Fang Liu

Fang Liu

Auto-generated profile, based on publicly available information

Fang Liu

College of Electronics and Information Engineering, Shenzhen University, Shenzhen, China

Research Focus & Metrics

Academic Career Length

24Years (2003-Present)

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

Representative Publications

View All Publications
31 October 2023 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
XSXiaoqin Song
Xiaoqin Song
YHYuqing Hua
Yuqing Hua
YYYang Yang
Yang Yang
GXGuoliang Xing
Guoliang Xing
FLFang Liu
Fang Liu
Tiecheng SongTiecheng Song
Tiecheng Song
In this paper, we propose a distributed sidelink resource allocation approach named federated multi-agent deep Q network (FMDQN). The objective is the network spectrum-energy efficiency (SEE) maximization. To address the optimization problem in a distributed manner, we merge federated learning and deep reinforcement learning. We establish an asynchronous federated learning framework by facilitating local model training, thus avoiding the transmission of user-sensitive raw data. Our proposed approach utilizes a client-server architecture and federated aggregation. Additionally, we formulate a sub-band and power allocation strategy that aims to maximize SEE while adhering to strict delay limits for the given task. Simulation results illustrate the favorable convergence of the proposed algorithm in highly dynamic and uncertain Internet of vehicles (IoV) environments.

8 August 2024 in Journal of Medical Internet Research
Verified
Journal Article
Open access
WQWenhao Qi
Wenhao Qi
XZXiaohong Zhu
Xiaohong Zhu
DHDanni He
Danni He
BWBin Wang
Bin Wang
SCShihua Cao
Shihua Cao
JKJunling Kang
Junling Kang
Background: With the rise of artificial intelligence (AI) in the field of dementia biomarker research, exploring its current developmental trends and research focuses has become increasingly important. This study, using literature data mining, analyzes and assesses the key contributions and development scale of AI in dementia biomarker research. Objective: The aim of this study was to comprehensively evaluate the current state, hot topics, and future trends of AI in dementia biomarker research globally. Methods: This study thoroughly analyzed the literature in the application of AI to dementia biomarkers across various dimensions, such as publication volume, authors, institutions, journals, and countries, based on the Web of Science Core Collection. In addition, scales, trends, and potential connections between AI and biomarkers were extracted and deeply analyzed through multiple expert panels. Results: To date, the field includes 1070 publications across 362 journals, involving 74 countries and 1793 major research institutions, with a total of 6455 researchers. Notably, 69.41% (994/1432) of the researchers ceased their studies before 2019. The most prevalent algorithms used are support vector machines, random forests, and neural networks. Current research frequently focuses on biomarkers such as imaging biomarkers, cerebrospinal fluid biomarkers, genetic biomarkers, and blood biomarkers. Recent advances have highlighted significant discoveries in biomarkers related to imaging, genetics, and blood, with growth in studies on digital and ophthalmic biomarkers. Conclusions: The field is currently in a phase of stable development, receiving widespread attention from numerous countries, institutions, and researchers worldwide. Despite this, stable clusters of collaborative research have yet to be established, and there is a pressing need to enhance interdisciplinary collaboration. Algorithm development has shown prominence, especially the application of support vector machines and neural networks in imaging studies. Looking forward, newly discovered biomarkers are expected to undergo further validation, and new types, such as digital biomarkers, will garner increased research interest and attention.

21 July 2023 in IEEE Transactions on Communications
Verified
Journal Article
Closed access
FLFang Liu
Fang Liu
WWWing Shing Wong
Wing Shing Wong
YLYuan-Hsun Lo
Yuan-Hsun Lo
YZYijin Zhang
Yijin Zhang
Chung Shue ChenChung Shue Chen
Chung Shue Chen
GXGuoliang Xing
Guoliang Xing
This paper considers a system in which multiple users send periodically generated status information to a common access point (AP) over a collision channel. To avoid high overhead, there is no time synchronization and no feedback information from the AP to indicate whether a transmission is successful or not. The performance metric that we focus on is the age-of-information (AoI), which represents the freshness of the status information received at the AP. For this model, we propose a sequence based MAC scheme in which each user is pre-assigned a periodic sequence to schedule transmissions. This scheme guarantees each user at least one successful packet transmission within a sequence period, in the absence of time synchronization and feedback information from the AP. To the best of our knowledge, this is the first study investigating AoI performance under a sequence based MAC scheme. We derive the closed-form expressions for average AoI, average peak AoI and average age penalty under the sequence based scheduling. Besides, we derive several critical properties of the sequences to optimize the AoI performance. Comparison results show that our proposed sequence scheme outperforms slotted ALOHA and framed ALOHA in various settings.

23 December 2020 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
LDLei Deng
Lei Deng
FLFang Liu
Fang Liu
YZYijin Zhang
Yijin Zhang
WWWing Shing Wong
Wing Shing Wong
Transparent topology is common in many mobile ad hoc networks (MANETs) such as vehicle ad hoc networks (VANETs), unmanned aerial vehicle (UAV) ad hoc networks, and wireless sensor networks due to their decentralization and mobility nature. There are many existing works on distributed scheduling scheme design for topology-transparent MANETs. Most of them focus on delay-unconstrained settings. However, with the proliferation of real-time applications over wireless communications, it becomes more and more important to support delay-constrained traffic in MANETs. In such applications, each packet has a given hard deadline: if it is not delivered before its deadline, its validity will expire and it will be removed from the system. This feature is fundamentally different from the traditional delay-unconstrained one. In this paper, we for the first time investigate distributed scheduling schemes for a topology-transparent MANET to support delay-constrained traffic. We analyze and compare probabilistic ALOHA scheme and deterministic sequence schemes, including the conventional time division multiple access (TDMA), the Galois Field (GF) sequence scheme proposed in \cite{chlamtac1994making}, and the combination sequence scheme that we propose for a special type of sparse network topology. We use both theoretical analysis and empirical simulations to compare all these schemes and summarize the conditions under which different individual schemes perform best.

7 October 2019 in IEEE Transactions on Communications
Verified
Journal Article
Closed access
FLFang Liu
Fang Liu
Kenneth ShumKenneth Shum
Kenneth Shum
WWWing Shing Wong
Wing Shing Wong
We consider a single-hop wireless sensor network in which each sensor node has an individual elastic data stream to transmit to each other node. We refer to this traffic pattern as unicast in this paper. The network has multiple slotted channels available for the data transmissions. To guarantee successful unicast within a bounded delay, we consider deterministic schemes that pre-assign each node a periodic schedule sequence to schedule transmitting and receiving at each time slot. The sequence period should be minimized since it upper bounds the unicast delay. We have investigated both synchronous TDMA sequences and asynchronous sequences. Since accurate time synchronization is difficult to achieve in sensor networks, we mainly present analysis and design for asynchronous sequences. In this paper, for a group-based channel assignment, we present a lower bound on the common period and propose a sequence construction method by which the period can achieve the same order as the lower bound. We also analyze optimal transmitting and receiving probabilities for two random schemes and compare their frequency utilization efficiency. Finally, unicast delay and energy consumption performance are compared by simulations.

19 April 2021 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
FLFang Liu
Fang Liu
Kenneth ShumKenneth Shum
Kenneth Shum
YZYijin Zhang
Yijin Zhang
WWWing Shing Wong
Wing Shing Wong
Ultra-reliable and low-latency communication (URLLC) is of capital importance for the 5G wireless networks and has received a lot of attention recently. Motivated by the hard delay requirement for URLLC, we aim at devising medium access control (MAC) schemes that can guarantee packet transmissions within a bounded delay. The model we considered is a multi-channel single-hop ad hoc network in which each node has packets to broadcast to other nodes. This broadcast scenario is common, and has been extensively studied in various scenarios such as safety message broadcast in vehicular ad hoc networks (VANETs). Most of the existing MAC schemes employ backoff algorithms to handle transmission failures caused by channel collisions, and thus cannot provide a hard guarantee on transmission delay. Moreover, these schemes rely on time synchronization and coordination among the nodes, which would cause heavy overhead for control message exchange. In this paper, we focus on devising MAC schemes to provide a hard guarantee on broadcast delay, without time synchronization and cooperation among the nodes. We employ the sequence scheme in which each node is pre-assigned a periodic sequence to schedule transmissions and receptions at each time slot. These sequences are referred to as schedule sequences. Since each node starts its transmission schedule independently, there exist relative time offsets among the schedule sequences they use. Previous sequence schemes for broadcast are only conducted in the single channel context. Our objective is to design schedule sequences such that in a multi-channel system, each node can transmit at least one packet to each of its neighbors successfully within a common period, no matter what the time offsets are. The sequence period should be designed as short as possible. In this paper, we analyze the lower bound on sequence period, and propose a sequence construction method by which the period can achieve the same order as the lower bound. We also consider the random scheme in which each node transmits or receives on a channel at each time slot with a pre-determined probability. The frame length and broadcast completion time under different schemes are compared by numerical studies.