Canshu Lin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Canshu Lin has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Dongbiao He
Southeast University, Nanjing, China
3 shared publications - Laizhong Cui
Shenzhen University, Shenzhen, China
3 shared publications - Zhongxing Ming
Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ), Shenzhen, China
3 shared publications - Cedric Westphal
University of California, Santa Cruz, Santa Cruz, California, USA
2 shared publications - J. J. Garcia-Luna-Aceves
Department of Electrical and Computer Engineering, University of Toronto, Toronto, Ontario, Canada
2 shared publications - Xian Yu
Computer Network Information Center, Chinese Academy of Sciences, Beijing, China
2 shared publications
Latest Publications
- Open Access
- 27 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (3) , 1-24
Enhancing Video Conference Applications with VCApather : A Network as a Service Perspective
The provision of performance-aware video conferencing services today relies on approaches that focus on data compression and client-side bitrate adaptation techniques to optimize transmission. However, these methods fail to quickly respond to fluctuations in network conditions, thereby compromising the quality of service for transmissions. For this reason, this article aims to propose a novel traffic scheduling-based video transmission optimization solution from the perspective of the network service provider. We first investigate the resource requirements of video conferences and present the experiential performance of video conferences under different network conditions and network competition. Based on these results, we design a service-customized routing mechanism called VCApather that minimizes network contention. We then provide implementation solutions for the control plane and the data plane of VCApather . We evaluate VCApather using a fully meshed topology with five nodes and real-world video conference traffic. The results show that VCApather is capable of achieving high link utilization and balance, while also meeting predefined user metrics. Compared to other schemes, VCApather could satisfy 69.8% more QoE requirements and yielded an average bitrate improvement of 1.74 \(\times\) .- 27 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (3) , 1-24
Open AccessEnhancing Video Conference Applications with VCApather : A Network as a Service Perspective
The provision of performance-aware video conferencing services today relies on approaches that focus on data compression and client-side bitrate adaptation techniques to optimize transmission. However, these methods fail to quickly respond to fluctuations in network conditions, thereby compromising the quality of service for transmissions. For this reason, this article aims to propose a novel traffic scheduling-based video transmission optimization solution from the perspective of the network service provider. We first investigate the resource requirements of video conferences and present the experiential performance of video conferences under different network conditions and network competition. Based on these results, we design a service-customized routing mechanism called VCApather that minimizes network contention. We then provide implementation solutions for the control plane and the data plane of VCApather . We evaluate VCApather using a fully meshed topology with five nodes and real-world video conference traffic. The results show that VCApather is capable of achieving high link utilization and balance, while also meeting predefined user metrics. Compared to other schemes, VCApather could satisfy 69.8% more QoE requirements and yielded an average bitrate improvement of 1.74 \(\times\) . - Open Access
- 15 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 57-63
VCApather: A Network as a Service Solution for Video Conference Applications
We propose a network service as a solution for video conference applications by constructing network layer routing strategies. Our approach takes into account the characteristics of conferencing flows, addresses various self-customized metrics, and proactively ensures a positive user experience by preventing contention. The performance of VCApather is evaluated using a fully-meshed topology with five nodes and real-world video conference traffic. The results show that VCApather is capable of achieving high link utilization and balance, while also meeting predefined user metrics. Compared to other schemes, VCApather was found to satisfy 69.8% more QoE requirement and to yield an average bitrate improvement of 1.74×.
- 15 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 57-63
Open AccessVCApather: A Network as a Service Solution for Video Conference Applications
We propose a network service as a solution for video conference applications by constructing network layer routing strategies. Our approach takes into account the characteristics of conferencing flows, addresses various self-customized metrics, and proactively ensures a positive user experience by preventing contention. The performance of VCApather is evaluated using a fully-meshed topology with five nodes and real-world video conference traffic. The results show that VCApather is capable of achieving high link utilization and balance, while also meeting predefined user metrics. Compared to other schemes, VCApather was found to satisfy 69.8% more QoE requirement and to yield an average bitrate improvement of 1.74×.
- Open Access
- 6 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 125-130
Felinet
The edge network has been introduced for providing computing capabilities to accelerate federated learning. However, the heterogeneity of edge networks increases the complexity of traffic scheduling, which can result in network congestion and decreased FL training efficiency. In this paper, we propose Felinet, a path-driven routing solution designed to effectively handle heterogeneous connections and unpredictable FL workloads. We evaluate Felinet through real-world network experiments, demonstrating a 26% improvement in network resource allocation compared to the benchmark routing solution.
- 6 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 125-130
Open AccessFelinet
The edge network has been introduced for providing computing capabilities to accelerate federated learning. However, the heterogeneity of edge networks increases the complexity of traffic scheduling, which can result in network congestion and decreased FL training efficiency. In this paper, we propose Felinet, a path-driven routing solution designed to effectively handle heterogeneous connections and unpredictable FL workloads. We evaluate Felinet through real-world network experiments, demonstrating a 26% improvement in network resource allocation compared to the benchmark routing solution.