregated network architecture, necessitates accurate, low-latency, and computationally efficient CTI classification and mitigation to support real-time control and maintain Quality of Service (QoS). Unfortunately, existing solutions predominantly rely on high-complexity, monolithic deep learning-based solutions that, while achieving high classification accuracy, incur significant latency and computational overhead This paper exploits the O-RAN functional split to leverage multi-domain raw signal representations (time, frequency, and Channel State Information (CSI)) directly from the same data stream. Each domain is processed locally, naturally interleaving CTI within the distributed, disaggregated O-RAN architecture. This distributed strategy enables a cost-aware, multi-domain fusion architecture that balances classification accuracy with computational overhead and latency. Our proposed multi-domain distributed architecture achieves a 400 $μs$ inference latency on standard CPUs. Compared to a state-of-the-art monolithic frequency-domain classifier, this represents an average 9x reduction in latency and an 11-fold decrease in computational cost, while sacrificing only 4% in classification performance and maintaining >90% accuracy in high-interference conditions.

eess.SP↗

arXiv2026

Resource Allocation in Hybrid Radio-Optical IoT Networks using GNN with Multi-task Learning

This paper addresses the problem of dual-technology scheduling in hybrid Internet-of-Things (IoT) networks that integrate Optical Wireless Communication (OWC) with Radio Frequency (RF). We first present an optimization formulation that jointly maximizes throughput and minimizes delivery-based Age of Information (AoI) between access points and IoT nodes under energy and link availability constraints. However, solving such NP-hard problems at scale is computationally intractable and typically assumes full channel observability, which is impractical in real deployments. To address this challenge, we propose the Dual-Graph Embedding with Transformer (DGET) framework, a supervised multi-task learning architecture that combines a two-stage Graph Neural Network (GNN) with a Transformer encoder. The first stage employs a transductive GNN to encode the known graph topology together with initial node and link states, such as energy levels, available links, and queued transmissions. The second stage introduces an inductive GNN for temporal refinement, enabling the model to generalize these embeddings to evolving network states while capturing variations in energy and queue dynamics over time through a consistency loss. The resulting embeddings are then processed by a Transformer-based classifier that models cross-link dependencies using multi-head self-attention. Simulation results show that hybrid RF-OWC networks outperform standalone RF systems by supporting higher traffic loads and reducing AoI by up to 20% while maintaining comparable energy consumption. Compared with optimization-based methods, the proposed DGET framework achieves near-optimal scheduling with over 90% classification accuracy, lower computational complexity, and improved robustness under partial channel observability.

cs.NI↗

arXiv2026

Data Augmentation and Attention for massive MIMO-based Indoor Localization in Changing Environments

The demand for high-precision indoor localization has grown significantly with the rise of smart environments, industrial automation, and location-aware applications. While massive Multiple-Input and Multiple-Output (MIMO) systems enable millimeter-level accuracy by leveraging rich Channel State Information (CSI), most existing solutions are optimized for static environments, where users or devices remain fixed during data collection and inference. Real-world applications, however, often require real-time localization in changing environments, where rapid movement, unpredictable blockages, and dynamic channel conditions pose significant challenges. To address these challenges, we introduce two data augmentation techniques designed to resemble blocked antennas, enhancing the generalizability of localization models to dynamic scenarios. Additionally, we enhance an existing Deep Learning (DL) model by incorporating attention modules, improving its ability to focus on relevant channel features and antennas. We train our model on data from a static scenario, augmented with the proposed techniques, and evaluate it on a dataset collected in changing scenarios. We investigate the performance enhancements achieved by the data augmentation techniques and the Attention modules, and observe a localization accuracy improvement from a mean error of 286 mm, when trained without Attention and without data augmentations, to 66 mm, when trained with Attention and data augmentation. This shows that high localization accuracy can be maintained in changing environments, even without training data from those scenarios.

eess.SY↗

arXiv2026

Machine Learning on the Edge for Sustainable IoT Networks: A Systematic Literature Review

The Internet of Things (IoT) has become integral to modern technology, enhancing daily life and industrial processes through seamless connectivity. However, the rapid expansion of IoT systems presents significant sustainability challenges, such as high energy consumption and inefficient resource management. Addressing these issues is critical for the long-term viability of IoT networks. Machine learning (ML), with its proven success across various domains, offers promising solutions for optimizing IoT operations. ML algorithms can learn directly from raw data, uncovering hidden patterns and optimizing processes in dynamic environments. Executing ML at the edge of IoT networks can further enhance sustainability by reducing bandwidth usage, enabling real-time decision-making, and improving data privacy. Additionally, testing ML models on actual hardware is essential to ensure satisfactory performance under real-world conditions, as it captures the complexities and constraints of real-world IoT deployments. Combining ML at the edge and actual hardware testing, therefore, increases the reliability of ML models to effectively improve the sustainability of IoT systems. The present systematic literature review explores how ML can be utilized to enhance the sustainability of IoT networks, examining current methodologies, benefits, challenges, and future opportunities. Through our analysis, we aim to provide insights that will drive future innovations in making IoT networks more sustainable.

eess.SY↗

arXiv2025

Analytical Modeling of Batteryless IoT Sensors Powered by Ambient Energy Harvesting

This paper presents a comprehensive mathematical model to characterize the energy dynamics of batteryless IoT sensor nodes powered entirely by ambient energy harvesting. The model captures both the energy harvesting and consumption phases, explicitly incorporating power management tasks to enable precise estimation of device behavior across diverse environmental conditions. The proposed model is applicable to a wide range of IoT devices and supports intelligent power management units designed to maximize harvested energy under fluctuating environmental conditions. We validated our model against a prototype batteryless IoT node, conducting experiments under three distinct illumination scenarios. Results show a strong correlation between analytical and measured supercapacitor voltage profiles, confirming the proposed model's accuracy.

eess.SP↗

arXiv2025

Cell-Free Massive MIMO under a Non-Linear Power Amplifier Consumption Model

Existing works on Cell-Free Massive MIMO primarily focus on optimising system throughput and energy efficiency under high-traffic scenarios with only a limited focus on variable user demand as required by higher network layers. Additionally, existing works only minimise the transmitted power instead of the consumed power at the power amplifier. This work introduces a penalty-method-based approach to minimise the amplifier's power consumption while scaling much better with network size than current solutions and promoting sparsity in the power allocated to each access point. Furthermore, we demonstrate substantial reductions in power consumption (up to 24%) by considering the non-linear power consumption.

eess.SP↗

arXiv2025

AoI in Context-Aware Hybrid Radio-Optical IoT Networks

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.

cs.CY↗

arXiv2025

Batteryless BLE and Light-based IoT Sensor Nodes for Reliable Environmental Sensing

The sustainable design of Internet of Things (IoT) networks encompasses considerations related to energy efficiency and autonomy as well as considerations related to reliable communications, ensuring no energy is wasted on undelivered data. Under these considerations, this work proposes the design and implementation of energy-efficient Bluetooth Low Energy (BLE) and Light-based IoT (LIoT) batteryless IoT sensor nodes powered by an indoor light Energy Harvesting Unit (EHU). Our design intends to integrate these nodes into a sensing network to improve its reliability by combining both technologies and taking advantage of their features. The nodes incorporate state-of-the-art components, such as low-power sensors and efficient System-on-Chips (SoCs). Moreover, we design a strategy for adaptive switching between active and sleep cycles as a function of the available energy, allowing the IoT nodes to continuously operate without batteries. Our results show that by adapting the duty cycle of the BLE and LIoT nodes depending on the environment's light intensity, we can ensure a continuous and reliable node operation. In particular, measurements show that our proposed BLE and LIoT node designs are able to communicate with an IoT gateway in a bidirectional way, every 19.3 and 624.6 seconds, respectively, in an energy-autonomous and reliable manner.

eess.SP↗

arXiv2024

Balanced Space- and Time-based Duty-cycle Scheduling for Light-based IoT

In this work, we propose a Multiple Access Control (MAC) protocol for Light-based IoT (LIoT) networks, where the gateway node orchestrates and schedules batteryless nodes duty-cycles based on their location and sleep time. The LIoT concept represents a sustainable solution for massive indoor IoT applications, offering an alternative communication medium through Visible Light Communication (VLC). While most existing scheduling algorithms for intermittent batteryless IoT aim to maximize data collection and enhance dataset size, our solution is tailored for environmental sensing applications, such as temperature, humidity, and air quality monitoring, optimizing measurement distribution and minimizing blind spots to achieve comprehensive and uniform environmental sensing. We propose a Balanced Space and Time-based Time Division Multiple Access scheduling (BST-TDMA) algorithm, which addresses environmental sensing challenges by balancing spatial and temporal factors to improve the environmental sensing efficiency of batteryless LIoT nodes. Our measurement-based results show that BST-TDMA was able to efficiently schedule duty-cycles with given intervals.

eess.SP↗

arXiv2024

Computational Efficient Width-Wise Early Exiting in Wireless Communication Systems

Deep learning (DL) techniques are increasingly pervasive across various domains, including wireless communication, where they extract insights from raw radio signals. However, the computational demands of DL pose significant challenges, particularly in distributed wireless networks like Cell-free networks, where deploying DL models on edge devices becomes hard due to heightened computational loads. These computational loads escalate with larger input sizes, often correlating with improved model performance. To mitigate this challenge, Early Exiting (EE) techniques have been introduced in DL, primarily targeting the depth of the model. This approach enables models to exit during inference based on specified criteria, leveraging entropy measures at intermediate exits. Doing so makes less complex samples exit early, reducing the average computational load and inference time. In our contribution, we propose a novel width-wise exiting strategy for Convolutional Neural Network (CNN)-based architectures. By selectively adjusting the input size, we aim to regulate computational demands effectively. Our approach aims to decrease the average computational load during inference while maintaining performance levels comparable to conventional models. We specifically investigate Modulation Classification, a well-established application of DL in wireless communication. Our experimental results show substantial reductions in computational load, with an average decrease of 26%, and particularly notable reductions of 60% in high-SNR scenarios. Through this work, we present a practical solution for reducing computational demands in deep learning applications, particularly within the domain of wireless communication.

eess.SP↗

arXiv2024

Distributed Deep Learning for Modulation Classification in 6G Cell-Free Wireless Networks

In the evolution of 6th Generation (6G) technology, the emergence of cell-free networking presents a paradigm shift, revolutionizing user experiences within densely deployed networks where distributed access points collaborate. However, the integration of intelligent mechanisms is crucial for optimizing the efficiency, scalability, and adaptability of these 6G cell-free networks. One application aiming to optimize spectrum usage is Automatic Modulation Classification (AMC), a vital component for classifying and dynamically adjusting modulation schemes. This paper explores different distributed solutions for AMC in cell-free networks, addressing the training, computational complexity, and accuracy of two practical approaches. The first approach addresses scenarios where signal sharing is not feasible due to privacy concerns or fronthaul limitations. Our findings reveal that maintaining comparable accuracy is remarkably achievable, yet it comes with an increase in computational demand. The second approach considers a central model and multiple distributed models collaboratively classifying the modulation. This hybrid model leverages diversity gain through signal combining and requires synchronization and signal sharing. The hybrid model demonstrates superior performance, achieving a 2.5% improvement in accuracy with equivalent total computational load. Notably, the hybrid model distributes the computational load across multiple devices, resulting in a lower individual computational load.

eess.SP↗

arXiv2024

On the Ground and in the Sky: A Tutorial on Radio Localization in Ground-Air-Space Networks

The inherent limitations in scaling up ground infrastructure for future wireless networks, combined with decreasing operational costs of aerial and space networks, are driving considerable research interest in multisegment ground-air-space (GAS) networks. In GAS networks, where ground and aerial users share network resources, ubiquitous and accurate user localization becomes indispensable, not only as an end-user service but also as an enabler for location-aware communications. This breaks the convention of having localization as a byproduct in networks primarily designed for communications. To address these imperative localization needs, the design and utilization of ground, aerial, and space anchors require thorough investigation. In this tutorial, we provide an in-depth systemic analysis of the radio localization problem in GAS networks, considering ground and aerial users as targets to be localized. Starting from a survey of the most relevant works, we then define the key characteristics of anchors and targets in GAS networks. Subsequently, we detail localization fundamentals in GAS networks, considering 3D positions, orientations, and velocities. Afterward, we thoroughly analyze radio localization systems in GAS networks, detailing the system model, design aspects, and considerations for each of the three GAS anchors. Preliminary results are presented to provide a quantifiable perspective on key design aspects in GAS-based localization scenarios. We then identify the vital roles 6G enablers are expected to play in radio localization in GAS networks.

cs.NI↗

arXiv2024

Location-Based Load Balancing for Energy-Efficient Cell-Free Networks

Cell-Free Massive MIMO (CF mMIMO) has emerged as a potential enabler for future networks. It has been shown that these networks are much more energy-efficient than classical cellular systems when they are serving users at peak capacity. However, these CF mMIMO networks are designed for peak traffic loads, and when this is not the case, they are significantly over-dimensioned and not at all energy efficient. To this end, Adaptive Access Point (AP) ON/OFF Switching (ASO) strategies have been developed to save energy when the network is not at peak traffic loads by putting unnecessary APs to sleep. Unfortunately, the existing strategies rely on measuring channel state information between every user and every access point, resulting in significant measurement energy consumption overheads. Furthermore, the current state-of-art approach has a computational complexity that scales exponentially with the number of APs. In this work, we present a novel convex feasibility testing method that allows checking per-user Quality-of-Service (QoS) requirements without necessarily considering all possible access point activations. We then propose an iterative algorithm for activating access points until all users' requirements are fulfilled. We show that our method has comparable performance to the optimal solution whilst avoiding solving costly mixed-integer problems and measuring channel state information on only a limited subset of APs.

eess.SP↗

arXiv2024

Electric Field Evaluation of Reconfigurable Intelligent Surface in Wireless Networks

Reconfigurable intelligent surface (RIS) used as infrastructure in wireless networks has been a trend, thanks to its low cost and high flexibility. Working in many ways including reflective mirrors and phase-shifted surfaces, RIS is able to enhance the coverage in communications and provide more degrees of freedom for sensing. However, the key issue lies in how to place RIS in accordance with the regulations for electromagnetic field (EMF) exposure, which requires refined evaluations. In this paper, we first investigate the regulations in terms of E-field. Then, relevant deployment characteristics are evaluated jointly: the minimum distance from the base station (BS) to the RIS, and the minimum height of the RIS are given for a given BS power limit and as function of the number of RIS elements. The ray-tracing simulations verify the correctness of our analysis. Besides, different frequency ranges (FRs) and radiation patterns of RIS elements are investigated. The results show that the EMF exposure risk is negligible when RIS works in the reflective-only (RO) mode. However, when it works in the beamforming (BO) mode, its placement should be well specified based on our analytical framework to comply with the regulations of E-field limit in general public scenarios. Finally, we provide an E-field measurement methodology and low-cost solutions in terms of general wireless networks and 5G standalone networks, which pave the way for real-world evaluation in future work.

eess.SY↗

arXiv2023

REM-U-net: Deep Learning Based Agile REM Prediction with Energy-Efficient Cell-Free Use Case

Radio environment maps (REMs) hold a central role in optimizing wireless network deployment, enhancing network performance, and ensuring effective spectrum management. Conventional REM prediction methods are either excessively time-consuming, e.g., ray tracing, or inaccurate, e.g., statistical models, limiting their adoption in modern inherently dynamic wireless networks. Deep-learning-based REM prediction has recently attracted considerable attention as an appealing, accurate, and time-efficient alternative. However, existing works on REM prediction using deep learning are either confined to 2D maps or use a limited dataset. In this paper, we introduce a runtime-efficient REM prediction framework based on u-nets, trained on a large-scale 3D maps dataset. In addition, data preprocessing steps are investigated to further refine the REM prediction accuracy. The proposed u-net framework, along with preprocessing steps, are evaluated in the context of the 2023 IEEE ICASSP Signal Processing Grand Challenge, namely, the First Pathloss Radio Map Prediction Challenge. The evaluation results demonstrate that the proposed method achieves an average normalized root-mean-square error (RMSE) of 0.045 with an average of 14 milliseconds (ms) runtime. Finally, we position our achieved REM prediction accuracy in the context of a relevant cell-free massive multiple-input multiple-output (CF-mMIMO) use case. We demonstrate that one can obviate consuming energy on large-scale fading measurements and rely on predicted REM instead to decide on which sleep access points (APs) to switch on in a CF-mMIMO network that adopts a minimum propagation loss AP switch ON/OFF strategy.

cs.NI↗

arXiv2023

ecoBLE: A Low-Computation Energy Consumption Prediction Framework for Bluetooth Low Energy

Bluetooth Low Energy (BLE) is a de-facto technology for Internet of Things (IoT) applications, promising very low energy consumption. However, this low energy consumption accounts only for the radio part, and it overlooks the energy consumption of other hardware and software components. Monitoring and predicting the energy consumption of IoT nodes after deployment can substantially aid in ensuring low energy consumption, calculating the remaining battery lifetime, predicting needed energy for energy-harvesting nodes, and detecting anomalies. In this paper, we introduce a Long Short-Term Memory Projection (LSTMP)-based BLE energy consumption prediction framework together with a dataset for a healthcare application scenario where BLE is widely adopted. Unlike radio-focused theoretical energy models, our framework provides a comprehensive energy consumption prediction, considering all components of the IoT node, including the radio, sensor as well as microcontroller unit (MCU). Our measurement-based results show that the proposed framework predicts the energy consumption of different BLE nodes with a Mean Absolute Percentage Error (MAPE) of up to 12%, giving comparable accuracy to state-of-the-art energy consumption prediction with a five times smaller prediction model size.

cs.NI↗

arXiv2023

QualityBLE: A QoS Aware Implementation for BLE Mesh Networks

Bluetooth Low Energy (BLE) Mesh is widely recognized as a driver technology for IoT applications. However, the lack of quality of service (QoS) in BLE Mesh, represented by packet prioritization, significantly limits its potential. This work implements a quality-of-service (QoS) method for BLE Mesh to prioritize the data packets and provide them with different network transmission settings according to their assigned priority. Unlike existing works on QoS for BLE Mesh, our proposed approach does not require any modifications to the BLE Mesh protocol and can be smoothly adopted in existing BLE Mesh networks. We conducted an extensive measurement campaign to evaluate our solution over a 15-node BLE Mesh network deployed to emulate a smart healthcare scenario where 45 sensors with an assigned priority transmit information over the network. The experiments provide performance results for single and multi channel network scenarios. The obtained results validate our solution, showing the difference between the established priorities and providing insights and guidelines to conduct further research on QoS over BLE Mesh and broadcast-based networks.

cs.NI↗