arXiv subjects

Giovanni Perin

Publications and source records attributed to Giovanni Perin.

15 recordsLinked to original sources

arXiv2026

A MEC-Based Optimization Framework for Dynamic Inductive Charging

Range anxiety and long recharging times remain critical barriers to electric vehicle adoption. Dynamic Inductive Charging (DIC) offers a compelling solution by enabling wireless power transfer while driving, potentially reducing battery size requirements and thus vehicle costs. However, DIC infrastructures are expensive and power-constrained, requiring intelligent resource allocation to maximize user satisfaction and economic viability. We propose a Model Predictive Control framework for optimal power allocation in DIC systems, using edge computing and vehicular communications to prioritize vehicles with critical battery states. The framework is implemented and evaluated through SUMO-based simulations on a realistic 10 km urban scenario in Istanbul, Turkey, under varying traffic intensities. Results demonstrate two critical limitations of uncoordinated allocation. First, resource utilization remains suboptimal despite available power when demand saturates system capacity. Second, when demand exceeds capacity, uniform distribution of power leaves a heavy tail of critically unsatisfied vehicles that may require emergency stops. Our MPC-based strategy addresses both regimes -- maximizing power utilization during saturation through dynamic stripe rebalancing, and improving satisfaction fairness under scarcity by aggressively prioritizing depleted batteries at the expense of well-charged vehicles. The framework and simulation tools are released as open-source to support further research in this emerging domain.

eess.SY↗

arXiv2026

BRISC: A Dataset of Channel Measurements at 5 GHz With a Reflective Intelligent Surface

We introduce the broadband reconfigurable intelligent surface (RIS) channel (BRISC) dataset. The dataset comprises measurements of channel state information (CSI) collected at 5.53 GHz using a 256-element RIS with binary states. In the measurement campaign, the transmitter and receiver are two software defined radios (SDRs), phase-synchronized via an OctoClock, where the transmitter (receiver) is equipped with one (two) antenna(s). To manage complexity, the RIS elements are grouped into blocks of different sizes, where all elements within a block share the same state. CSIs have been captured for multiple a) transmitter positions (and fixed receiver location), b) pilot block sizes, and c) state configurations. Furthermore, we calibrated the parameters of state-of-the-art RIS channel models to fit the measured CSI. With approximately 10000 configurations explored per transmitting position, BRISC serves as a robust benchmark in communication applications. We also show here an example of its use for physical-layer authentication.

eess.SP↗

arXiv2026

Federated Learning Meets Random Access: Energy-Efficient Uplink Resource Allocation

Artificial intelligence-generated traffic is changing the shape of wireless networks. Specifically, as the amount of data generated to train machine learning models is massive, network resources must be carefully allocated to continue supporting standard applications. In this paper, we tackle the problem of allocating radio resources for two sets of concurrent devices communicating in uplink with a gateway over the same bandwidth. A set of devices performs federated learning (FL), and accesses the medium in FDMA, uploading periodically large models. The other set is throughput-oriented and accesses the medium via random access (RA), either with ALOHA or slotted-ALOHA protocols. We derive close-to-optimal solutions to the non-convex problem of minimizing the system energy consumption subject to FL latency and RA throughput constraints. Our solutions show that ALOHA can sustain high FL efficiency, yielding up to 48% lower consumption when the system is dominated by FL traffic. On the other hand, slotted-ALOHA becomes more efficient when RA traffic dominates, yielding 6% lower consumption.

cs.NI↗

arXiv2025

Feature-Based Semantics-Aware Scheduling for Energy-Harvesting Federated Learning

Federated Learning (FL) on resource-constrained edge devices faces a critical challenge: The computational energy required for training Deep Neural Networks (DNNs) often dominates communication costs. However, most existing Energy-Harvesting FL (EHFL) strategies fail to account for this reality, resulting in wasted energy due to redundant local computations. For efficient and proactive resource management, algorithms that predict local update contributions must be devised. We propose a lightweight client scheduling framework using the Version Age of Information (VAoI), a semantics-aware metric that quantifies update timeliness and significance. Crucially, we overcome VAoI's typical prohibitive computational cost, which requires statistical distance over the entire parameter space, by introducing a feature-based proxy. This proxy estimates model redundancy using intermediate-layer extraction from a single forward pass, dramatically reducing computational complexity. Experiments conducted under extreme non-IID data distributions and scarce energy availability demonstrate superior learning performance while achieving energy reduction compared to existing baseline selection policies. Our framework establishes semantics-aware scheduling as a practical and vital solution for EHFL in realistic scenarios where training costs dominate transmission costs.

cs.LG↗

arXiv2025

ADMM-Based Training for Spiking Neural Networks

In recent years, spiking neural networks (SNNs) have gained momentum due to their high potential in time-series processing combined with minimal energy consumption. However, they still lack a dedicated and efficient training algorithm. The popular backpropagation with surrogate gradients, adapted from stochastic gradient descent (SGD)-derived algorithms, has several drawbacks when used as an optimizer for SNNs. Specifically, the approximation introduced by the use of surrogate gradients leads to numerical imprecision, poor tracking of SNN firing times at training time, and, in turn, poor scalability. In this paper, we propose a novel SNN training method based on the alternating direction method of multipliers (ADMM). Our ADMM-based training aims to solve the problem of the SNN step function's non-differentiability by taking an entirely new approach with respect to gradient backpropagation. For the first time, we formulate the SNN training problem as an ADMM-based iterative optimization, derive closed-form updates, and empirically show the optimizer's convergence, its great potential, and discuss future and promising research directions to improve the method to different layer types and deeper architectures.

cs.LG↗

arXiv2025

LightSNN: Lightweight Architecture Search for Sparse and Accurate Spiking Neural Networks

Spiking Neural Networks (SNNs) are highly regarded for their energy efficiency, inherent activation sparsity, and suitability for real-time processing in edge devices. However, most current SNN methods adopt architectures resembling traditional artificial neural networks (ANNs), leading to suboptimal performance when applied to SNNs. While SNNs excel in energy efficiency, they have been associated with lower accuracy levels than traditional ANNs when utilizing conventional architectures. In response, in this work we present LightSNN, a rapid and efficient Neural Network Architecture Search (NAS) technique specifically tailored for SNNs that autonomously leverages the most suitable architecture, striking a good balance between accuracy and efficiency by enforcing sparsity. Based on the spiking NAS network (SNASNet) framework, a cell-based search space including backward connections is utilized to build our training-free pruning-based NAS mechanism. Our technique assesses diverse spike activation patterns across different data samples using a sparsity-aware Hamming distance fitness evaluation. Thorough experiments are conducted on both static (CIFAR10 and CIFAR100) and neuromorphic datasets (DVS128-Gesture). Our LightSNN model achieves state-of-the-art results on CIFAR10 and CIFAR100, improves performance on DVS128Gesture by 4.49\%, and significantly reduces search time most notably offering a $98\times$ speedup over SNASNet and running 30\% faster than the best existing method on DVS128Gesture. Code is available on Github at: https://github.com/YesmineAbdennadher/LightSNN.

cs.NE↗

arXiv2025

Static and Repeated Cooperative Games for the Optimization of the AoI in IoT Networks

Wireless sensing and the internet of things (IoT) are nowadays pervasive in 5G and beyond networks, and they are expected to play a crucial role in 6G. However, a centralized optimization of a distributed system is not always possible and cost-efficient. In this paper, we analyze a setting in which two sensors collaboratively update a common server seeking to minimize the age of information (AoI) of the latest sample of a common physical process. We consider a distributed and uncoordinated setting where each sensor lacks information about whether the other decides to update the server. This strategic setting is modeled through game theory (GT) and two games are defined: i) a static game of complete information with an incentive mechanism for cooperation, and ii) a repeated game over a finite horizon where the static game is played at each stage. We perform a mathematical analysis of the static game finding three Nash Equilibria (NEs) in pure strategies and one in mixed strategies. A numerical simulation of the repeated game is also presented and novel and valuable insight into the setting is given thanks to the definition of a new metric, the price of delayed updates (PoDU), which shows that the decentralized solution provides results close to the centralized optimum.

cs.NI↗

arXiv2024

Decentralized LLM Inference over Edge Networks with Energy Harvesting

Large language models have significantly transformed multiple fields with their exceptional performance in natural language tasks, but their deployment in resource-constrained environments like edge networks presents an ongoing challenge. Decentralized techniques for inference have emerged, distributing the model blocks among multiple devices to improve flexibility and cost effectiveness. However, energy limitations remain a significant concern for edge devices. We propose a sustainable model for collaborative inference on interconnected, battery-powered edge devices with energy harvesting. A semi-Markov model is developed to describe the states of the devices, considering processing parameters and average green energy arrivals. This informs the design of scheduling algorithms that aim to minimize device downtimes and maximize network throughput. Through empirical evaluations and simulated runs, we validate the effectiveness of our approach, paving the way for energy-efficient decentralized inference over edge networks.

cs.DC↗

arXiv2023

VREM-FL: Mobility-Aware Computation-Scheduling Co-Design for Vehicular Federated Learning

Assisted and autonomous driving are rapidly gaining momentum and will soon become a reality. Artificial intelligence and machine learning are regarded as key enablers thanks to the massive amount of data that smart vehicles will collect from onboard sensors. Federated learning is one of the most promising techniques for training global machine learning models while preserving data privacy of vehicles and optimizing communications resource usage. In this article, we propose vehicular radio environment map federated learning (VREM-FL), a computation-scheduling co-design for vehicular federated learning that combines mobility of vehicles with 5G radio environment maps. VREM-FL jointly optimizes learning performance of the global model and wisely allocates communication and computation resources. This is achieved by orchestrating local computations at the vehicles in conjunction with transmission of their local models in an adaptive and predictive fashion, by exploiting radio channel maps. The proposed algorithm can be tuned to trade training time for radio resource usage. Experimental results demonstrate that VREM-FL outperforms literature benchmarks for both a linear regression model (learning time reduced by 28%) and a deep neural network for semantic image segmentation (doubling the number of model updates within the same time window).

eess.SY↗

arXiv2023

Static and dynamic jamming games over wireless channels with mobile strategic players

We study a wireless jamming problem consisting of the competition between a legitimate receiver and a jammer, modeled as a zero-sum game where the value to maximize/minimize is the channel capacity at the receiver's side. Most approaches in the literature consider the two players to be stationary nodes. Instead, we investigate what happens when they can change location, specifically moving along a linear geometry. We frame this at first as a static game, which can be solved in closed form, and subsequently we extend it to a dynamic game under three different versions concerning completeness/perfection of mutual information about the adversary's position, corresponding to different assumptions of concealment/sequentiality of the moves, respectively. We first provide some theoretical conditions that hold for the static game and also help identify good strategies valid under any setup, including dynamic games. Because dynamic games, although more realistic, have a significantly larger strategy space, we use reinforcement learning to obtain efficient strategies that lead to equilibrium outcomes. We show how theoretical findings can be used to train smart agents to play the game and validate our approach in practical settings.

cs.GT↗

arXiv2023

Divide and Save: Splitting Workload Among Containers in an Edge Device to Save Energy and Time

The increasing demand for edge computing is leading to a rise in energy consumption from edge devices, which can have significant environmental and financial implications. To address this, in this paper we present a novel method to enhance the energy efficiency while speeding up computations by distributing the workload among multiple containers in an edge device. Experiments are conducted on two Nvidia Jetson edge boards, the TX2 and the AGX Orin, exploring how using a different number of containers can affect the energy consumption and the computational time for an inference task. To demonstrate the effectiveness of our splitting approach, a video object detection task is conducted using an embedded version of the state-of-the-art YOLO algorithm, quantifying the energy and the time savings achieved compared to doing the computations on a single container. The proposed method can help mitigate the environmental and economic consequences of high energy consumption in edge computing, by providing a more sustainable approach to managing the workload of edge devices.

cs.DC↗

arXiv2022

Energy Efficient Deployment and Orchestration of Computing Resources at the Network Edge: a Survey on Algorithms, Trends and Open Challenges

Mobile networks are becoming energy hungry, and this trend is expected to continue due to a surge in communication and computation demand. Multi-access Edge Computing (MEC), will entail energy-consuming services and applications, with non-negligible impact in terms of ecological sustainability. In this paper, we provide a comprehensive review of existing approaches to make edge computing networks greener, including but not limited to the exploitation of renewable energy resources, and context-awareness. We hence provide an updated account of recent developments on MEC from an energetic sustainability perspective, addressing the initial deployment of computing resources, their dynamic (re)allocation, as well as distributed and federated learning designs. In doing so, we highlight the energy aspects of these algorithms, advocating the need for energy-sustainable edge computing systems that are aligned with Sustainable Development Goals (SDGs) and the Paris agreement. To the best of our knowledge, this is the first work providing a systematic literature review on the efficient deployment and management of energy harvesting MEC, with special focus on the deployment, provisioning, and scheduling of computing tasks, including federated learning for distributed edge intelligence, toward making edge networks greener and more sustainable. At the end of the paper, open research avenues and challenges are identified for all the surveyed topics.

cs.NI↗

arXiv2021

EASE: Energy-Aware Job Scheduling for Vehicular Edge Networks With Renewable Energy Resources

The energy sustainability of multi-access edge computing (MEC) platforms is here addressed by developing Energy-Aware job Scheduling at the Edge (EASE), a computing resource scheduler for edge servers co-powered by renewable energy resources and the power grid. The scenario under study involves the optimal allocation and migration of time-sensitive computing tasks in a resource-constrained internet of vehicles (IoV) context. This is achieved by tackling, as the main objective, the minimization of the carbon footprint of the edge network, whilst delivering adequate quality of service (QoS) to the end users (e.g., meeting task execution deadlines). EASE integrates i) a centralized optimization step, solved through model predictive control (MPC), to manage the renewable energy that is locally collected at the edge servers and their local computing resources, estimating their future availability, and ii) a distributed consensus step, solved via dual ascent in closed form, to reach agreement on service migrations. EASE is compared with four existing migration strategies. Quantitative results demonstrate its greater energy efficiency, which often gets close to complete carbon neutrality, while also improving the QoS.

cs.NI↗

arXiv2019

Impact of cation redox chemistry on continuous hydrothermal synthesis of 2D-Ni(Co/Fe) hydroxides

Continuous hydrothermal flow synthesis (CHFS) is a facile, upscalable and cost-efficient synthetic method enabling the nanostructuring of advanced functional materials in steady conditions, i.e. not in batch synthesis. In this paper, we use CHFS to crystallize NiCo- and NiFe-hydroxides in water solution with 2D nanofeatures. By tuning the synthetic parameters, we disclose the key role of the cation redox chemistry in the transition between two competitive phases: from 2D-nanoplatelets of brucite to layered double hydroxides (LDH). For controlling the precipitation of different Ni, Fe, Co-hydroxide phases, we propose the combined use of an oxidizing (H2O2) and a complexing (NH3) agent. At temperatures as low as 80 °C, the presence of H2O2 and a low concentration of NH3 favour the Ni2+/Co3+ over Ni2+/Co2+ oxidation states, shifting the product structure from brucite phase (temperatures > 80 °C) to LDH. Conversely, for the NiFe-hydroxides the transition from LDH (temperatures < 80 °C) to brucite phase (temperatures > 80 °C) is controlled by the reaction temperature only. Due to the high stability of Fe3+, the synthesis of NiFe products by CHFS does not require oxidizing and complexing agents, resulting in a robust process for large-scale production.

cond-mat.mtrl-sci↗

arXiv2019

Gd0.2Ce0.8O1.9/Y0.16Zr0.84O1.92 nanocomposite thin films for low temperature ionic conductivity

Gd0.2Ce0.8O1.9/Y0.16Zr0.84O1.92 (GDC/YSZ) nanocomposite is synthesized by a novel hybrid chemical route, where colloidal crystalline GDC nanoparticles from continuous hydrothermal synthesis are dispersed into a metalorganic YSZ matrix precursor. The result is a mixture of metal oxides in which GDC nanoparticles are finely distributed in a continuous metalorganic polymeric matrix to be crystallized after calcination. The GDC nanoparticles reduce the temperature necessary to obtain crystalline YSZ, which is already formed at 400 °C. The nanocomposite reveals structural stability up to 800 °C when treated in both air and reducing atmosphere, showing the onset of diffusion below 1000 °C. The diffusional processes are largely dependent on the nanometric grain size, with Zr4+ diffusing abruptly towards GDC in air at 1000 °C and GDC/YSZ interdiffusion being hindered in reducing environment despite the onset temperature of 900 °C. The nanocomposite precursor is an inkjet-printable reactive water-based material, suitable for the deposition of thin films with a thickness below 100 nm after calcination at 750 °C. The crystal structure of the film reveals no interaction between GDC and YSZ but a microstrain (0.3% tensile strain for YSZ). The thin film microstructure shows a compact layer with 94% density. The nanocomposite shows high oxygen ionic conductivity at low temperatures (> 5 * 10-3 S cm-1 at 500 °C), low activation energy (0.55 eV), and dominant oxygen ionic conductivity even in reducing conditions (pO2 < 10-25 atm). We show that these properties arise from the large interface between the components of the composite, due to the embedding of the GDC nanoparticles in the YSZ matrix, while ZrO-CeO intermixing can be avoided and no n-type conductivity is observed even at low oxygen activities and high temperatures.

cond-mat.mtrl-sci↗