Xinchen Wei

Xinchen Wei

Auto-generated profile, based on publicly available information

Xinchen Wei

Department of Electronic Engineering, University of York, York, U.K.

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
24 May 2024 in IEEE Transactions on Communications
Verified
Journal Article
Closed access
WCWenchao Chen
Wenchao Chen
XWXinchen Wei
Xinchen Wei
KCKaikai Chi
Kaikai Chi
KYKeping Yu
Keping Yu
Amr TolbaAmr Tolba
Amr Tolba
Mohsen GuizaniMohsen Guizani
Mohsen Guizani
Wireless powered mobile edge computing (WP-MEC), which combines mobile edge computing (MEC) and wireless power transfer (WPT), is a promising paradigm for coping with the computing power and energy constraints of wireless devices. However, how to realize the online optimal offloading decision and resource allocation in the WP-MEC system is very challenging. This paper studies the system computation completion time (SCCT) minimization problems for WP-MEC networks using non-orthogonal multiple access (NOMA) communication under binary and partial offloading modes. Due to the complexity of the optimization problems and the time-varying nature of the channel state information, we decouple the original problems into a top-problem of optimizing WPT duration and a sub-problem of optimizing resource allocation, and then propose a convolutional deep reinforcement learning online (CDRO) algorithm. For the top-problem, a deep reinforcement learning framework is used to obtain the near-optimal WPT duration, and an incremental exploration policy is designed to balance the exploration accuracy and exploration range to improve the convergence performance of the CDRO algorithm. For the sub-problems, we propose their corresponding low-complexity algorithms based on in-depth analysis and derivation of the optimal offloading decision’s properties. Finally, numerical results show that the proposed CDRO algorithm achieves near-optimal SCCT with low computational complexity, enabling online decision-making in time-varying channel environments.

6 May 2022 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
XWXinchen Wei
Xinchen Wei
HAHaitham AlObiedollah
Haitham AlObiedollah
KCKanapathippillai Cumanan
Kanapathippillai Cumanan
ZDZhiguo Ding
Zhiguo Ding
ODOctavia A. Dobre
Octavia A. Dobre
In this paper, we consider an energy efficient resource allocation technique for a hybrid time division multiple access (TDMA) - non-orthogonal multiple access (NOMA) system. In such a hybrid system, the available time for transmission is divided into several sub-time slots, and a sub-time slot is allocated to serve a group of users (i.e., cluster). Furthermore, signals for the users in each cluster are transmitted based on the NOMA approach. With NOMA, multiple users can be served simultaneously through utilizing power domain multiplexing at transmitter and successive interference cancellation (SIC) at receiver. In this paper, to maximize the energy efficiency (EE), we jointly allocate both the available time slots and the available transmit power in the hybrid TDMA-NOMA system. In particular, we formulate an EE maximization (EE-Max) problem aiming to maximize the overall EE of the system with a per-user minimum rate and transmit power constraints. However, this joint optimization problem is non-convex in nature, and thus, cannot be solved directly. Therefore, we develop an iterative algorithm by approximating the original problem into a convex one with sequential convex approximation (SCA) and a novel second-order cone (SOC) approach. Simulation results demonstrate that the performance of the proposed hybrid TDMA-NOMA system with joint resource allocation outperforms the system with equal time allocation in terms of the overall EE. Simulation results further confirm that the proposed iterative approaches with SCA and SOC techniques converge within a few number of iterations while yielding the solution to the original non-convex problem.

11 January 2022 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
XWXinchen Wei
Xinchen Wei
HAHaitham AlObiedollah
Haitham AlObiedollah
KCKanapathippillai Cumanan
Kanapathippillai Cumanan
WWWei Wang
Wei Wang
ZDZhiguo Ding
Zhiguo Ding
ODOctavia A. Dobre
Octavia A. Dobre
The combination of time division multiple access (TDMA) and non-orthogonal multiple access (NOMA), referred to as hybrid TDMA-NOMA system, is considered as a potential solution to meet the unprecedented requirements for future wireless networks. While recent resource allocation techniques aiming to individually maximize either spectral efficiency (SE) or energy efficiency (EE), this paper considers an SE-EE trade-off based technique for a hybrid TDMA-NOMA system. This design offers an additional degree of freedom in resource allocation. The proposed design is formulated as a non-convex multi-objective optimization (MOO) problem. The MOO framework is reformulated as a single-objective optimization (SOO) problem by combining the multi-objectives through a weighted-sum objective function. With this, each of the original objectives is assigned with a weight factor to reflect its importance in the design. Then, the sequential convex approximation (SCA) and a second-order cone (SOC) approach are jointly utilized to deal with the non-convexity issues of the SOO problem. Simulation results reveal that the proposed trade-off based design strikes a good balance between the objective functions, while meeting the instantaneous requirements of the system.

3 December 2024 in Computer Networks
Verified
Journal Article
Closed access
GSGuanqun Shen
Guanqun Shen
XWXinchen Wei
Xinchen Wei
KCKaikai Chi
Kaikai Chi
FAFayez Hussain Alqahtani
Fayez Hussain Alqahtani
Amr TolbaAmr Tolba
Amr Tolba

26 August 2023 in IET Communications
Verified
Journal Article
Open access
XCXi Chen
Xi Chen
GJGuodong Jiang
Guodong Jiang
KCKaikai Chi
Kaikai Chi
SZShubin Zhang
Shubin Zhang
XWXinchen Wei
Xinchen Wei
GCGang Chen
Gang Chen
The 0–1 mixed integer programming problem of binary offloading on wireless powered mobile edge computing (WP‐MEC) networks requires joint optimization of binary and continuous variables, which is computationally expensive for traditional techniques and difficult to solve within the channel coherence time under time‐varying condition. Using machine learning models to output variable values is also challenging. Hence, designing efficient and low‐complexity algorithms is crucial for optimal network performance. This paper focuses on the computation latency of the FDMA‐based WP‐MEC network and proposes a task‐offloading algorithm to minimize the total completion delay (TCD). The TCD minimization is modelled as a 0–1 MIP problem and is decomposed into a master problem of optimizing the offloading decision and the sub‐problem of optimizing other parameters under a given offloading decision. The sub‐problem is solved using optimization method, while the master problem is solved using a deep reinforcement learning algorithm. Simulation results show that the proposed algorithm can achieve almost minimal TCD with low complexity.