Deqing Zhu

Deqing Zhu

Auto-generated profile, based on publicly available information

Deqing Zhu

Hangzhou Normal University, Hangzhou, Zhejiang, 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
16 January 2025 in Computer Networks
Verified
Journal Article
Closed access
DZDeqing Zhu
Deqing Zhu
LWLidong Wang
Lidong Wang
GPGenmei Pan
Genmei Pan
SLShenji Luan
Shenji Luan

28 January 2026 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
DZDeqing Zhu
Deqing Zhu
GPGenmei Pan
Genmei Pan
HSHaiyan Shi
Haiyan Shi
SQShuwei Qiu
Shuwei Qiu
SLShenji Luan
Shenji Luan
Restricted target wake time (R-TWT) is designed to prioritize latency-sensitive traffic within a basic service set (BSS). However, when transmissions from overlapping BSSs (OBSSs) overlap with the R-TWT service period (SP), the resulting interference may still disrupt latency sensitive traffic. The IEEE 802.11bn amendment addresses this issue by introducing coordinated R-TWT (CR-TWT), which mandates neighboring BSSs using the same primary channel to remain silent, thereby extending SP protection. While nonoverlapping R-TWT schedules guarantee protection, they may not satisfy the stringent timing requirements of delay-sensitive traffic. Conversely, allowing concurrent transmissions among multiaccess points (Multi-APs) in the same SP through spatial reuse can be beneficial. This poses challenges since spatial reuse typically requires a centralized coordinator to collect received signal strength indicator (RSSI) information and compute the optimal transmit power of involved access points (APs), incurring significant signaling overhead. Meanwhile, the traffic identifier (TID) is critical for negotiating R-TWT SPs, as Multi-APs can use this information to prioritize high-priority latency-sensitive traffic. These considerations motivate our proposal of distributed CR-TWT (DCR-TWT), which incorporates TID prioritization and formulates weighted data rate maximization as a convex optimization problem. In DCR-TWT, each AP independently solves its local optimization problem to determine whether to transmit within the SP and its corresponding optimal transmit power, greatly reducing signaling overhead. We evaluate the performance of CR-TWT and DCR-TWT using a full physical layer model coupled with a simplified medium access control (MAC) layer model. An adaptive modulation and coding scheme (MCS) selection mechanism is devised to enhance simulation realism. Results show that DCR-TWT outperforms CR-TWT by achieving more than two times the weighted data rate, a significant reduction in the packet error rate and the packet transmission latency. Furthermore, we identify a limitation in DCR-TWT: excessive scheduling for concurrent transmissions. This overscheduling results in up to 30% of cases failing to meet the minimum signal-to-interference-plus-noise ratio (SINR) threshold. To overcome this, we propose enhanced DCR-TWT (EDCR-TWT), which excludes BSSs causing difficulties in the convergence of the convex optimization problem. Experimental results demonstrate a significant improvement: the proportion of cases below the minimum SINR threshold decreases from 30% to below 10% under EDCR-TWT. A comparative analysis reveals that EDCR-TWT outperforms prior work in both weighted data rate and packet error rate.

1 January 2022 in SSRN Electronic Journal
Verified
Preprint
Open access
德朱德庆 朱
德庆 朱
GPGenmei Pan
Genmei Pan
The uplink OFDMA-based random access (UORA) is a key mechanism in the 802.11ax protocol. The low efficiency of UORA could be the bottleneck of a network. When evaluating and improving UORA, most of the existing work assumes the random-access RUs (RA-RUs) allocated in different trigger frame is a fixed number. However, the number of RA-RUs allocated could vary according to network conditions and the node’s activity degrees. In this work, we formulate an analytical model for the UORA procedure in which the number of RA-RUs and the contention window picked are taken as random variables. We verify the proposed model via simulation. Based on the model, we define the delay for successfully delivering a frame in terms of trigger frame intervals and the packet delivery ratio (PDR) to evaluate the network performance. We then investigate the impact of the number of allocated RA-RUs and the allowed maximum transmission times on network performance. With the analysis results, we suggest how to use UORA in practice. The comparison results show that our scheme outperforms conventional 802.11ax and previous work in terms of delay and PDR.

31 March 2019 in KSII Transactions on Internet and Information Systems
Verified
Journal Article
Closed access
SLShenji Luan
Shenji Luan
JBJianrong Bao
Jianrong Bao
Chao LiuChao Liu
Chao Liu
JLJie Li
Jie Li
DZDeqing Zhu
Deqing Zhu

31 December 2021 in KSII Transactions on Internet and Information Systems
Verified
Journal Article
Closed access
DZDeqing Zhu
Deqing Zhu
SLShenji Luan
Shenji Luan
KSII Transactions on Internet and Information Systems TIIS Paper Details