Skip to Main content Skip to Navigation
Journal articles

Multi-tenancy and URLLC on unlicensed spectrum: Performance and design

Abstract : We study in this paper the transport of Ultra-Reliable Low-Latency Communication (URLLC) in a scenario of Industry 4.0 where transmission in uplink is possible in both licensed and unlicensed spectrum. We propose a transmission policy where the packet first attempts transmission in unlicensed spectrum during a time budget smaller than the delay constraint, if the transmission fails then the packet uses the remaining time budget to attempt transmission in licensed spectrum. The goal of using this policy is to minimize the cost of licensed bandwidth required for URLLC. We first consider the case of one tenant managing the industrial area and quantify the needed licensed bandwidth to attain reliability target within the target delay. We optimize the system by choosing the transmission policy which minimizes this cost. We then study the case of multiple tenants present in the same area and sharing unlicensed spectrum. Each tenant tries to minimize its cost of licensed bandwidth by utilizing unlicensed resources, which may result in the tragedy of the commons like situation. We formulate the problem using a game-theoretic approach to model the non-cooperative multi-tenant scenario. We model the medium access of unlicensed system for this case and derive the strategies that minimize individual cost functions. We then prove the existence of pure Nash equilibria analytically and identify them numerically. Finally, we quantify the so-called price of anarchy, i.e., ratio of the utility yielded by the competitive setting to the outcome of a cooperative scenario.
Document type :
Journal articles
Complete list of metadata
Contributor : Tijani Chahed <>
Submitted on : Tuesday, January 12, 2021 - 2:07:18 PM
Last modification on : Thursday, May 27, 2021 - 10:34:29 AM

Links full text



Ayat Zaki-Hindi, Salah-Eddine Elayoubi, Tijani Chahed. Multi-tenancy and URLLC on unlicensed spectrum: Performance and design. Computer Networks, Elsevier, 2020, 177, pp.107311:1-107311:10. ⟨10.1016/j.comnet.2020.107311⟩. ⟨hal-03107199⟩



Record views