arXiv subjects

Khaled Ben Letaief

Publications and source records attributed to Khaled Ben Letaief.

At least 19 recordsLinked to original sources

arXiv2026

PCDM: A Diffusion-Based Data Poisoning Attack Against Federated Learning Systems

Federated learning (FL) is vulnerable to data poisoning attacks due to its distributed nature. Although recent GAN-based data poisoning methods have indicated the potential of using generative AI to generate seemingly legitimate poisoned data, the inherent consistency of GAN outputs can still reveal a sign of data poisoning. In this paper, we propose a diffusion-based data poisoning framework against FL systems, which leverages a Poisoning-Oriented Conditional Diffusion Model (PCDM) to enable fine-grained control over the local generation of poisoned data while ensuring both attack effectiveness and stealthiness. Our PCDM incorporates an adjustable poisoning vector within the global context to precisely control the generation of poisoned data, with theoretical guarantees on attack performance. Furthermore, it employs a novel jumping diffusion strategy for lightweight and efficient poisoned data generation. We conduct the most systematic and broad experimental evaluation for FL poisoning attacks against various defenses, including advanced Byzantine robust aggregation mechanisms, on four open datasets: MNIST, Fashion-MNIST, CIFAR-10, CIFAR-100, and a real-world wireless-specific dataset VRAI. Our results demonstrate that PCDM is less likely to exhibit statistical anomalies compared with the state-of-the-art methods while more effectively degrading global FL performance, which poses a significant risk to data security in FL.

cs.CR↗

arXiv2026

A Model Consistency-Based Countermeasure to GAN-Based Data Poisoning Attack in Federated Learning

In federated learning (FL), although the original intention of available but not visible data is to allay data privacy concerns, it potentially brings new security threats, particularly poisoning attacks that target such not visible local data. Intuitively, such data poisoning attacks have great potential in stealthily degrading global FL outcomes, and are expected to be even stealthier if being enhanced by generative models like generative adversarial networks (GANs). However, existing defense methods have not been thoroughly challenged in this regard and generally fail to be aware of a local generation of seemingly legitimate poisoned data. With a growing concern on potentially stealthier attacks, in this paper, a cost-effective defense mechanism named Model Consistency-Based Defense (MCD) is proposed, which offers a comprehensive examination of available local models across multiple feature dimensions, providing an indirect yet effective means of identifying hidden data poisoning attackers. To push the limit of MCD against stealthier attacks, we propose a new GAN-based data poisoning attack model named VagueGAN and an unsupervised variant of it, which can be flexibly deployed to generate seemingly legitimate but noisy poisoned data. The consistency of GAN outputs revealed by VagueGAN helps strengthen MCD to work against stealthier GAN-based attacks as well as other mainstream ones. Extensive experiments on multiple open datasets (MNIST, Fashion-MNIST, CIFAR-10, CIFAR-100, and Mini-Imagenet) indicate that our attack method better balances the trade-off between attack effectiveness and stealthiness with low complexity. More importantly, our defense mechanism is shown to be more competent in identifying a variety of poisoned data, particularly stealthier GAN-poisoned ones.

cs.CR↗

arXiv2026

Modular Foundation Model Inference at the Edge: Network-Aware Microservice Optimization

Foundation models (FMs) unlock unprecedented multimodal and multitask intelligence, yet their cloud-centric deployment precludes real-time responsiveness and compromises user privacy. Meanwhile, monolithic execution at the edge remains infeasible under stringent resource limits and uncertain network dynamics. To bridge this gap, we propose a microservice-based FM inference framework that exploits the intrinsic functional asymmetry between heavyweight core services and agile light services. Our two-tier deployment strategy ensures robust Quality of Service (QoS) under resource contention. Specifically, core services are placed statically via a long-term network-aware integer program with sparsity constraints to form a fault-tolerant backbone. On the other hand, light services are orchestrated dynamically by a low-complexity online controller that integrates effective capacity theory with Lyapunov optimization, providing probabilistic latency guarantees under real-time workload fluctuations. Simulations demonstrate that our framework achieves over 84% average on-time task completion with moderate deployment costs and maintains strong robustness as the system load scales.

cs.DC↗

arXiv2025

Satellite Federated Fine-Tuning for Foundation Models in Space Computing Power Networks

Advancements in artificial intelligence (AI) and low-earth orbit (LEO) satellites have promoted the application of large remote sensing foundation models for various downstream tasks. However, direct downloading of these models for fine-tuning on the ground is impeded by privacy concerns and limited bandwidth. Satellite federated learning (FL) offers a solution by enabling model fine-tuning directly on-board satellites and aggregating model updates without data downloading. Nevertheless, for large foundation models, the computational capacity of satellites is insufficient to support effective on-board fine-tuning in traditional satellite FL frameworks. To address these challenges, we propose a satellite-ground collaborative federated fine-tuning framework. The key of the framework lies in how to reasonably decompose and allocate model components to alleviate insufficient on-board computation capabilities. During fine-tuning, satellites exchange intermediate results with ground stations or other satellites for forward propagation and back propagation, which brings communication challenges due to the special communication topology of space transmission networks, such as intermittent satellite-ground communication, short duration of satellite-ground communication windows, and unstable inter-orbit inter-satellite links (ISLs). To reduce transmission delays, we further introduce tailored communication strategies that integrate both communication and computing resources. Specifically, we propose a parallel intra-orbit communication strategy, a topology-aware satellite-ground communication strategy, and a latency-minimalization inter-orbit communication strategy to reduce space communication costs. Simulation results demonstrate significant reductions in training time with improvements of approximately 33%.

cs.LG↗

arXiv2024

Two-Timescale Model Caching and Resource Allocation for Edge-Enabled AI-Generated Content Services

Generative AI (GenAI) has emerged as a transformative technology, enabling customized and personalized AI-generated content (AIGC) services. In this paper, we address challenges of edge-enabled AIGC service provisioning, which remain underexplored in the literature. These services require executing GenAI models with billions of parameters, posing significant obstacles to resource-limited wireless edge. We subsequently introduce the formulation of joint model caching and resource allocation for AIGC services to balance a trade-off between AIGC quality and latency metrics. We obtain mathematical relationships of these metrics with the computational resources required by GenAI models via experimentation. Afterward, we decompose the formulation into a model caching subproblem on a long-timescale and a resource allocation subproblem on a short-timescale. Since the variables to be solved are discrete and continuous, respectively, we leverage a double deep Q-network (DDQN) algorithm to solve the former subproblem and propose a diffusion-based deep deterministic policy gradient (D3PG) algorithm to solve the latter. The proposed D3PG algorithm makes an innovative use of diffusion models as the actor network to determine optimal resource allocation decisions. Consequently, we integrate these two learning methods within the overarching two-timescale deep reinforcement learning (T2DRL) algorithm, the performance of which is studied through comparative numerical simulations.

cs.LG↗

arXiv2023

A GAN-based Semantic Communication for Text without CSI

Recently, semantic communication (SC) has been regarded as one of the potential paradigms of 6G. Current SC frameworks require channel state information (CSI) to handle severe signal distortion induced by channel fading. Since the channel estimation overhead for obtaining CSI cannot be neglected, we therefore propose a generative adversarial network (GAN) based SC framework (Ti-GSC) that doesn't require CSI. In Ti-GSC, two main modules, i.e., an autoencoder-based encoder-decoder module (AEDM) and a GAN-based signal distortion suppression module (GSDSM) are included where AEDM first encodes the data at the source before transmission, and then GSDSM suppresses the distortion of the received signals in both syntactic and semantic dimensions at the destination. At last, AEDM decodes the distortion-suppressed signal at the destination. To measure signal distortion, syntactic distortion and semantic distortion terms are newly added to the total loss function. To achieve better training results, joint optimization-based training (JOT) and alternating optimization-based training (AOT) are designed for the proposed Ti-GSC. Experimental results show that JOT is more efficient for Ti-GSC. Moreover, without CSI, bilingual evaluation understudy (BLEU) score achieved by Ti-GSC is about 40% and 62% higher than that achieved by existing SC frameworks in Rician and Rayleigh fading, respectively. (*Due to the notification of arXiv "The Abstract field cannot be longer than 1,920 characters", the appeared Abstract is shortened. For the full Abstract, please download the Article.)

cs.IT↗

arXiv2023

Multi-Channel Attentive Feature Fusion for Radio Frequency Fingerprinting

Radio frequency fingerprinting (RFF) is a promising device authentication technique for securing the Internet of things. It exploits the intrinsic and unique hardware impairments of the transmitters for RF device identification. In real-world communication systems, hardware impairments across transmitters are subtle, which are difficult to model explicitly. Recently, due to the superior performance of deep learning (DL)-based classification models on real-world datasets, DL networks have been explored for RFF. Most existing DL-based RFF models use a single representation of radio signals as the input. Multi-channel input model can leverage information from different representations of radio signals and improve the identification accuracy of the RF fingerprint. In this work, we propose a novel multi-channel attentive feature fusion (McAFF) method for RFF. It utilizes multi-channel neural features extracted from multiple representations of radio signals, including IQ samples, carrier frequency offset, fast Fourier transform coefficients and short-time Fourier transform coefficients, for better RF fingerprint identification. The features extracted from different channels are fused adaptively using a shared attention module, where the weights of neural features from multiple channels are learned during training the McAFF model. In addition, we design a signal identification module using a convolution-based ResNeXt block to map the fused features to device identities. To evaluate the identification performance of the proposed method, we construct a WiFi dataset, named WFDI, using commercial WiFi end-devices as the transmitters and a Universal Software Radio Peripheral (USRP) as the receiver. ...

eess.SP↗

arXiv2022

How Far Are Wireless Networks from Being Truly Deterministic?

With the rapid development of Internet-of-Things (IoT) technology and machine-type communications, various emerging applications appear in industrial productions and our daily lives. Among these, applications like industrial sensing and controlling, remote surgery, and automatic driving require an extremely low latency and a very small jitter. Delivering information deterministically has become one of the the biggest challenges for modern wire-line and wireless communications. In this paper, we present a review of currently available wire-line deterministic networks and discuss the main challenges to build wireless deterministic networks. We also discuss and propose several potential techniques enabling wireless networks to provide deterministic communications. By elaborating the coding/modulation schemes of the physical layer and managing the channel-access/packet-scheduling at the media access control (MAC) layer, it is believed that wireless deterministic communications can be realized in the near future.

cs.IT↗

arXiv2019

Energy Harvesting Powered Sensing in IoT: Timeliness Versus Distortion

We consider an Internet-of-Things (IoT) system in which an energy harvesting powered sensor node monitors the phenomenon of interest and transmits its observations to a remote monitor over a Gaussian channel. We measure the timeliness of the signals recovered by the monitor using age of information (AoI), which could be reduced by transmitting more observations to the monitor. We evaluate the corresponding distortion with the mean-squared error (MSE) metric, which would be reduced if a larger transmit power and a larger source coding rate were used. Since the energy harvested by the sensor node is random and limited, however, the timeliness and the distortion of the received signals cannot be optimized at the same time. Thus, we shall investigate the timeliness-distortion trade-off of the system by minimizing the average weighted-sum AoI and distortion over all possible transmit powers and transmission intervals. First, we explicitly present the optimal transmit powers for the performance limit achieving save-and-transmit policy and the easy-implementing fixed power transmission policy. Second, we propose a backward water-filling based offline power allocation algorithm and a genetic based offline algorithm to jointly optimize the transmission interval and transmit power. Third, we formulate the online power control as an Markov Decision Process (MDP) and solve the problem with an iterative algorithm, which closely approach the trade-off limit of the system. Also, we show that the optimal transmit power is a monotonic and bi-valued function of current AoI and distortion. Finally, we present our results via numerical simulations and extend results on the save-and-transmit policy to fading sensing systems.

cs.IT↗

arXiv2019

Age-upon-Decisions Minimizing Scheduling in Internet of Things: To be Random or to be Deterministic?

We consider an Internet of Things (IoT) system in which a sensor delivers updates to a monitor with exponential service time and first-come-first-served (FCFS) discipline. We investigate the freshness of the received updates and propose a new metric termed as \textit{Age upon Decisions (AuD)}, which is defined as the time elapsed from the generation of each update to the epoch it is used to make decisions (e.g., estimations, inferences, controls). Within this framework, we aim at improving the freshness of updates at decision epochs by scheduling the update arrival process and the decision making process. Theoretical results show that 1) when the decisions are made according to a Poisson process, the average AuD is independent of decision rate and would be minimized if the arrival process is periodic (i.e., deterministic); 2) when both the decision process and the arrive process are periodic, the average AuD is larger than, but decreases with decision rate to, the average AuD of the corresponding system with Poisson decisions (i.e., random); 3) when both the decision process and the arrive process are periodic, the average AuD can be further decreased by optimally controlling the offset between the two processes. For practical IoT systems, therefore, it is suggested to employ periodic arrival processes and random decision processes. Nevertheless, making periodical updates and decisions with properly controlled offset also is a promising solution if the timing information of the two processes can be accessed by the monitor.

cs.IT↗

arXiv2019

Fog-Assisted Multi-User SWIPT Networks: Local Computing or Offloading

This paper investigates a fog computing-assisted multi-user simultaneous wireless information and power transfer (SWIPT) network, where multiple sensors with power splitting (PS) receiver architectures receive information and harvest energy from a hybrid access point (HAP), and then process the received data by using local computing mode or fog offloading mode. For such a system, an optimization problem is formulated to minimize the sensors' required energy while guaranteeing their required information transmissions and processing rates by jointly optimizing the multi-user scheduling, the time assignment, the sensors' transmit powers and the PS ratios. Since the problem is a mixed integer programming (MIP) problem and cannot be solved with existing solution methods, we solve it by applying problem decomposition, variable substitutions and theoretical analysis. For a scheduled sensor, the closed-form and semi-closedform solutions to achieve its minimal required energy are derived, and then an efficient multi-user scheduling scheme is presented, which can achieve the suboptimal user scheduling with low computational complexity. Numerical results demonstrate our obtained theoretical results, which show that for each sensor, when it is located close to the HAP or the fog server (FS), the fog offloading mode is the better choice; otherwise, the local computing mode should be selected. The system performances in a frame-by-frame manner are also simulated, which show that using the energy stored in the batteries and that harvested from the signals transmitted by previous scheduled sensors can further decrease the total required energy of the sensors.

cs.IT↗

arXiv2019

Optimal Online Transmission Policy in Wireless Powered Networks with Urgency-aware Age of Information

This paper investigates the age of information (AoI) for a radio frequency (RF) energy harvesting (EH) enabled network, where a sensor first scavenges energy from a wireless power station and then transmits the collected status update to a sink node. To capture the thirst for the fresh update becoming more and more urgent as time elapsing, urgency-aware AoI (U-AoI) is defined, which increases exponentially with time between two received updates. Due to EH, some waiting time is required at the sensor before transmitting the status update. To find the optimal transmission policy, an optimization problem is formulated to minimize the long-term average U-AoI under constraint of energy causality. As the problem is non-convex and with no known solution, a two-layer algorithm is presented to solve it, where the outer loop is designed based on Dinklebach's method, and in the inner loop, a semi-closed-form expression of the optimal waiting time policy is derived based on Karush-Kuhn-Tucker (KKT) optimality conditions. Numerical results shows that our proposed optimal transmission policy outperforms the the zero time waiting policy and equal time waiting policy in terms of long-term average U-AoI, especially when the networks are non-congested. It is also observed that in order to achieve the lower U-AoI, the sensor should transmit the next update without waiting when the network is congested while should wait a moment before transmitting the next update when the network is non-congested. Additionally, it also shows that the system U-AoI first decreases and then keep unchanged with the increments of EH circuit's saturation level and the energy outage probability.

cs.IT↗

arXiv2019

Optimal Design of SWIPT-Aware Fog Computing Networks

This paper studies a simultaneous wireless information and power transfer (SWIPT)-aware fog computing network, where a multiple antenna fog function integrated hybrid access point (F-HAP) transfers information and energy to multiple heterogeneous single-antenna sensors and also helps some of them fulfill computing tasks. By jointly optimizing energy and information beamforming designs at the F-HAP, the bandwidth allocation and the computation offloading distribution, an optimization problem is formulated to minimize the required energy under communication and computation requirements, as well as energy harvesting constraints. Two optimal designs, i.e., fixed offloading time (FOT) and optimized offloading time (OOT) designs, are proposed. As both designs get involved in solving non-convex problems, there are no known solutions to them. Therefore, for the FOT design, the semidefinite relaxation (SDR) is adopted to solve it. It is theoretically proved that the rank-one constraints are always satisfied, so the global optimal solution is guaranteed. For the OOT design, since its non-convexity is hard to deal with, a penalty dual decomposition (PDD)-based algorithm is proposed, which is able to achieve a suboptimal solution. The computational complexity for two designs are analyzed. Numerical results show that the partial offloading mode is superior to binary benchmark modes. It is also shown that if the system is with strong enough computing capability, the OOT design is suggested to achieve lower required energy; Otherwise, the FOT design is preferred to achieve a relatively low computation complexity.

cs.CY↗

arXiv2017

Directivity-Beamwidth Tradeoff of Massive MIMO Uplink Beamforming for High Speed Train Communication

High-mobility adaption and massive Multiple-input Multiple-output (MIMO) application are two primary evolving objectives for the next generation high speed train (HST) wireless communication system. In this paper, we consider how to design a location-aware beamforming for the massive MIMO system in the high traffic density HST network. We first analyze the tradeoff between beam directivity and beamwidth, based on which we present the sensitivity analysis of positioning accuracy. Then, in order to guarantee a high efficient transmission, we derive an optimal problem to maximize the beam directivity under the restriction of diverse positioning accuracies. After that, we present a low-complexity beamforming design by utilizing location information, which requires neither eigen-decomposing (ED) the uplink channel covariance matrix (CCM) nor ED the downlink CCM (DCCM). Finally, we study the beamforming scheme in future high traffic density HST network, where a two HSTs encountering scenario is emphasized. By utilizing the real-time location information, we propose an optimal adaptive beamforming scheme to maximize the achievable rate region under limited channel source constraint. Numerical simulation indicates that a massive MIMO system with less than a certain positioning error can guarantee a required performance with satisfying transmission efficiency in the high traffic density HST scenario and the achievable rate region when two HSTs encounter is greatly improved as well.

cs.IT↗

arXiv2017

Group Cooperation with Optimal Resource Allocation in Wireless Powered Communication Networks

This paper considers a wireless powered communication network (WPCN) with group cooperation, where two communication groups cooperate with each other via wireless power transfer and time sharing to fulfill their expected information delivering and achieve "win-win" collaboration. To explore the system performance limits, we formulate optimization problems to respectively maximize the weighted sum-rate and minimize the total consumed power. The time assignment, beamforming vector and power allocation are jointly optimized under available power and quality of service requirement constraints of both groups. For the WSR-maximization, both fixed and flexible power scenarios are investigated. As all problems are non-convex and have no known solution methods, we solve them by using proper variable substitutions and the semi-definite relaxation. We theoretically prove that our proposed solution method guarantees the global optimum for each problem. Numerical results are presented to show the system performance behaviors, which provide some useful insights for future WPCN design. It shows that in such a group cooperation-aware WPCN, optimal time assignment has the greatest effect on the system performance than other factors.

cs.IT↗

arXiv2017

LRC: Dependency-Aware Cache Management for Data Analytics Clusters

Memory caches are being aggressively used in today's data-parallel systems such as Spark, Tez, and Piccolo. However, prevalent systems employ rather simple cache management policies--notably the Least Recently Used (LRU) policy--that are oblivious to the application semantics of data dependency, expressed as a directed acyclic graph (DAG). Without this knowledge, memory caching can at best be performed by "guessing" the future data access patterns based on historical information (e.g., the access recency and/or frequency), which frequently results in inefficient, erroneous caching with low hit ratio and a long response time. In this paper, we propose a novel cache replacement policy, Least Reference Count (LRC), which exploits the application-specific DAG information to optimize the cache management. LRC evicts the cached data blocks whose reference count is the smallest. The reference count is defined, for each data block, as the number of dependent child blocks that have not been computed yet. We demonstrate the efficacy of LRC through both empirical analysis and cluster deployments against popular benchmarking workloads. Our Spark implementation shows that, compared with LRU, LRC speeds up typical applications by 60%.

cs.DC↗

arXiv2016

Joint Subcarrier and CPU Time Allocation for Mobile Edge Computing

In mobile edge computing systems, mobile devices can offload compute-intensive tasks to a nearby cloudlet,so as to save energy and extend battery life. Unlike a fully-fledged cloud, a cloudlet is a small-scale datacenter deployed at a wireless access point, and thus is highly constrained by both radio and compute resources. We show in this paper that separately optimizing the allocation of either compute or radio resource, as most existing works did, is highly suboptimal: the congestion of compute resource leads to the waste of radio resource, and vice versa. To address this problem, we propose a joint scheduling algorithm that allocates both radio and compute resources coordinately. Specifically, we consider a cloudlet in an Orthogonal Frequency-Division Multiplexing Access (OFDMA) system with multiple mobile devices, where we study subcarrier allocation for task offloading and CPU time allocation for task execution in the cloudlet. Simulation results show that the proposed algorithm significantly outperforms per-resource optimization, accommodating more offloading requests while achieving salient energy saving.

cs.IT↗

arXiv2016

Optimum Transmission Policies for Energy Harvesting Sensor Networks Powered By a Mobile Control Center

Wireless energy transfer, namely RF-based energy harvesting, is a potential way to prolong the lifetime of energy-constrained devices, especially in wireless sensor networks. However, due to huge propagation attenuation, its energy efficiency is regarded as the biggest bottleneck to widely applications. It is critical to find appropriate transmission policies to improve the global energy efficiency in this kind of systems. To this end, this paper focuses on the sensor networks scenario, where a mobile control center powers the sensors by RF signal and also collects information from them. Two related schemes, called as harvest-and-use scheme and harvest-store-use scheme, are investigated, respectively. In harvest-and-use scheme, as a benchmark, both constant and adaptive transmission modes from sensors are discussed. To harvest-store-use scheme, we propose a new concept, the best opportunity for wireless energy transfer, and use it to derive an explicit closed-form expression of optimal transmission policy. It is shown by simulation that a considerable improvement in terms of energy efficiency can be obtained with the help of the transmission policies developed in this paper. Furthermore, the transmission policies is also discussed under the constraint of fixed information rate. The minimal required power, the performance loss from the new constraint as well as the effect of fading are then presented.

cs.IT↗