Federated learning (FL) is susceptible to poisoning attacks. To defend against such threats, robust aggregation rules (AGRs) are typically deployed on the server to identify or filter clients’ potentially malicious submissions based on statistical similarity. Recently, knowledge distillation (KD) has been widely used in FL to facilitate collaborative learning among clients that have heterogeneous model architectures by aggregating and distilling architecture-independent model outputs (i.e., logits). However, the KD process introduces a novel poisoning attack surface, where adversaries can manipulate local model output logits to ruin the global model performance. To fully reveal and explore such a new security vulnerability and effectively poison the global model in the existence of robust AGRs, in this paper, we propose the first untargeted poisoning attack scheme to KD-based FL under robust AGRs, named ManipulatingKD. It manipulates compromised clients to send well-designed malicious logits during the KD process. To ensure attack effectiveness and stealthiness, ManipulatingKD models attacks as constrained optimization problems. This allows for crafting satisfactory malicious logits that are statistically similar to benign logits but can generate poisoned aggregated logits to provide deviated supervision and mislead the global model. Extensive experiments demonstrate the effectiveness of ManipulatingKD under both non-robust and robust AGRs. Particularly, under robust AGRs, the global model accuracy degradation caused by our attacks can exceed 2× that of state-of-the-art attacks.

In mobile edge computing (MEC), users can offload tasks to nearby MEC servers to reduce computation cost. Considering that the size of offloaded tasks could disclose user location information, several location privacy-preserving task offloading mechanisms have been proposed under the single-server scenario. However, to the best of our knowledge, none of them could provide a strict privacy protection guarantee or be applicable to the multi-server scenario where the user's location can be inferred more accurately if servers collude with each other. In this paper, we propose a novel location privacy-aware task offloading framework (LPA-Offload) for both single-server and multi-server scenarios, which provides strict and provable location privacy protection while achieving efficient task offloading. Specifically, we propose a location perturbation mechanism that allows each user to perturb its real location within a rational perturbation region and provides a differential privacy guarantee. To make a satisfactory offloading strategy, we propose a perturbation region determination mechanism and an offloading strategy generation mechanism that adaptively select a proper perturbation region according to the customized privacy factor, and then generate an optimal offloading strategy based on the perturbed location within the decided region. The determination of the perturbation region could achieve personalized privacy requirements while reducing computation cost. LPA-Offload is proved to satisfy$(\epsilon,\delta)$-differential privacy, and the experiments demonstrate the effectiveness of our framework.

Traditional traffic classification methods struggle to identify emerging network traffic due to the need for model retraining, which hampers the real-time response of deployed edge devices. Furthermore, emerging network traffic samples are often scarce, traditional methods often treat a session as a single image, thereby overlooking essential structural features. These factors can result in poor generalization ability of the trained model. To overcome these challenges, we propose ConViTML (Convolutional Vision Transformer-based Meta-Learning), a real-time end-to-end network traffic classification framework that employs meta-learning to avoid model retraining. We propose a novel feature extraction network, Convolutional Visual Transformer (ConViT), merging Convolutional Neural Network (CNN) and Visual Transformer (ViT). ConViT can directly extract low-dimensional discriminative features containing basic and structural features of the session, which is vital for improving detection accuracy and accelerating convergence in a data-scarce environment. Furthermore, we employ a Packet-based Relation Network (PRN) to analyze the matching degree of support samples and query samples. Therefore, accurate classification in novel traffic identification tasks can be achieved with just a few labeled samples, eliminating extensive data collection and labeling operations. Finally, we replace various feature extractors and compare our approach with the classic meta-learning framework Relation Network (RelationNet). Extensive experimental results demonstrate that ConViTML outperforms others with various performance indicators.

Mobile edge computing (MEC) can use wireless access network (RAN) to provide users with nearby information technology (IT) services and cloud computing functions, which creates a high-performance and low latency service environment. By caching the popular content at small base station (SBS) can reduce the heavy backhaul load and the content retransmission. However, the time-varying and dynamic of the content requests may lead to the base station to cache the useless contents. In this paper, we study a distributed caching optimization problem in edge networks (ENs) with the spatio-temporal requirements. In the considered ENs, the cache control is described as a stochastic differential game (SDG) in which each SBS defines a caching strategy to reduce the total cost in terms of the service delay and backhaul link load. To reduce the computational complexity, the original optimization problem is transformed into a mean field game (MFG). We propose a distributed caching iterative control algorithm that decouples the information interaction between the general SBS and others through the mean field distribution. In addition, we obtain the optimal edge caching control strategy, while the existence and uniqueness of the mean field equilibrium (MFE) can also be guaranteed. Simulation results demonstrate that our proposed caching control algorithm can average reduce 27.12% storage cost and achieve better performance than other existing schemes.

Truth discovery is an effective paradigm which could reveal the truth from crowdsouced data with conflicts, enabling data-driven decision-making systems to make quick and smart decisions. The increasing privacy concern promotes users to perturb or encrypt their private data before outsourcing, which poses significant challenges for truth discovery. Although several privacy-preserving truth discovery mechanisms have been proposed, none of them take personal privacy expectation into consideration. In this work, we propose a novel personalized privacy-preserving truth discovery (PPPTD) framework over crowdsourced data streams to achieve timely and accurate truth discovery while guaranteeing the protection of individual privacy. The key challenges of PPPTD lie in improving the accuracy of truth estimation from the perturbed streaming data with personalized protection level. To address these challenges, we first develop a personalized budget initialization mechanism to quantify each user’s privacy protection requirement, and allocate personalized privacy budgets to users according to their privacy requirements. Then we propose a deviation-aware weighted aggregation method to improve the accuracy of truth discovery from streaming data with varying degrees of perturbation. In order to achieve privacy-utility tradeoff, we further propose an influence-aware adaptive budget adjustment mechanism that adaptively re-allocates privacy budgets to users based on the evolution of their influence in the weighted aggregation. We prove that PPPTD can achieve$\epsilon $-differential privacy over the whole data generated by users and satisfy individual personalized privacy requirements. Extensive experiments on two real-world datasets demonstrate the effectiveness of PPPTD.

With the aggravation of data explosion and backhaul loads on 5 G edge network, it is difficult for traditional centralized cloud to meet the low latency requirements for content access. The federated learning (FL)-basedproactive contentcaching (FPC) can alleviate the matter by placing content in local cache to achieve fast and repetitive data access while protecting the users’ privacy. However, due to the non-independent and identically distributed (Non-IID) data across the clients and limited edge resources, it is unrealistic for FL to aggregate all participated devices in parallel for model update and adopt the fixed iteration frequency in local training process. To address this issue, we propose a distributed resources-efficient FPC policy to improve the content caching efficiency and reduce the resources consumption. Through theoretical analysis, we first formulate the FPC problem into a stacked autoencoders (SAE) model loss minimization problem while satisfying resources constraint. We then propose an adaptive FPC (AFPC) algorithm combined deep reinforcement learning (DRL) consisting of two mechanisms of client selection and local iterations number decision. Next, we show that when training data are Non-IID, aggregating the model parameters of all participated devices may be not an optimal strategy to improve the FL-based content caching efficiency, and it is more meaningful to adopt adaptive local iteration frequency when resources are limited. Finally, experimental results in three real datasets demonstrate that AFPC can effectively improve cache efficiency up to 38.4$\%$and 6.84$\%$, and save resources up to 47.4$\%$and 35.6$\%$, respectively, compared with traditional multi-armed bandit (MAB)-based and FL-based algorithms.