29 March 2018 in IEEE Transactions on Wireless Communications
This paper considers the channel estimation (CE) and multi-user detection (MUD) problems in cloud radio access network (C-RAN). By taking into account of the sparsity of user activities in C-RAN, we solve the CE and MUD problems with compressed sensing to greatly reduce the large pilot overhead. A mixed ℓ 2,1 -regularization penalty functional is proposed to exploit the inherent sparsity existing in both the user activities and remote radio heads with which active users are associated. An iteratively re-weighted strategy is adopted to further enhance the estimation accuracy, and empirical and theoretical guidelines are also provided to assist in choosing tuning parameters. To speed up the optimization procedure, three low-complexity methods under different computing setups are proposed to provide differentiated services. With a centralized setting at the baseband unit pool, we propose a sequential method based on block coordinate descent (BCD). With a modern distributed computing setup, we propose two parallel methods based on alternating direction method of multipliers (ADMM) and hybrid BCD (HBCD), respectively. Specifically, the ADMM is guaranteed to converge but has a high computational complexity, while the HBCD has low complexity but works under empirical guidance. Numerical results are provided to verify the effectiveness of the proposed functional and methods.
16 January 2024 in IEEE Transactions on Vehicular Technology
Semantic communications can significantly improve transmission efficiency, especially in the low signal-to-noise (SNR) regime. However, two-way semantic communications still remain an unexplored topic, which is a critical issue in many machine communication scenarios. Simply extending existing semantic communication systems to the two-way situation requires bidirectional information feedback during the training process, resulting in significant communication overhead. To fill this gap, we investigate a two-way semantic communication (TW-SC) system, where the information feedback can be omitted by exploiting the weight reciprocity in the transceiver. Particularly, the channel simulator and semantic transceiver are implemented on both TWSC nodes and the channel distribution is modeled by a conditional generative adversarial network. Simulation results demonstrate that the proposed TW-SC system performs closing to the stateof-the-art one-way semantic communication systems but requiring no feedback between the transceiver in training process.
19 May 2025 in IEEE Transactions on Vehicular Technology
Beyond the traditional quality of experience (QoE) optimization that focuses on bit transmission, taking into account semantic QoE can better optimize the network to improve user experience. This paper develops a novel distributed hierarchical deep reinforcement learning (DHDRL) framework with a two-layer control strategy in different timescales to optimize QoE in the downlink multi-cell network including the semantic rate and accuracy. Specifically, we decompose the optimization problem into two sub-problems, beamforming and power allocation as well as semantic resource allocation, resulting in an infinite-horizon discrete and a finite-horizon continuous-time Markov decision processes. Two-layer neural networks (NNs) are proposed to solve the sub-problems. At the network level, a distributed high-level NN is proposed to mitigate the inter-cell interference and improve the network performance on a large timescale, while at the link level, another distributed low-level NN to improve the semantic compression rate and meet various semantic accuracy and rate requirements on a small timescale. Numerical results show that DHDRL achieves at least a 19.07% performance gain over the single-layer method, maintains effectiveness in high-dimensional scenarios, and exhibits strong generalization across a variety of multi-objective optimization problems.
23 July 2018 in IEEE Communications Letters
We address the cloud radio access network with wireless fronthaul links for massive machine-type communication as distributed input distributed output (DIDO) system for simplicity. In this paper, the channel estimation and user activity detection problems in the DIDO system are studied. We notice that there are two type of sparsity in DIDO systems: the first is the sparsity of user equipment (UE) activities, and the second is the spatial sparsity of UE signals. A two-stage compressed sensing process is proposed in response, in which UE activities and the overall channel states from active UEs to baseband unit pool are identified at the first stage, and channel states from active UEs to remote radio heads are estimated at the second stage. A low-complexity method is proposed to accelerate the process in the first stage. Simulation results are also presented to show the performance of the proposed approach.
13 June 2025 in IEEE Transactions on Communications
Semantic communications have the potential to improve transmission efficiency and support intelligent tasks. However, the commonly used global sampling-based pattern ignores the fact that the data processing ability of edge transmitters is strictly limited and only a small part of information is available for a single sample in some scenarios. This paper proposes a novel partial sampling-based semantic communication (PSSC) framework where an edge transmitter is guided by feedback from the receiver to locate and collect only part of content relevant to the target task. Taking the vision-based task as an example, the transmitter selectively samples a small patch of a large-size image until the intelligent task is successfully executed at the receiver. The selection of sampling location is modeled as a partially observable Markov decision process problem and an intelligent approach based on reinforcement learning is proposed to solve the problem. In addition, a recurrent neural network-based receiver is designed to fuse information received over multiple transmission rounds. Besides, we prove that the feedback does not increase the semantic channel capacity. Simulation results demonstrate that the proposed framework can locate the informative areas accurately and achieve competitive performance compared to the existing global sampling-based methods.
1 August 2025 in Digital Communications and Networks