Vehicular Technology Chapter Technical Seminar
Age-based Scheduling: Improving Data Freshness for Wireless Real-Time Traffic
Abstract: We consider the problem of scheduling real-time traffic with hard deadlines in a wireless ad hoc network. In contrast to existing real-time scheduling policies that merely ensure a minimal timely throughput, our design goal is to provide guarantees on both the timely throughput and data freshness in terms of age-of-information (AoI), which is a newly proposed metric that captures the “age” of the most recently received information at the destination of a link. The main idea is to introduce the AoI as one of the driving factors in making scheduling decisions. We first prove that the proposed scheduling policy is feasibility-optimal, i.e., satisfying the per-traffic timely throughput requirement. Then, we derive an upper bound on a considered data freshness metric in terms of AoI, demonstrating that the network-wide data freshness is guaranteed and can be tuned under the proposed scheduling policy. Interestingly, we reveal that the improvement of network data freshness is at the cost of slowing down the convergence of the timely throughput. Extensive simulations are performed to validate our analytical results. Both analytical and simulation results confirm the capability of the proposed scheduling policy to improve the data freshness without sacrificing the feasibility optimality.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
- 200 Univ. Ave. West
- Waterloo, Ontario
- Canada N2L 3G1
- Building: EIT
- Room Number: 4152
Speakers
Ning Lu of Thompson Rivers University, Canada
Age-based Scheduling: Improving Data Freshness for Wireless Real-Time Traffic
Abstract: We consider the problem of scheduling real-time traffic with hard deadlines in a wireless ad hoc network. In contrast to existing real-time scheduling policies that merely ensure a minimal timely throughput, our design goal is to provide guarantees on both the timely throughput and data freshness in terms of age-of-information (AoI), which is a newly proposed metric that captures the “age” of the most recently received information at the destination of a link. The main idea is to introduce the AoI as one of the driving factors in making scheduling decisions. We first prove that the proposed scheduling policy is feasibility-optimal, i.e., satisfying the per-traffic timely throughput requirement. Then, we derive an upper bound on a considered data freshness metric in terms of AoI, demonstrating that the network-wide data freshness is guaranteed and can be tuned under the proposed scheduling policy. Interestingly, we reveal that the improvement of network data freshness is at the cost of slowing down the convergence of the timely throughput. Extensive simulations are performed to validate our analytical results. Both analytical and simulation results confirm the capability of the proposed scheduling policy to improve the data freshness without sacrificing the feasibility optimality.
Biography:
Ning Lu is an assistant professor in the Dept. of Computing Science at Thompson Rivers University, Canada. Previously he was a postdoctoral fellow with the Coordinated Science Laboratory in the University of Illinois at Urbana-Champaign. He received his Ph.D. degree in Electrical and Computer Engineering from the University of Waterloo, Canada, in 2015. He received the B.E. and M.E. degrees both in electrical engineering from Tongji University, Shanghai, China, in 2007 and 2010, respectively. He also spent the summer of 2009 as an intern in the National Institute of Informatics, Tokyo, Japan. His current research interests include real-time scheduling, distributed algorithms, and reinforcement learning for wireless communication networks.
Email:
Address:805 TRU Way, , Kamloops, British Columbia, Canada, V2C 0C8