With the growing number of antennas in massive multiple-input multiple-output (MIMO) systems, robust and fast channel estimation becomes increasingly critical yet remains highly challenging. In this work, we propose a lightweight zero-shot self-supervised (ZS-SS) learning framework. It leverages non-local self-similarity in wireless channels to construct a channel-coefficient bank and generate training pairs via randomized and non-contiguous spatial permutations to decorrelate noise. These pairs then train a compact convolutional neural network (CNN) with a specially designed composite loss for robust channel estimation. To further improve adaptability and efficiency, we incorporate a meta-learning approach for fast inference time to dynamic channel environments. Simulations under Gaussian and representative non-Gaussian scenarios show that our method achieves up to 90% gains over traditional estimators and consistent improvements over state-of-the-art baselines, while running nearly 100 times faster. This demonstrates its practicality and suitability for real-time deployment in resource-limited massive MIMO systems.
Zijun Gao, Wenqiang Yi, Fatma Benkhelifa et al.· IEEE Transactions on Wireles...· 0 citations
The escalating complexity of deep neural networks introduces substantial challenges to deploying federated learning (FL) in resource-limited edge environments. To address these limitations, split federated learning (SFL) has emerged as a promising paradigm, alleviating client-side computational and communication burdens via strategic model splitting, and periodically aggregating client-side and server-side models consistent with the principles of FL. Nevertheless, existing SFL frameworks encounter significant performance degradation arising from data heterogeneity and imbalance, client heterogeneity, as well as constrained wireless resources. To overcome these issues, this paper introduces a novel data distribution deviation-aware split federated learning (DA-SFL) framework. DA-SFL dynamically adjusts aggregation weights according to the deviation of clients’ data distributions from a global distribution, effectively mitigating biases induced by data imbalance and heterogeneity. Furthermore, we theoretically establish the convergence bound of DA-SFL under a non-convex loss function setting, demonstrating that minimizing the data deviation in each training round enhances learning efficacy. Motivated by this, we formulate a mixed-integer nonlinear programming to optimize learning performance under long-term energy constraints. Leveraging the Lyapunov optimization framework, we decompose the problem into a series of tractable subproblems in each learning round, and propose efficient algorithms to find the client scheduling, adaptive cut layer selection, bandwidth allocation, and aggregation weighting policies. Extensive experimental evaluations conducted on Fashion-MNIST, CIFAR-10, and CINIC-10 datasets across diverse scenarios of data heterogeneity and imbalance demonstrate that DA-SFL significantly outperforms baselines regarding test accuracy, time and energy efficiency, while exhibiting notable robustness and scalability.
Chunfeng Xie, Zhixiong Chen, Wenqiang Yi et al.· IEEE Transactions on Communi...· 1 citation
