Shenghui Song

Dr. Shenghui Song


Research Focus & Metrics

Academic Career Length

20Years (2007-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
9 November 2022 in IEEE Transactions on Wireless Communications
Verified
Journal Article
Closed access
YSYifei Shen
Yifei Shen
JZJun Zhang
Jun Zhang
SSShenghui Song
Shenghui Song
KLKhaled B. Letaief
Khaled B. Letaief
Deep learning-based approaches have been developed to solve challenging problems in wireless communications, leading to promising results. Early attempts adopted neural network architectures inherited from applications such as computer vision. They often yield poor performance in large scale networks (i.e., poor scalability) and unseen network settings (i.e., poor generalization). To resolve these issues, graph neural networks (GNNs) have been recently adopted, as they can effectively exploit the domain knowledge, i.e., the graph topology in wireless communications problems. GNN-based methods can achieve near-optimal performance in large-scale networks and generalize well under different system settings, but the theoretical underpinnings and design guidelines remain elusive, which may hinder their practical implementations. This paper endeavors to fill both the theoretical and practical gaps. For theoretical guarantees, we prove that GNNs achieve near-optimal performance in wireless networks with much fewer training samples than traditional neural architectures. Specifically, to solve an optimization problem on an n -node graph (where the nodes may represent users, base stations, or antennas), GNNs’ generalization error and required number of training samples are O(n) and O(n2) times lower than the unstructured multi-layer perceptrons. For design guidelines, we propose a unified framework that is applicable to general design problems in wireless networks, which includes graph modeling, neural architecture design, and theory-guided performance enhancement. Extensive simulations, which cover a variety of important problems and network settings, verify our theory and the effectiveness of the proposed design framework.

20 July 2022 in IEEE Transactions on Wireless Communications
Verified
Journal Article
Closed access
LLLumin Liu
Lumin Liu
JZJun Zhang
Jun Zhang
SSShenghui Song
Shenghui Song
KLKhaled B. Letaief
Khaled B. Letaief
Federated learning (FL) is a powerful distributed machine learning framework where a server aggregates models trained by different clients without accessing their private data. Hierarchical FL, with a client-edge-cloud aggregation hierarchy, can effectively leverage both the cloud server’s access to many clients’ data and the edge servers’ closeness to the clients to achieve a high communication efficiency. Neural network quantization can further reduce the communication overhead during model uploading. To fully exploit the advantages of hierarchical FL, an accurate convergence analysis with respect to the key system parameters is needed. Unfortunately, existing analysis is loose and does not consider model quantization. In this paper, we derive a tighter convergence bound for hierarchical FL with quantization. The convergence result leads to practical guidelines for important design problems such as the client-edge aggregation and edge-client association strategies. Based on the obtained analytical results, we optimize the two aggregation intervals and show that the client-edge aggregation interval should slowly decay while the edge-cloud aggregation interval needs to adapt to the ratio of the client-edge and edge-cloud propagation delay. Simulation results shall verify the design guidelines and demonstrate the effectiveness of the proposed aggregation strategy.

5 June 2023 in IEEE Journal of Selected Topics in Signal Processing
Verified
Journal Article
Open access
WYWentao Yu
Wentao Yu
YSYifei Shen
Yifei Shen
HHHengtao He
Hengtao He
XYXianghao Yu
Xianghao Yu
SSShenghui Song
Shenghui Song
KLKhaled B. Letaief
Khaled B. Letaief
Terahertz ultra-massive MIMO (THz UM-MIMO) is envisioned as one of the key enablers of 6G wireless networks, for which channel estimation is highly challenging. Traditional analytical estimation methods are no longer effective, as the enlarged array aperture and the small wavelength result in a mixture of far-field and near-field paths, constituting a hybrid-field channel. Deep learning (DL)-based methods, despite the competitive performance, generally lack theoretical guarantees and scale poorly with the size of the array. In this paper, we propose a general DL framework for THz UM-MIMO channel estimation, which leverages existing iterative channel estimators and is with provable guarantees. Each iteration is implemented by a fixed point network (FPN), consisting of a closed-form linear estimator and a DL-based non-linear estimator. The proposed method perfectly matches the THz UM-MIMO channel estimation due to several unique advantages. First, the complexity is low and adaptive. It enjoys provable linear convergence with a low per-iteration cost and monotonically increasing accuracy, which enables an adaptive accuracy-complexity tradeoff. Second, it is robust to practical distribution shifts and can directly generalize to a variety of heavily out-of-distribution scenarios with almost no performance loss, which is suitable for the complicated THz channel conditions. For practical usage, the proposed framework is further extended to wideband THz UM-MIMO systems with beam squint effect. Theoretical analysis and extensive simulation results are provided to illustrate the advantages over the state-of-the-art methods in estimation accuracy, convergence rate, complexity, and robustness.

1 December 2023 in IEEE Communications Surveys & Tutorials
Verified
Journal Article
Closed access
HGHuixian Gu
Huixian Gu
LZLiqiang Zhao
Liqiang Zhao
ZHZhu Han
Zhu Han
GZGan Zheng
Gan Zheng
SSShenghui Song
Shenghui Song
The cloud-edge-terminal collaborative network (CETCN) is considered as a novel paradigm for emerging applications owing to its huge potential in providing low-latency and ultra-reliable computing services. However, achieving such benefits is very challenging due to the heterogeneous computing power of terminal devices and the complex environment faced by the CETCN. In particular, the high-dimensional and dynamic environment states cause difficulties for the CETCN to make efficient decisions in terms of task offloading, collaborative caching and mobility management. To this end, artificial intelligence (AI), especially deep reinforcement learning (DRL) has been proven effective in solving sequential decision-making problems in various domains, and offers a promising solution for the above-mentioned issues due to several reasons. Firstly, accurate modelling of the CETCN, which is difficult to obtain for real-world applications, is not required for the DRL-based method. Secondly, DRL can effectively respond to high-dimensional and dynamic tasks through iterative interactions with the environment. Thirdly, due to the complexity of tasks and the differences in resource supply among different vendors, collaboration is required between different vendors to complete tasks. The multi-agent DRL (MADRL) methods are very effective in solving collaborative tasks, where the collaborative tasks can be jointly completed by cloud, edge and terminal devices which provided by different vendors. This survey provides a comprehensive overview regarding the applications of DRL and MADRL in the context of CETCN. The first part of this survey provides a depth overview of the key concepts of the CETCN and the mathematical underpinnings of both DRL and MADRL. Then, we highlight the applications of RL algorithms in solving various challenges within CETCN, such as task offloading, resource allocation, caching and mobility management. In addition, we extend discussion to explore how DRL and MADRL are making inroads into emerging CETCN scenarios like intelligent transportation system (ITS), the industrial Internet of Things (IIoT), smart health and digital agriculture. Furthermore, security considerations related to the application of DRL within CETCN are addressed, along with an overview of existing standards that pertain to edge intelligence. Finally, we list several lessons learned in this evolving field and outline future research opportunities and challenges that are critical for the development of the CETCN. We hope this survey will attract more researchers to investigate scalable and decentralized AI algorithms for the design of CETCN.

6 February 2023 in IEEE Journal on Selected Areas in Communications
Verified
Journal Article
Closed access
ZXZhijie Xie
Zhijie Xie
SSShenghui Song
Shenghui Song
One of the fundamental issues for Federated Learning (FL) is data heterogeneity, which causes accuracy degradation, slow convergence, and the communication bottleneck issue. Although the impact of data heterogeneity on supervised FL has been widely studied, the related investigation for Federated Reinforcement Learning (FRL) is still in its infancy. In this paper, we first define the type and level of data heterogeneity for policy gradient based FRL systems. By inspecting the connection between the global and local objective functions, we prove that local training can benefit the global objective, if the local update is properly penalized by the total variation (TV) distance between the local and global policies. A necessary condition for the global policy to be learn-able from the local environments is also derived, which is directly related to the heterogeneity level. Based on the theoretical result, a Kullback-Leibler (KL) divergence based penalty is proposed to directly constrain the model outputs in the distribution space and the convergence proof of the proposed algorithm is also provided. By jointly penalizing the divergence of the local policy from the global policy with a global penalty and penalizing each iteration of the local training with a local penalty, the proposed method achieves a better trade-off between training speed (step size) and convergence. Experiment results on two popular Reinforcement Learning (RL) experiment platforms demonstrate the advantage of the proposed algorithm over existing methods in accelerating and stabilizing the training process with heterogeneous data.

10 October 2023 in IEEE Transactions on Wireless Communications
Verified
Journal Article
Closed access
HHHengtao He
Hengtao He
XYXianghao Yu
Xianghao Yu
JZJun Zhang
Jun Zhang
SSShenghui Song
Shenghui Song
KLKhaled B. Letaief
Khaled B. Letaief
As one of the core technologies for 5G systems, massive multiple-input multiple-output (MIMO) introduces dramatic capacity improvements along with very high beamforming and spatial multiplexing gains. When developing efficient physical layer algorithms for massive MIMO systems, message passing is one promising candidate owing to its superior performance. However, as their computational complexity increases dramatically with the problem size, the state-of-the-art message passing algorithms cannot be directly applied to future 6G systems, where an exceedingly large number of antennas are expected to be deployed. To address this issue, we propose a model-driven deep learning (DL) framework, namely the AMP-GNN for massive MIMO transceiver design, by considering the low complexity of the AMP algorithm and adaptability of GNNs. Specifically, the structure of the AMP-GNN network is customized by unfolding the approximate message passing (AMP) algorithm and introducing a graph neural network (GNN) module into it. The permutation equivariance property of AMP-GNN is proved, which enables the AMP-GNN to learn more efficiently and to adapt to different numbers of users. We also reveal the underlying reason why GNNs improve the AMP algorithm from the perspective of expectation propagation, which motivates us to amalgamate various GNNs with different message passing algorithms. In the simulation, we take the massive MIMO detection to exemplify that the proposed AMP-GNN significantly improves the performance of the AMP detector, achieves comparable performance as the state-of-the-art DL-based MIMO detectors, and presents strong robustness to various mismatches.