arXiv subjects

Shenghui Song

Publications and source records attributed to Shenghui Song.

At least 73 records · Page 4Linked to original sources

arXiv2024

Tackling Distribution Shifts in Task-Oriented Communication with Information Bottleneck

Task-oriented communication aims to extract and transmit task-relevant information to significantly reduce the communication overhead and transmission latency. However, the unpredictable distribution shifts between training and test data, including domain shift and semantic shift, can dramatically undermine the system performance. In order to tackle these challenges, it is crucial to ensure that the encoded features can generalize to domain-shifted data and detect semanticshifted data, while remaining compact for transmission. In this paper, we propose a novel approach based on the information bottleneck (IB) principle and invariant risk minimization (IRM) framework. The proposed method aims to extract compact and informative features that possess high capability for effective domain-shift generalization and accurate semantic-shift detection without any knowledge of the test data during training. Specifically, we propose an invariant feature encoding approach based on the IB principle and IRM framework for domainshift generalization, which aims to find the causal relationship between the input data and task result by minimizing the complexity and domain dependence of the encoded feature. Furthermore, we enhance the task-oriented communication with the label-dependent feature encoding approach for semanticshift detection which achieves joint gains in IB optimization and detection performance. To avoid the intractable computation of the IB-based objective, we leverage variational approximation to derive a tractable upper bound for optimization. Extensive simulation results on image classification tasks demonstrate that the proposed scheme outperforms state-of-the-art approaches and achieves a better rate-distortion tradeoff.

eess.SP↗

arXiv2024

Approximate Message Passing-Enhanced Graph Neural Network for OTFS Data Detection

Orthogonal time frequency space (OTFS) modulation has emerged as a promising solution to support high-mobility wireless communications, for which, cost-effective data detectors are critical. Although graph neural network (GNN)-based data detectors can achieve decent detection accuracy at reasonable computational cost, they fail to best harness prior information of transmitted data. To further minimize the data detection error of OTFS systems, this letter develops an AMP-GNN-based detector, leveraging the approximate message passing (AMP) algorithm to iteratively improve the symbol estimates of a GNN. Given the inter-Doppler interference (IDI) symbols incur substantial computational overhead to the constructed GNN, learning-based IDI approximation is implemented to sustain low detection complexity. Simulation results demonstrate a remarkable bit error rate (BER) performance achieved by the proposed AMP-GNN-based detector compared to existing baselines. Meanwhile, the proposed IDI approximation scheme avoids a large amount of computations with negligible BER degradation.

eess.SP↗

arXiv2024

Mutual Information Density of Massive MIMO Systems over Rayleigh-Product Channels

The Rayleigh-product channel model is utilized to characterize the rank deficiency caused by keyhole effects. However, the finite blocklength analysis for Rayleigh product channels is not available in the literature. In this paper, we will characterize the mutual information density (MID) and perform the FBL analysis to reveal the impact of rank-deficiency in Rayleigh-product channels. To this end, we first set up a central limit theorem for the MID over Rayleigh-product MIMO channels in the asymptotic regime where the number of scatterers, number of antennas, and blocklength go to infinity at the same pace. Then, we utilize the CLT to obtain the upper and lower bounds for the packet error probability, whose approximations in the high and low signal to noise ratio regimes are then derived to illustrate the impact of rank deficiency. One interesting observation is that rank-deficiency degrades the performance of MIMO systems with FBL and the fundamental limits of Rayleigh-product channels degenerate to those of the Rayleigh case when the number of scatterers approaches infinity.

cs.IT↗

arXiv2024

Fundamental Limits of Optical Fiber MIMO Channels With Finite Blocklength

The multiple-input and multiple-output (MIMO) technique is regarded as a promising approach to boost the throughput and reliability of optical fiber communications. However, the fundamental limits of optical fiber MIMO systems with finite block-length (FBL) are not available in the literature. This paper studies the fundamental limits of optical fiber multicore/multimode systems in the FBL regime when the coding rate is a perturbation within $\mathcal{O}(\frac{1}{\sqrt{ML}})$ of the capacity, where M and L represent the number of transmit channels and blocklength, respectively. Considering the Jacobi MIMO channel, which was proposed to model the nearly lossless propagation and the crosstalks in optical fiber systems, we derive the upper and lower bounds for the optimal error probability. For that purpose, we first set up the central limit theorem for the information density in the asymptotic regime where the number of transmit, receive, available channels and the blocklength go to infinity at the same pace. The result is then utilized to derive the upper and lower bounds for the optimal average error probability with the concerned rate. The derived theoretical results reveal interesting physical insights for Jacobi MIMO channels with FBL. First, the derived bounds for Jacobi channels degenerate to those for Rayleigh channels when the number of available channels approaches infinity. Second, the high signal-to-noise (SNR) approximation indicates that a larger number of available channels results in a larger error probability. Numerical results validate the accuracy of the theoretical results and show that the derived bounds are closer to the performance of practical LDPC codes than outage probability.

cs.IT↗

arXiv2024

Model-Driven Deep Learning for Distributed Detection with Binary Quantization

Within the realm of rapidly advancing wireless sensor networks (WSNs), distributed detection assumes a significant role in various practical applications. However, critical challenge lies in maintaining robust detection performance while operating within the constraints of limited bandwidth and energy resources. This paper introduces a novel approach that combines model-driven deep learning (DL) with binary quantization to strike a balance between communication overhead and detection performance in WSNs. We begin by establishing the lower bound of detection error probability for distributed detection using the maximum a posteriori (MAP) criterion. Furthermore, we prove the global optimality of employing identical local quantizers across sensors, thereby maximizing the corresponding Chernoff information. Subsequently, the paper derives the minimum MAP detection error probability (MAPDEP) by inplementing identical binary probabilistic quantizers across the sensors. Moreover, the paper establishes the equivalence between utilizing all quantized data and their average as input to the detector at the fusion center (FC). In particular, we derive the Kullback-Leibler (KL) divergence, which measures the difference between the true posterior probability and output of the proposed detector. Leveraging the MAPDEP and KL divergence as loss functions, the paper proposes model-driven DL method to separately train the probability controller module in the quantizer and the detector module at the FC. Numerical results validate the convergence and effectiveness of the proposed method, which achieves near-optimal performance with reduced complexity for Gaussian hypothesis testing.

cs.IT↗

arXiv2024

Model-Driven Sensing-Node Selection and Power Allocation for Tracking Maneuvering Targets in Perceptive Mobile Networks

Maneuvering target tracking will be an important service of future wireless networks to assist innovative applications such as intelligent transportation. However, tracking maneuvering targets by cellular networks faces many challenges. For example, the dense network and high-speed targets make the selection of the sensing nodes (SNs) and the associated power allocation very challenging. Existing methods demonstrated engaging performance, but with high computational complexity. In this paper, we propose a model-driven deep learning (DL)-based approach for SN selection. To this end, we first propose an iterative SN selection method by jointly exploiting the majorization-minimization (MM) framework and the alternating direction method of multipliers (ADMM). Then, we unfold the iterative algorithm as a deep neural network and prove its convergence. The proposed method achieves lower computational complexity, because the number of layers is less than the number of iterations required by the original algorithm, and each layer only involves simple matrix-vector additions/multiplications. Finally, we propose an efficient power allocation method based on fixed point (FP) water filling and solve the joint SN selection and power allocation problem under the alternative optimization framework. Simulation results show that the proposed method achieves better performance than the conventional optimization-based methods with much lower computational complexity.

cs.IT↗

arXiv2024

URLLC in IRS-Aided MIMO Systems: Finite Blocklength Analysis and Design

This paper investigates the ultra reliable and low latency communication (URLLC) performance of the IRS-aided MIMO system. The upper and lower bounds of the optimal average error probability (OAEP) for the coding rate 1/sqrt(Mn) of the capacity are derived, where n and M represent the blocklength and the number of transmit antennas, respectively. To achieve this goal, a new central limit theorem (CLT) for the mutual information density over the IRS-aided MIMO system is derived in the asymptotic regime where the block-length, the IRS size, and number of the antennas go to infinity with the same pace. The CLT is then utilized to derive the closed form upper and lower bounds for the OAEP. Based on the analysis result, a gradient-based algorithm is proposed to minimize the lower bound of the OAEP by optimizing the phase shift of the IRS. Simulation results validate the fitness of the CLT and the effectiveness of the proposed algorithm in optimizing the theoretical bound, as well as the performance of practical LDPC code.

cs.IT↗

arXiv2024

Fundamental Limits of Non-Centered Non-Separable Channels and Their Application in Holographic MIMO Communications

The classical Rician Weichselberger channel and the emerging holographic multiple-input multiple-output (MIMO) channel share a common characteristic of non-separable correlation, which captures the interdependence between transmit and receiver antennas. However, this correlation structure makes it very challenging to characterize the fundamental limits of non-centered (Rician), non-separable MIMO channels. In fact, there is a dearth of existing literature that addresses this specific aspect, underscoring the need for further research in this area. In this paper, we investigate the mutual information (MI) of non-centered non-separable MIMO channels, where both the line-of-sight and non-line-of-sight components are considered. By utilizing random matrix theory (RMT), we set up a central limit theorem for the MI and give the closed-form expressions for its mean and variance. The derived results are then utilized to approximate the ergodic MI and outage probability of holographic MIMO channels. Numerical simulations validate the accuracy of the theoretical results.

cs.IT↗

arXiv2024

Interference Mitigation for Network-Level ISAC: An Optimization Perspective

Future wireless networks are envisioned to simultaneously provide high data-rate communication and ubiquitous environment-aware services for numerous users. One promising approach to meet this demand is to employ network-level integrated sensing and communications (ISAC) by jointly designing the signal processing and resource allocation over the entire network. However, to unleash the full potential of network-level ISAC, some critical challenges must be tackled. Among them, interference management is one of the most significant ones. In this article, we build up a bridge between interference mitigation techniques and the corresponding optimization methods, which facilitates efficient interference mitigation in network-level ISAC systems. In particular, we first identify several types of interference in network-level ISAC systems, including self-interference, mutual interference, crosstalk, clutter, and multiuser interference. Then, we present several promising techniques that can be utilized to suppress specific types of interference. For each type of interference, we discuss the corresponding problem formulation and identify the associated optimization methods. Moreover, to illustrate the effectiveness of the proposed interference mitigation techniques, two concrete network-level ISAC systems, namely coordinated cellular network-based and distributed antenna-based ISAC systems, are investigated from interference management perspective. Experiment results indicate that it is beneficial to collaboratively employ different interference mitigation techniques and leverage the network structure to achieve the full potential of network-level ISAC. Finally, we highlight several promising future research directions for the design of ISAC systems.

cs.IT↗

arXiv2024

Fundamental Limits of Two-Hop MIMO Channels: An Asymptotic Approach

Multi-antenna relays and intelligent reflecting surfaces (IRSs) have been utilized to construct favorable channels to improve the performance of wireless systems. A common feature between relay systems and IRS-aided systems is the two-hop multiple-input multiple-output (MIMO) channel. As a result, the mutual information (MI) of two-hop MIMO channels has been widely investigated with very engaging results. However, a rigorous investigation on the fundamental limits of two-hop MIMO channels, i.e., the first and second-order analysis, is not yet available in the literature, due to the difficulties caused by the two-hop (product) channel and the noise introduced by the relay (active IRS). In this paper, we employ large-scale random matrix theory (RMT), specifically Gaussian tools, to derive the closed-form deterministic approximation for the mean and variance of the MI. Additionally, we determine the convergence rate for the mean, variance and the characteristic function of the MI, and prove the asymptotic Gaussianity. Furthermore, we also investigate the analytical properties of the fundamental equations that describe the closed-form approximation and prove the existence and uniqueness of the solution. An iterative algorithm is then proposed to obtain the solution for the fundamental equations. Numerical results validate the accuracy of the theoretical analysis.

cs.IT↗

arXiv2024

Sensing Mutual Information with Random Signals in Gaussian Channels: Bridging Sensing and Communication Metrics

Sensing performance is typically evaluated by classical radar metrics, such as Cramer-Rao bound and signal-to-clutter-plus-noise ratio. The recent development of the integrated sensing and communication (ISAC) framework motivated the efforts to unify the performance metric for sensing and communication, where mutual information (MI) was proposed as a sensing performance metric with deterministic signals. However, the need of communication in ISAC systems necessitates the transmission of random signals for sensing applications, whereas an explicit evaluation for the sensing mutual information (SMI) with random signals is not yet available in the literature. This paper aims to fill the research gap and investigate the unification of sensing and communication performance metrics. For that purpose, we first derive the explicit expression for the SMI with random signals utilizing random matrix theory. On top of that, we further build up the connections between SMI and traditional sensing metrics, such as ergodic minimum mean square error (EMMSE), ergodic linear minimum mean square error (ELMMSE), and ergodic Bayesian Cramér-Rao bound (EBCRB). Such connections open up the opportunity to unify sensing and communication performance metrics, which facilitates the analysis and design for ISAC systems. Finally, SMI is utilized to optimize the precoder for both sensing-only and ISAC applications. Simulation results validate the accuracy of the theoretical results and the effectiveness of the proposed precoding designs.

cs.IT↗

arXiv2024

Learning Bayes-Optimal Channel Estimation for Holographic MIMO in Unknown EM Environments

Holographic MIMO (HMIMO) has recently been recognized as a promising enabler for future 6G systems through the use of an ultra-massive number of antennas in a compact space to exploit the propagation characteristics of the electromagnetic (EM) channel. Nevertheless, the promised gain of HMIMO could not be fully unleashed without an efficient means to estimate the high-dimensional channel. Bayes-optimal estimators typically necessitate either a large volume of supervised training samples or a priori knowledge of the true channel distribution, which could hardly be available in practice due to the enormous system scale and the complicated EM environments. It is thus important to design a Bayes-optimal estimator for the HMIMO channels in arbitrary and unknown EM environments, free of any supervision or priors. This work proposes a self-supervised minimum mean-square-error (MMSE) channel estimation algorithm based on powerful machine learning tools, i.e., score matching and principal component analysis. The training stage requires only the pilot signals, without knowing the spatial correlation, the ground-truth channels, or the received signal-to-noise-ratio. Simulation results will show that, even being totally self-supervised, the proposed algorithm can still approach the performance of the oracle MMSE method with an extremely low complexity, making it a competitive candidate in practice.

eess.SP↗

arXiv2024

Task-Oriented Communication with Out-of-Distribution Detection: An Information Bottleneck Framework

Task-oriented communication is an emerging paradigm for next-generation communication networks, which extracts and transmits task-relevant information, instead of raw data, for downstream applications. Most existing deep learning (DL)-based task-oriented communication systems adopt a closed-world scenario, assuming either the same data distribution for training and testing, or the system could have access to a large out-of-distribution (OoD) dataset for retraining. However, in practical open-world scenarios, task-oriented communication systems need to handle unknown OoD data. Under such circumstances, the powerful approximation ability of learning methods may force the task-oriented communication systems to overfit the training data (i.e., in-distribution data) and provide overconfident judgments when encountering OoD data. Based on the information bottleneck (IB) framework, we propose a class conditional IB (CCIB) approach to address this problem in this paper, supported by information-theoretical insights. The idea is to extract distinguishable features from in-distribution data while keeping their compactness and informativeness. This is achieved by imposing the class conditional latent prior distribution and enforcing the latent of different classes to be far away from each other. Simulation results shall demonstrate that the proposed approach detects OoD data more efficiently than the baselines and state-of-the-art approaches, without compromising the rate-distortion tradeoff.

eess.SP↗

arXiv2024

Deep Learning-Based Adaptive Joint Source-Channel Coding using Hypernetworks

Deep learning-based joint source-channel coding (DJSCC) is expected to be a key technique for {the} next-generation wireless networks. However, the existing DJSCC schemes still face the challenge of channel adaptability as they are typically trained under specific channel conditions. In this paper, we propose a generic framework for channel-adaptive DJSCC by utilizing hypernetworks. To tailor the hypernetwork-based framework for communication systems, we propose a memory-efficient hypernetwork parameterization and then develop a channel-adaptive DJSCC network, named Hyper-AJSCC. Compared with existing adaptive DJSCC based on the attention mechanism, Hyper-AJSCC introduces much fewer parameters and can be seamlessly combined with various existing DJSCC networks without any substantial modifications to their neural network architecture. Extensive experiments demonstrate the better adaptability to channel conditions and higher memory efficiency of Hyper-AJSCC compared with state-of-the-art baselines.

cs.IT↗

arXiv2024

How Robust is Federated Learning to Communication Error? A Comparison Study Between Uplink and Downlink Channels

Because of its privacy-preserving capability, federated learning (FL) has attracted significant attention from both academia and industry. However, when being implemented over wireless networks, it is not clear how much communication error can be tolerated by FL. This paper investigates the robustness of FL to the uplink and downlink communication error. Our theoretical analysis reveals that the robustness depends on two critical parameters, namely the number of clients and the numerical range of model parameters. It is also shown that the uplink communication in FL can tolerate a higher bit error rate (BER) than downlink communication, and this difference is quantified by a proposed formula. The findings and theoretical analyses are further validated by extensive experiments.

cs.LG↗

arXiv2023

FedSDD: Scalable and Diversity-enhanced Distillation for Model Aggregation in Federated Learning

Recently, innovative model aggregation methods based on knowledge distillation (KD) have been proposed for federated learning (FL). These methods not only improved the robustness of model aggregation over heterogeneous learning environment, but also allowed training heterogeneous models on client devices. However, the scalability of existing methods is not satisfactory, because the training cost on the server increases with the number of clients, which limits their application in large scale systems. Furthermore, the ensemble of existing methods is built from a set of client models initialized from the same checkpoint, causing low diversity. In this paper, we propose a scalable and diversity-enhanced federated distillation scheme, FedSDD, which decouples the training complexity from the number of clients to enhance the scalability, and builds the ensemble from a set of aggregated models with enhanced diversity. In particular, the teacher model in FedSDD is an ensemble built by a small group of aggregated (global) models, instead of all client models, such that the computation cost will not scale with the number of clients. Furthermore, to enhance diversity, FedSDD only performs KD to enhance one of the global models, i.e., the \textit{main global model}, which improves the performance of both the ensemble and the main global model. While partitioning client model into more groups allow building an ensemble with more aggregated models, the convergence of individual aggregated models will be slow down. We introduce the temporal ensembling which leverage the issues, and provide significant improvement with the heterogeneous settings. Experiment results show that FedSDD outperforms other FL methods, including FedAvg and FedDF, on the benchmark datasets.

cs.LG↗

arXiv2023

Sensing-Enhanced Secure Communication: Joint Time Allocation and Beamforming Design

The integration of sensing and communication enables wireless communication systems to serve environment-aware applications. In this paper, we propose to leverage sensing to enhance physical layer security (PLS) in multiuser communication systems in the presence of a suspicious target. To this end, we develop a two-phase framework to first estimate the location of the potential eavesdropper by sensing and then utilize the estimated information to enhance PLS for communication. In particular, in the first phase, a dual-functional radar and communication (DFRC) base station (BS) exploits a sensing signal to mitigate the sensing information uncertainty of the potential eavesdropper. Then, in the second phase, to facilitate joint sensing and secure communication, the DFRC BS employs beamforming and artificial noise to enhance secure communication. The design objective is to maximize the system sum rate while alleviating the information leakage by jointly optimizing the time allocation and beamforming policy. Capitalizing on monotonic optimization theory, we develop a two-layer globally optimal algorithm to reveal the performance upper bound of the considered system. Simulation results show that the proposed scheme achieves a significant sum rate gain over two baseline schemes that adopt existing techniques. Moreover, our results unveil that ISAC is a promising paradigm for enhancing secure communication in wireless networks.

cs.IT↗

arXiv2023

Sensing Mutual Information with Random Signals in Gaussian Channels

Sensing performance is typically evaluated by classical metrics, such as Cramer-Rao bound and signal-to-clutter-plus-noise ratio. The recent development of the integrated sensing and communication (ISAC) framework motivated the efforts to unify the metric for sensing and communication, where researchers have proposed to utilize mutual information (MI) to measure the sensing performance with deterministic signals. However, the need to communicate in ISAC systems necessitates the use of random signals for sensing applications and the closed-form evaluation for the sensing mutual information (SMI) with random signals is not yet available in the literature. This paper investigates the achievable performance and precoder design for sensing applications with random signals. For that purpose, we first derive the closed-form expression for the SMI with random signals by utilizing random matrix theory. The result reveals some interesting physical insights regarding the relation between the SMI with deterministic and random signals. The derived SMI is then utilized to optimize the precoder by leveraging a manifold-based optimization approach. The effectiveness of the proposed methods is validated by simulation results.

cs.IT↗