Lianhong Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Lianhong Zhang has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Chao Li
School of Computer Science, Guangzhou University, Guangzhou, China
3 shared publications - Lisheng Fan
School of Computer Science, Guangzhou University, Guangzhou, China
3 shared publications - Yuxin Wu
School of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou, China
3 shared publications - Arumugam Nallanathan
Queen Mary University of London
2 shared publications Chongzhi Gao
2 shared publicationsJunjuan Xia
2 shared publications
Latest Publications
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 75 (9) , 1-12
Towards Long-Tailed Federated Edge Learning: A Strategic Data Selection Framework
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 75 (9) , 1-12
Towards Long-Tailed Federated Edge Learning: A Strategic Data Selection Framework
- 28 November 2024
- Published by Elsevier in Physical Communication
- Vol. 68
Deep joint source-channel coding empowered two-way relay networks for wireless image transmission
- 28 November 2024
- Published by Elsevier in Physical Communication
- Vol. 68
Deep joint source-channel coding empowered two-way relay networks for wireless image transmission
- 29 May 2024
- Published by Springer Nature in Wireless Personal Communications
- Vol. 136 (1) , 143-164
Priority-Aware Resource Allocation for RIS-assisted Mobile Edge Computing Networks: A Deep Reinforcement Learning Approach
- 29 May 2024
- Published by Springer Nature in Wireless Personal Communications
- Vol. 136 (1) , 143-164
Priority-Aware Resource Allocation for RIS-assisted Mobile Edge Computing Networks: A Deep Reinforcement Learning Approach
- 4 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 70 (1) , 2188-2199
Resilient Machine Learning-Based Semantic-Aware MEC Networks for Sustainable Next-G Consumer Electronics
In this paper, we investigate a semantic-aware mobile edge computing (MEC) network for sustainable next-G consumer electronics, which leverages advanced semantic communication technology to overcome the limitations of available bandwidth and thereby improve communication efficiency. For this network, consumer electronic devices can offload the semantic information extracted from the task instead of transmitting the whole task in the conventional MEC networks, where the latency and energy consumption can be significantly reduced through proper semantic encoding, offloading, and bandwidth and computation allocation decisions. However, the non-convexity of the issue makes it difficult to obtain the optimal decision. To address this issue, a two-level optimization framework is proposed. Specifically, in the upper-level optimization, a resilient deep reinforcement learning (DRL) approach is utilized to enable adaptive offloading and semantic encoding decisions within a dynamic network. In the lower-level optimization, we design three criteria for allocating bandwidth and computation resources by carefully considering the trade-off between computational complexity and implementation efficiency. Finally, extensive simulations are conducted to validate the effectiveness of our proposed strategy. The findings in this paper can help reduce the energy consumption of consumer electronics, hence supporting the development of sustainable next-G consumer electronics.
- 4 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 70 (1) , 2188-2199
Resilient Machine Learning-Based Semantic-Aware MEC Networks for Sustainable Next-G Consumer Electronics
In this paper, we investigate a semantic-aware mobile edge computing (MEC) network for sustainable next-G consumer electronics, which leverages advanced semantic communication technology to overcome the limitations of available bandwidth and thereby improve communication efficiency. For this network, consumer electronic devices can offload the semantic information extracted from the task instead of transmitting the whole task in the conventional MEC networks, where the latency and energy consumption can be significantly reduced through proper semantic encoding, offloading, and bandwidth and computation allocation decisions. However, the non-convexity of the issue makes it difficult to obtain the optimal decision. To address this issue, a two-level optimization framework is proposed. Specifically, in the upper-level optimization, a resilient deep reinforcement learning (DRL) approach is utilized to enable adaptive offloading and semantic encoding decisions within a dynamic network. In the lower-level optimization, we design three criteria for allocating bandwidth and computation resources by carefully considering the trade-off between computational complexity and implementation efficiency. Finally, extensive simulations are conducted to validate the effectiveness of our proposed strategy. The findings in this paper can help reduce the energy consumption of consumer electronics, hence supporting the development of sustainable next-G consumer electronics.
- 2 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (6) , 3341-3348
Scoring Aided Federated Learning on Long-Tailed Data for Wireless IoMT Based Healthcare System
In this paper, we propose a novel federated learning (FL) framework for wireless Internet of Medical Things (IoMT) based healthcare systems, where multiple mobile clients and one edge server (ES) collaboratively train a shared model on long-tail data through wireless channels. However, the presence of long-tailed data in this system may introduce a biased global model which fails to handle the tail classes. Additionally, the occurrence of severe fading in wireless channels may prevent mobile clients from successfully uploading local models to the ES, thereby excluding them from participating in the model aggregation. These situations adversely affect the performance of FL. To overcome these challenges, we propose a novel scoring aided FL framework that uses a scoring-based sampling strategy to select mobile clients with more tailed data and better transmission conditions to upload their local models. Specifically, we leverage the logits to explore the data distribution among local clients and propose a logits based scoring client selection method to alleviate the impact of long-tailed data. Moreover, we address the impact of severe fading by incorporating the channel state information (CSI) and data rate of clients into the logits based scoring and proposing a novel logits and model upload rate based client selection method. Experimental results demonstrate the effectiveness of our proposed framework. In particular, compared to the conventional FedAvg, the proposed framework can achieve accuracy gains ranging from 4.44% to 28.36% on the CIFAR-10-LT dataset with an imbalance factor (IF) of 50.
- 2 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (6) , 3341-3348
Scoring Aided Federated Learning on Long-Tailed Data for Wireless IoMT Based Healthcare System
In this paper, we propose a novel federated learning (FL) framework for wireless Internet of Medical Things (IoMT) based healthcare systems, where multiple mobile clients and one edge server (ES) collaboratively train a shared model on long-tail data through wireless channels. However, the presence of long-tailed data in this system may introduce a biased global model which fails to handle the tail classes. Additionally, the occurrence of severe fading in wireless channels may prevent mobile clients from successfully uploading local models to the ES, thereby excluding them from participating in the model aggregation. These situations adversely affect the performance of FL. To overcome these challenges, we propose a novel scoring aided FL framework that uses a scoring-based sampling strategy to select mobile clients with more tailed data and better transmission conditions to upload their local models. Specifically, we leverage the logits to explore the data distribution among local clients and propose a logits based scoring client selection method to alleviate the impact of long-tailed data. Moreover, we address the impact of severe fading by incorporating the channel state information (CSI) and data rate of clients into the logits based scoring and proposing a novel logits and model upload rate based client selection method. Experimental results demonstrate the effectiveness of our proposed framework. In particular, compared to the conventional FedAvg, the proposed framework can achieve accuracy gains ranging from 4.44% to 28.36% on the CIFAR-10-LT dataset with an imbalance factor (IF) of 50.
- 23 September 2022
- Published by Elsevier in Physical Communication
- Vol. 55
Deep reinforcement learning based IRS-assisted mobile edge computing under physical-layer security
- 23 September 2022
- Published by Elsevier in Physical Communication
- Vol. 55
Deep reinforcement learning based IRS-assisted mobile edge computing under physical-layer security
- Open Access
- 25 January 2022
- Published by Springer Nature in Research Square
DQN Based Mobile Edge Computing for Smart Internet of Vehicle
In this paper, we investigate a multiuser mobile edge computing (MEC)-aided smart internet of vehicle (IoV) network, where one edge server can help accomplish the intensive computational tasks from the vehicular users. For the MEC networks, most of existing works mainly focus on minimizing the system latency to guarantee the user's quality of service (QoS) through designing some offloading strategies, which however fail to consider the pricing from the server and hence fail to take into account the the budget constraint from the users. To address this issue, we jointly incorporate the budget constraint into the system design of the MEC based IoV networks, and then propose a joint deep reinforcement learning (DRL) approach combined with the convex optimization algorithm. Specifically, a deep Q-network (DQN) is firstly used to make the offloading decision, and then the Lagrange multiplier method is employed to allocate the computational capability of the server to multiple users. Simulations are finally presented to demonstrate that the proposed schemes outperforms the conventional ones. In particular, the proposed scheme can effective reduce the system latency up to 56% compared to the conventional schemes.
- 25 January 2022
- Published by Springer Nature in Research Square
Open AccessDQN Based Mobile Edge Computing for Smart Internet of Vehicle
In this paper, we investigate a multiuser mobile edge computing (MEC)-aided smart internet of vehicle (IoV) network, where one edge server can help accomplish the intensive computational tasks from the vehicular users. For the MEC networks, most of existing works mainly focus on minimizing the system latency to guarantee the user's quality of service (QoS) through designing some offloading strategies, which however fail to consider the pricing from the server and hence fail to take into account the the budget constraint from the users. To address this issue, we jointly incorporate the budget constraint into the system design of the MEC based IoV networks, and then propose a joint deep reinforcement learning (DRL) approach combined with the convex optimization algorithm. Specifically, a deep Q-network (DQN) is firstly used to make the offloading decision, and then the Lagrange multiplier method is employed to allocate the computational capability of the server to multiple users. Simulations are finally presented to demonstrate that the proposed schemes outperforms the conventional ones. In particular, the proposed scheme can effective reduce the system latency up to 56% compared to the conventional schemes.