Distributed multi-stage image compression—where visual content traverses multiple processing nodes under varying quality requirements—poses challenges. Progressive methods enable bitstream truncation but underutilize available compute resources; successive compression repeats costly pixel-domain operations and suffers cumulative quality loss and inefficiency; fixed-parameter models lack post-encoding flexibility. In this work, we developed the Hierarchical Cascade Framework (HCF) that achieves high rate-distortion performance and better computational efficiency through direct latent-space transformations across network nodes in distributed multi-stage image compression systems. Under HCF, we introduced policy-driven quantization control to optimize rate–distortion trade-offs, and established the edge quantization principle through differential entropy analysis. The configuration based on this principle demonstrates up to 0.6dB PSNR gains over other configurations. When comprehensively evaluated on the Kodak, CLIC, and CLIC2020-mobile datasets, HCF outperforms successive-compression methods by up to 5.56% BD-Rate in PSNR on CLIC, while saving up to 97.8% FLOPs, 96.5% GPU memory, and 90.0% execution time. It also outperforms state-of-the-art progressive compression methods by up to 12.64% BD-Rate on Kodak and enables retraining-free cross-quality adaptation with 7.13-10.87% BD-Rate reductions on CLIC2020-mobile.

This work considers a parallel task execution strategy in vehicular edge computing (VEC) networks, where edge servers are deployed along the roadside to process offloaded computational tasks of vehicular users. To minimize the overall waiting delay among vehicular users, a novel task offloading solution is implemented based on the network cooperation balancing resource under-utilization and load congestion. Dual evaluation through theoretical and numerical ways shows that the developed solution achieves a globally optimal delay reduction performance compared to existing methods, which is also validated by the feasibility test over a real-map virtual environment. The in-depth analysis reveals that predicting the instantaneous processing power of edge servers facilitates the identification of overloaded servers, which is critical for determining network delay. By considering discrete variables of the queue, the proposed technique's precise estimation can effectively address these combinatorial challenges to achieve optimal performance.

This paper develops a cooperative deployment strategy for multi-access point (AP) coordination in IEEE 802.11 networks. This strategy focuses on the radio resource utilization and the link performance in overlapping basic service set (OBSS) environments. A key challenge arises in prioritizing the assignment of resource units (RUs) to OBSS regions, as the RU contention within the coverage of an AP can propagate interference to neighboring APs, causing network-wide inefficiencies. Existing strategies often rely on dedicated resource management agents, which are incompatible with the ad-hoc deployment nature of IEEE 802.11 networks. To address this, a cooperative framework is proposed that enables APs and users to collaboratively optimize the RU allocation through local coordination. The underlying principle focuses on adaptively scheduling the RU allocation in OBSS regions through distributed coordination among APs. Numerical results demonstrate performance improvements in both indoor and outdoor environments, as supported by theoretical guarantees on the convergence and global optimality. In particular, the proposed strategy ensures a high proportion of users meeting the SINR threshold when the number of users remains below twice the number of available RUs, highlighting its effectiveness in dense network deployments.

This work develops a distributed approach to user allocation in hybrid Internet of Things (IoT) networks, where nonorthogonal multiple access (NOMA) and orthogonal multiple access (OMA) techniques coexist for wireless user access. This hybrid approach facilitates massive connections of IoT devices with limited resources, exploiting the best of both multiplexing strategies. The directional beamforming environment is considered for mitigating user interference and providing spatial gains. Under this configuration, it aims at determining transmission techniques for individual users with the dual consideration of resource availability and geometric adjacency. To this challenge, an efficient distributed algorithm is developed via a novel realization techniques with a message-passing framework. The developed algorithm is rigorously investigated by in-depth theoretical analysis that ensures the optimality and the convergence of the global solution. Furthermore, numerical results also approve to outperform existing schemes consistently over a variety of network configurations that happen in wireless IoT networks.

Mobile edge computing (MEC) offloads computation intensive tasks of individual users to computing clouds to alleviate the computing loads. Virtual machines (VMs), in practice, are often adopted to realize the parallel computing feature of MEC clouds. A careful local interaction among VMs further reduces the overall computing latency. However, their management turns out quite challenging in practical wireless MEC networks. This paper aims at minimizing the latency of the overall MEC task with the min-max criterion. To this end, a novel distributed strategy is developed for the joint management of the task allocation and the offloading balance among VMs. This task offloading protocol is carried out through a message-passing framework that enables a simultaneous consideration of the min-max criterion about multiple MEC tasks. The numerical results demonstrate that the proposed scheduling for distributed MEC operations achieves a 40% improvement in network utility performance over existing optimization techniques.