Federated learning (FL) utilizes distributed edge devices for training based on local datasets which preserves data privacy at the cost of frequent communications of model parameters. The channel state between clients and the aggregator affects the successful delivery of model parameters. A client under poor channel state may fail to deliver its local model parameters and thus results in energy waste. Besides, obtaining the channel state takes extra overhead, which may degrade communication efficiency. It motivates us to investigate the client selection problem for FL with unknown channel state. We first derive an upper bound of the convergence for FL, which reflects the effects of the channel state and client selection decisions. We then formulate a client selection problem considering both the convergence and energy consumption. To solve this problem, we further transform it into a restless multi-armed bandit (RMAB) problem. We prove its indexability and propose an index-based client selection algorithm, termed IDXSel , which has low time complexity, is easy to implement, and is proved to be asymptotically optimal. We compare our IDXSel algorithm with the FedAvg, TransP, IS, FedNorm, UCB-based, and & varepsilon; -greedy-based algorithms on the MNIST and CIFAR-10 datasets. Results show that our algorithm achieves comparable or higher accuracy than the baselines, but wastes more than 5 & times; less energy than the worst of the baselines among all the evaluated scenarios.

Edge computing gained popularity for its promises of low latency and high-quality computing services to users. However, it has also introduced the challenge of mutual untrust between user and edge devices for service level agreement (SLA) compliance. This obstacle hampers wide adoption of edge computing, especially in pervasive edge computing (PEC) where edge devices can freely enter or exit the market, which makes verifying and enforcing SLAs significantly more challenging. In this paper, we propose a framework for verifying and enforcing SLAs in PEC, allowing a user to assess SLA compliance of an edge service and ensure correctness of the service results. Our solution, called VeriEdge, employs a verifiable delayed sampling approach to sample a small number of computation steps, and relies on randomly selected verifiers to verify correctness of the computation results. To make sure the verification process is non-manipulable, we employ verifiable random functions to post-select the verifier(s). A dispute protocol is designed to resolve disputes for potential misbehavior. Building upon the base VeriEdge framework, we further extend it with an anonymity-enhanced design to support privacy-sensitive applications, using blind signature-based tokens to enable anonymous access to edge services. Rigorous security analysis demonstrates that VeriEdge achieves a high probability of detecting SLA violation with a minimal overhead. Experimental results indicate that VeriEdge is lightweight, practical, and efficient.

Sharding is a key approach to improving blockchain scalability for parallel transaction processing. Among sharding strategies, state sharding splits the ledger across multiple shards, allowing each validator to process only a subset of transactions; however, maintaining consistency for cross-shard transactions remains challenging. Most existing sharding systems lack scalability due to reliance on Merkle proofs, centralized intermediaries, or redundant ledger storage. In this paper, we propose ShardTree, a cross-shard protocol that enables high throughput and low confirmation latency for sharded blockchains without these limitations. ShardTree leverages a multi-party virtual payment channel (MPC) to efficiently manage cross-shard transaction processing in batches. We first design an algorithm to select validators to form an MPC with the maximum capacity over a payment channel (PC) path. Based on this, ShardTree constructs an MPC over a PC tree. We also design a rollback scheme to ensure the atomicity of cross-shard transactions. We theoretically prove that ShardTree guarantees security in the presence of Byzantine adversaries. Finally, we implement ShardTree and assess its performance using real-world Ethereum transactions. Evaluation results demonstrate that ShardTree efficiently processes cross-shard transactions and outperforms state-of-the-art protocols in terms of transaction throughput and confirmation latency.

With the development of Low Earth Orbit (LEO) satellite networks (LSNs) and the decreasing costs of satellite manufacturing and deployment, an increasing number of satellites are being placed in low Earth orbit to provide global, lowlatency internet services. However, current satellite communications are characterized by high dynamism, unstable connections, routine configurations, and overall higher latency compared to terrestrial networks, making them inherently less stable. This creates new opportunities for attackers, who can establish a fast and stable tunnel through terrestrial networks to redirect messages or traffic within the satellite network, thereby enabling a specific form of attack. In this paper, we present a novel attack in which an adversary deploys wormholes on satellites in an LSN to mislead routing and aggressively attract traffic from target areas, thereby enabling large-scale eavesdropping and severely compromising network security. We formally formulate the objectives and constraints of the attack, taking into account the wormhole’s unique space-ground coupling in the LSN, its predictable dynamics, coverage limitations, and resource constraints. Additionally, we propose an algorithm to derive the optimal attack strategy. We demonstrate that strategically placed wormholes can intercept a significant portion of traffic across target areas, with the extent of impact varying based on the available budget and temporal variation in traffic demand. Finally, we propose countermeasures of wormhole attack, and highlight the potential of enhancing the attack in future work.

Large Language Models (LLMs) are advanced neural networks pre-trained on massive datasets, enabling them to perform a wide range of natural language processing tasks with exceptional generalization and adaptability. Federated learning (FL) enables fine-tuning of LLMs across distributed clients while preserving data privacy, making it ideal for sensitive domains with diverse datasets. However, full fine-tuning (FFT) of LLMs in FL environments presents significant challenges, including high communication overhead, data privacy concerns, and low performance with heterogeneous client data. This paper introduces Private Low-Rank Adaptation (PrivLoRA), a novel privacy-preserving FL algorithm designed to address these challenges. PrivLoRA combines Low-Rank Adaptation (LoRA), an efficient parameter-efficient fine-tuning (PEFT) technique, with a novel layer-wise fine-tuning strategy and Singular Value Decomposition (SVD)-based initialization to mitigate performance degradation in highly heterogeneous data settings. The proposed framework incorporates advanced privacy mechanisms, including homomorphic encryption (HE) and differential privacy (DP), to ensure robust data protection while maintaining strong model performance. Extensive experiments demonstrate that PrivLoRA achieves superior accuracy and communication efficiency compared to state-of-the-art baselines, particularly in heterogeneous data environments. This research offers a significant advancement for privacy-preserving, communication-efficient LLM fine-tuning in collaborative and sensitive applications.

In the Internet of Things (IoT) field, where interconnected devices generate sensitive data, ensuring privacy is a major challenge. Federated Learning (FL) addresses this by allowing devices to collaboratively train a model without sharing their local data, improving privacy in IoT systems. While the traditional two-layer FL framework is commonly used, adopting a hierarchical client-edge-cloud architecture can significantly accelerate model training, especially in resource-constrained IoT networks. Hierarchical Federated Learning (HFL) offers significant advantages, yet concerns persist regarding potential privacy breaches from analyzing client or edge server data. To address these concerns, we propose PrivHFL, a privacy-preserving solution for HFL that leverages threshold homomorphic encryption. Security and performance analyses demonstrate that the proposed scheme is scalable, supporting larger FL scenarios, including diverse IoT environments, while ensuring data privacy. PrivHFL is resilient to collusion among nearly half of the clients and effectively handles client dropouts. Our approach achieves high accuracy in IID and non-IID scenarios, as demonstrated using the MNIST, CIFAR-10, and CIFAR-100 datasets. Additionally, we show that the added encryption overhead is reasonable, making our solution feasible for real-world IoT applications.

The inclusion of pervasive computing devices in a democratized edge computing ecosystem can significantly expand the capability and coverage of near-end computing for large-scale applications. However, offloading user tasks to heterogeneous and decentralized edge devices comes with the dual risk of both endangered user data security and privacy due to the curious base station or malicious edge servers, and unfair offloading and malicious attacks targeting edge servers from other edge servers and/or users. Existing solutions to edge access control and offloading either rely on "always-on" cloud servers with reduced edge benefits or fail to protect sensitive user service information. To address these challenges, this paper presents SA2FE, a novel framework for edge access control, offloading and accounting. We design a rerandomizable puzzle primitive and a corresponding scheme to protect sensitive service information from eavesdroppers and ensure fair offloading decisions, while a blind token-based scheme safeguards user privacy, prevents double spending, and ensures usage accountability. The security of SA2FE is proved under the Universal Composability framework, and its performance and scalability are demonstrated with implementation on commodity mobile devices and edge servers.

Low Earth Orbit Satellite Networks (LSNs), as the new generation of backbone networks, can provide low-latency network connectivity anywhere on Earth. However, their dynamic topology and unpredictable global usage patterns hinder reliable communication, limiting their application in supporting real-time applications that require predictable performance. Specifically, the highly dynamic LSN may experience congestion and energy depletion due to uneven user demands and the periodic movement of satellites. In this paper, we design a Congestion and Energy-Aware pricing and resource Reservation algorithm, CEAR, which enables a LSN to reserve network resources for online arriving real-time communication requests, ensuring reliable communication to support performance-critical applications such as disaster monitoring and remote teleconferencing. To maintain the long-term performance of the network, the LSN operator sets resource prices for link bandwidth and satellite energy consumption across the network. The resource prices act as a proxy between the resource reservation decisions for each communication request and the operator’s objective to maximize throughput and network utility and/or to balance network-wide resource depletion. CEAR is guided by online competitive algorithm design and achieves a competitive social welfare. Extensive simulations using real-world LSN topology show that CEAR achieves high social welfare while maintaining low network-wide congestion and energy deficit.

Operating a quantum network incurs high capital and operational expenditures, which are expected to be compensated by the high value of enabled quantum applications. However, existing mechanisms mainly focus on maximizing the entanglement distribution rate and neglect the cost incurred on users. This paper aims to address how to utilize quantum network resources in a cost-efficient manner while sustaining high-quantity entanglement distribution. We first consider how to establish a steady stream of entanglements between remote nodes with the minimum cost. Utilizing a recent flow-based abstraction and a novel graph representation, we design an optimal algorithm for min-cost remote entanglement distribution. Our algorithm provides rigorous tools for supporting high-performance quantum network applications with financial consideration and offers longterm data transmission in remote distances. Extensive simulation results validate the advantageous performance compared to existing solutions and heuristics.

Federated learning (FL) is an emerging distributed machine learning framework that enables models to be trained on multiple decentralized devices or servers without transferring data to a centralized server. However, due to its distributed nature, FL is vulnerable to attacks from malicious clients. Although most Byzantine-robust FL methods are designed against model poisoning attacks, they lose effectiveness as the intensity of attacks increases or when new attack strategies emerge. To address these challenges, we propose a novel robust FL method, called FLCom, which leverages outlier detection to defend against model poisoning attacks. FLCom enhances the robustness of FL and outperforms the state-of-the-art methods in accuracy. Additionally, we propose an improved model poisoning attack, called vector-scaling attack (VSA), which exhibits stronger stealthiness against robust aggregation methods. We evaluate both our defense and attack methods under IID and Non-IID settings across three different datasets. The results demonstrate that FLCom achieves higher accuracy than other methods under various attacks, particularly in the Non-IID case. Furthermore, FLCom effectively defends against our proposed VSA, while VSA successfully breaches existing defense mechanisms.

Payment channel networks (PCNs) have been proposed to tackle the scalability issues in blockchains by enabling off-chain transaction settlement. However, the balance depletion problem caused by unidirectional transactions may jeopardize the payments in PCNs. Existing works address this problem by sending artificial payments to rebalance the payment channels. In this paper, we take advantage of a unique property in PCNs, where the payments of opposite directions between two users can cancel each other out, to mitigate this channel depletion problem. Specifically, we design BAR, a distributed balance-aware payment routing protocol, subject to fee-based conservation, timeliness, and feasibility constraints. Moreover, to ensure payment security, we modify the original Hashed Time-Lock Contract (HTLC) protocol to adapt it to BAR, such that BAR achieves efficiency and atomicity. Extensive simulations demonstrate that BAR outperforms the state-of-the-art algorithms Spider [1] and LND [2] in terms of success ratio and success volume.

Operating a quantum network incurs high capital and operational expenditures, which are expected to be compensated by the high value of enabled quantum applications. However, existing mechanisms mainly focus on maximizing the entanglement distribution rate and neglect the cost incurred on users. This paper aims to address how to utilize quantum network resources in a cost-efficient manner while sustaining high-quantity and high-quality entanglement distribution. We first consider how to establish a steady stream of entanglements between remote nodes with the minimum cost. Utilizing a recent flow-based abstraction and a novel graph representation, we design an optimal algorithm for min-cost remote entanglement distribution. Next, we consider distributing entanglements with the highest fidelity subject to a cost bound and prove its NP-hardness. To explore the cost-fidelity trade-off due to swapping and purification, we propose an approximation scheme for maximizing fidelity while satisfying an arbitrary cost bound. Our algorithms provide rigorous tools for supporting high-performance quantum network applications with financial consideration and offer strong theoretical guarantees. Extensive simulation results validate the advantageous performance in cost efficiency and/or fidelity compared to existing solutions and heuristics.

Personalized federated learning (PFL) addresses the limitations of traditional federated learning (FL) in statistically heterogeneous scenarios, where diverse client data distributions reduce model applicability. However, many PFL methods compromise client privacy or depend on external client information. This paper proposes a novel PFL method that enhances privacy and autonomy by enabling each client to share only partial model components and adaptively optimize local aggregation weights to align with its local objective. This approach minimizes reliance on other clients' information while ensuring robust personalization. Comparative experiments on three real-world datasets against seven baseline methods demonstrate that our method achieves higher or comparable accuracies, validating its effectiveness in non-IID settings.

Federated learning (FL) enables collaborative training of a global model among clients without centralized data collection. However, it is non-trivial to select clients during each communication round due to wireless communication channels between clients and the server, especially the difficulty of obtaining the accurate channel state. To deal with these challenges, we first demonstrate the effectiveness of using link reliability as a metric in client selection to achieve high model accuracy. To tackle the difficulty of obtaining accurate link reliabilities, we then propose an online client selection strategy for FL based on predicted link reliabilities, denoted as FedPRCS, by formulating the problem as a combinatorial multi-armed bandit (CMAB) problem. Our FedPRCS algorithm dynamically assesses link reliabilities and selects clients accordingly, without prior knowledge of link reliabilities. The theoretical analysis on regret bound demonstrates the asymptotic optimality of our algorithm. Simulation results on diverse datasets and different link reliability settings under both IID and Non-IID cases demonstrate the advantages of our algorithm in achieving high model accuracy and low loss.

Device-to-device (D2D) communication supports direct communications between nearby devices, which has potential to improve network capacity, spectrum efficiency and energy efficiency. Considering the high overhead to obtain the complete channel state information (CSI), we investigate resource allocation problems of D2D communication underlaying cellular networks with incomplete CSI to minimize the total power consumption. To deal with the challenge brought by incomplete CSI in estimating the instantaneous rates of D2D pairs (DPs), we consider two QoS metrics in terms of the expected rate and outage probability using statistical CSI. Based on that, we first investigate the energy-efficient power control problem for single cellular user (CU) and single DP sharing the same spectrum. With rigorous theoretical analysis of the intrinsic properties of CU rate and two QoS metrics, we design an optimal energy-efficient power control (EPO) algorithm for single CU and single DP. Using EPO as a building block, we then propose an energy-efficient resource allocation algorithm for multiple CUs and multiple DPs with incomplete CSI. Simulation results show that our algorithms consume the lowest powers compared with two baseline algorithms.

Device-to-device (D2D) communication is a new technology in cellular networks which enables direct communications between nearby devices. It could reuse the uplink spectrum of the cellular users (CUs) thus will introduce interference. Efficient resource allocation methods can coordinate interference and further improve network capacity. The difficulty of obtaining perfect channel state information (CSI) makes the resource allocation more complex. By using the statistical information of the unknown channel fading factors, we propose an optimal power control algorithm for any CU and any D2D pair sharing the same channel. Based on that, we formulate the resource allocation problem for multiple CUs and multiple D2D pairs with incomplete CSI as a maximum weight matching problem. We then solve the problem by combining the optimal power control algorithm and the maximum weight matching algorithm. Simulation results demonstrate the efficiency of our algorithms in achieving high D2D rates while guaranteeing the minimum rates required by CUs.

Edge computing gained popularity for its promises of low latency and high-quality computing services to users. However, it has also introduced the challenge of mutual untrust between user and edge devices for service level agreement (SLA) compliance. This obstacle hampers wide adoption of edge computing, especially in pervasive edge computing (PEC) where edge devices can freely enter or exit the market, which makes verifying and enforcing SLAs significantly more challenging. In this paper, we propose a framework for verifying and enforcing SLAs in PEC, allowing a user to assess SLA compliance of an edge service and ensure correctness of the service results. Our solution, called VeriEdge, employs a verifiable delayed sampling approach to sample a small number of computation steps, and relies on randomly selected verifiers to verify correctness of the computation results. To make sure the verification process is non-manipulable, we employ verifiable random functions to post-select the verifier(s). A dispute protocol is designed to resolve disputes for potential misbehavior. Rigorous security analysis demonstrates that VeriEdge achieves a high probability of detecting SLA violation with a minimal overhead. Experimental results indicate that VeriEdge is lightweight, practical, and efficient.

Low Earth Orbit (LEO) Earth Observation (EO) satellites have changed the way we monitor Earth. Acting like moving cameras, EO satellites are formed in constellations with different missions and priorities, and capture vast data that needs to be transmitted to the ground for processing. However, EO satellites have very limited downlink communication capability, limited by transmission bandwidth, number and location of ground stations, and small transmission windows due to high velocity satellite movement. To optimize resource utilization, EO constellations are expected to share communication spectrum and ground stations for maximum communication efficiency. In this paper, we investigate a new attack surface exposed by resource competition in EO constellations, targeting the delay or drop of Earth monitoring data using legitimate EO services. Specifically, an attacker can inject high-priority requests to temporarily preempt low-priority data transmission windows. Furthermore, we show that by utilizing predictable satellite dynamics, an attacker can intelligently target critical data from low-priority satellites, either delaying its delivery or irreversibly dropping the data. We formulate two attacks, the data delay attack and the data overflow attack, design algorithms to assist attackers in devising attack strategies, and analyze their feasibility or optimality in typical scenarios. We then conduct trace-driven simulations using real-world satellite images and orbit data to evaluate the success probability of launching these attacks under realistic satellite communication settings. We also discuss possible defenses against these attacks.

Scalability is a critical challenge for blockchain-based cryptocurrencies. Payment channel networks (PCNs) have emerged as a promising solution for this challenge. However, channel balance depletion can significantly limit the capacity and usability of a PCN. Specifically, frequent transactions that result in unbalanced payment flows from two ends of a channel can quickly deplete the balance on one end, thus blocking future payments from that direction. In this paper, we propose Fence, an online balance-aware fee setting algorithm to prevent channel depletion and improve PCN sustainability and long-term throughput. In our algorithm, PCN routers set transaction fees based on the current balance and level of congestion on each channel, in order to incentivize payment senders to utilize paths with more balance and less congestion. Our algorithm is guided by online competitive algorithm design, and achieves an asymptotically tight competitive ratio with constant violation in a unidirectional PCN. We further prove that no online algorithm can achieve a finite competitive ratio in a general PCN. Extensive simulations under a real-world PCN topology show that Fence achieves high throughput and keeps network channels balanced, compared to state-of-the-art PCN routing algorithms.

Payment Channel Networks (PCNs) offer an efficient off-chain alternative to the blockchain for transactions. Router nodes in PCNs facilitate transactions between non-adjacent nodes in exchange for a fee. PCN topology tends to be centralized, with a select number of routers known as hubs dominating all payment services. The fee-setting choices of hubs in order to maximize their revenue present fertile grounds for the study of PCN communications and economics. In this paper, we conduct a comprehensive analysis of the Hub Price-Setting (HPS) game. In particular, we define approximate Best Response strategies (epsilon-BR) as well as approximate Nash equilibria (epsilon-NE). We prove that for any epsilon > 0, an epsilon-BR always exists, and can be computed in polynomial time. We also prove that for some. > 0, an epsilon-NE may not exist. We furthermore introduce the notion of conservative estimate and present a max-min approach to the HPS game. Extensive evaluation results demonstrate the power of our proposed approach.