Spectral Efficiency of Dynamic Licensed Shared Access

Onidare, Samuel and Navaie, Keivan and Ni, Qiang (2020) Spectral Efficiency of Dynamic Licensed Shared Access. IEEE Transactions on Vehicular Technology, 69 (12). 15149 - 15161. ISSN 0018-9545

Text (corrected LSACAPACITY)
corrected_LSACAPACITY.pdf - Accepted Version
Available under License Creative Commons Attribution-NonCommercial.
Download (666kB)

Abstract

In licensed shared access (LSA) the radio spectrum is dynamically shared between an incumbent and one or more licensee systems. Protective measures are then applied to the licensees' communication activity to protect normal operation of the incumbent system. Such measures are therefore crucial components of the LSA, and thus fundamentally affect the achievable spectrum efficiency. In this paper, we investigate a vertical LSA including an airport traffic control system, as the incumbent, and a mobile network as the licensee. While some previous works only consider the licensee uplink, we analytically obtain the interference received by the incumbent from the licensee's transmission both in the uplink and downlink. We then obtain optimal uplink and downlink power allocation in the licensee using an optimisation problem with the objective of maximizing licensee's spectral efficiency (SE) subject to the incumbent interference threshold. Furthermore, we investigate the effect of the number of users and cell size on the SE. Our results provide quantitative insights for practical system design and deployment of LSA system. We then examine the whole LSA spectrum utilization by characterising the availability of the LSA spectrum using a tandem queue setting. Using this model we obtain an expression for the spectral utilization as a function of the licensee's achievable spectral efficiency and the statistics of the LSA spectrum availability. Simulation results show more than a seven-fold improvement in the licensee SE using the optimal power allocation. It is also seen that a higher SE gain is achieved with the proposed optimal power allocation in cases where the number of user equipment in the eNodeB coverage area is very small. Furthermore, higher spectrum utilization efficiency is achieved as a result of shorter busy periods and higher achievable SE for distant cells.