Xinran Zhang

Xinran Zhang

Auto-generated profile, based on publicly available information

Xinran Zhang

State Key Laboratory of Networking and Switching Technology (SKL-NST), Beijing University of Posts and Telecommunications, Beijing, China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
27 December 2019 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
XZXinran Zhang
Xinran Zhang
MPMugen Peng
Mugen Peng
SYShi Yan
Shi Yan
YSYaohua Sun
Yaohua Sun
Cellular vehicle-to-everything (V2X) communication is crucial to support future diverse vehicular applications. However, for safety-critical applications, unstable vehicle-to-vehicle (V2V) links and high signalling overhead of centralized resource allocation approaches become bottlenecks. In this paper, we investigate a joint optimization problem of transmission mode selection and resource allocation for cellular V2X communications. In particular, the problem is formulated as a Markov decision process, and a deep reinforcement learning (DRL) based decentralized algorithm is proposed to maximize the sum capacity of vehicle-to-infrastructure users while meeting the latency and reliability requirements of V2V pairs. Moreover, considering training limitation of local DRL models, a two-timescale federated DRL algorithm is developed to help obtain robust model. Wherein, the graph theory based vehicle clustering algorithm is executed on a large timescale and in turn the federated learning algorithm is conducted on a small timescale. Simulation results show that the proposed DRL-based algorithm outperforms other decentralized baselines, and validate the superiority of the two-timescale federated DRL algorithm for newly activated V2V pairs.

6 January 2022 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
XZXinran Zhang
Xinran Zhang
MPMugen Peng
Mugen Peng
SYShi Yan
Shi Yan
YSYaohua Sun
Yaohua Sun
To satisfy the low-latency requirements of emerging computation-intensive vehicular services, offloading these services to edge or cloud servers has been recognized as an effective solution. Due to the limited resources of edge servers and the faraway distance of cloud servers, it is challenging to provide an efficient resource allocation strategy to balance the latency, throughput and the resource utilization. In this paper, an end–edge–cloud collaboration paradigm is presented for computation offloading in fog-based vehicular networks (FVNETs) by incorporating vehicles with idle resources as fog user equipments (F-UEs). To adaptively orchestrate end–edge–cloud resources in different load cases, a two-timescale resource reservation and allocation framework is proposed. Wherein, a Stackelberg-game-based dynamic F-UE incentive problem is first formulated with the cloud server as the leader and multiple F-UEs as the followers, and then an iterative algorithm is proposed to achieve the Stackelberg equilibrium of the computation resource pricing and reservation. On a small timescale, the joint communication and computation resource allocation problem is transferred into a multiagent stochastic game and a lenient multiagent deep-reinforcement-learning-based distributed algorithm is developed to minimize the sum latency. When latency performance deteriorates, F-UE incentive optimization will be triggered to reserve more resources of F-UEs. Simulation results show that the proposed end–edge–cloud orchestrated computation offloading scheme in FVNETs outperforms baselines in terms of average latency.

29 January 2019 in IEEE Access
Verified
Journal Article
Open access
SYShi Yan
Shi Yan
XZXinran Zhang
Xinran Zhang
HXHongyu Xiang
Hongyu Xiang
WWWenbin Wu
Wenbin Wu
The explosive growth of data with the requirements of high reliability and low latency has posed huge challenges to the vehicular networks. One potential solution is to deploy the fog computing servers geographically closer to the vehicles to serve the vehicle-based applications in real time. However, due to the constraint of caching storage space as well as lack of tractable access mode selection and spectrum allocation algorithm, it is very challenging to balance the network transmission performance and fronthaul savings. In this paper, we investigate the joint optimization problem of access mode selection and spectrum allocation in fog computing based vehicular networks. To solve the problem in an efficient way, the original high complexity optimization problem is divided into two subproblems. Wherein, the vehicle access mode selection problem is solved by Q-learning based algorithm by considering spectrum allocation profiles and the fronthaul link cost. The randomly vehicular network topologies are modeled as CoX processes and the closed-form payoff expressions are derived by the stochastic geometry tools. On the other hand, the optimal spectrum allocation indicator value can be finally obtained via convex optimization. The analytical results for the proposed algorithm as well as the traditional baseline approaches are evaluated with different weight factors, which verify our theoretical analysis and confirm the proposed approach can achieve significant performance gains.

23 August 2022 in IEEE Communications Letters
Verified
Journal Article
Closed access
ZSZezhong Sun
Zezhong Sun
SYShi Yan
Shi Yan
XZXinran Zhang
Xinran Zhang
MPMugen Peng
Mugen Peng
Channel state information (CSI) based sensing approaches have unique advantages for fine-grained motion sensing. However, due to the mismatch between the sensing position and detection probability, mapping motion-sensing precision to optimal detection zones is an involved problem. In this letter, we propose a general CSI-based motion detection model, in which the Fresnel Zones and radar detection theory are used to analyze the key factors affecting detection probability. Besides, analytic expressions for the detection probability at arbitrary positions with available and unavailable phases are derived. Finally, numerical results corroborate the theoretical analysis and show that appropriate transceiver deployment and sensing frequency selection can effectively improve sensing performance.