Capacity Optimization of Spatial Preemptive Scheduling for Joint URLLC-eMBB Traffic in 5G New Radio

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Ultra-reliable and low-latency communication (URLLC) is envisioned as a primary service class of the fifth generation mobile networks. URLLC applications demand stringent radio latency requirements of 1 millisecond with 99.999\% confidence. Obviously, the coexistence of the URLLC services and enhanced mobile broadband (eMBB) applications on the same spectrum imposes a challenging scheduling problem. In this paper, we propose an enhanced spatial preemptive scheduling framework for URLLC-eMBB traffic coexistence. The proposed scheduler ensures an instant and interference-free signal subspace for critical URLLC transmissions, while achieving best-effort eMBB performance. Furthermore, the impacted eMBB capacity is then recovered by limited network-assisted signaling. The performance of the proposed scheduler is evaluateeed by highly detailed system level simulations of the major performance indicators. Compared to the state-of-the-art multi-traffic schedulers from industry and academia, the proposed scheduler meets the stringent URLLC latency requirements, while significantly improving the achievable ergodic capacity.
Close

Details

Ultra-reliable and low-latency communication (URLLC) is envisioned as a primary service class of the fifth generation mobile networks. URLLC applications demand stringent radio latency requirements of 1 millisecond with 99.999\% confidence. Obviously, the coexistence of the URLLC services and enhanced mobile broadband (eMBB) applications on the same spectrum imposes a challenging scheduling problem. In this paper, we propose an enhanced spatial preemptive scheduling framework for URLLC-eMBB traffic coexistence. The proposed scheduler ensures an instant and interference-free signal subspace for critical URLLC transmissions, while achieving best-effort eMBB performance. Furthermore, the impacted eMBB capacity is then recovered by limited network-assisted signaling. The performance of the proposed scheduler is evaluateeed by highly detailed system level simulations of the major performance indicators. Compared to the state-of-the-art multi-traffic schedulers from industry and academia, the proposed scheduler meets the stringent URLLC latency requirements, while significantly improving the achievable ergodic capacity.
Original languageEnglish
JournalIEEE Global Communications Conference (GLOBECOM) 2018
Publication statusPublished - 2018
Publication categoryResearch
Peer-reviewedYes

Download statistics

No data available
ID: 286766977