TY - JOUR

T1 - On the Spectrum and Energy Efficiency in Dynamic Licensed Shared Access Systems

T2 - A Multiobjective Optimization Approach

AU - Onidare, Samuel

AU - Navaie, Keivan

AU - Ni, Qiang

PY - 2019/11/13

Y1 - 2019/11/13

N2 - In licensed shared access (LSA) systems, the protection of the incumbent in the shared spectrum may degrade the spectrum and energy efficiency of the licensee. In this paper, we examine the optimization of these two performance metrics in a LSA vertical sharing scenario between an airport incumbent, and a mobile network operator licensee. Considering a restriction zone of a pre-defined radius, we derive the probability of the incumbent's interference threshold and then formulate a power allocation scheme as a multi-objective optimization of both energy and spectrum efficiency. We then adopt the weighted sum method to convert this multi-objective optimization into a single objective optimization and convert that into a quasi concave optimization problem. The optimum power allocation is then obtained using fractional programming. We further investigate the impact of various critical operational parameters in conjunction with the two performance metrics. Simulation results indicate a significantly improved energy efficiency in the licensee network as well as the spectrum efficiency comparable to even when the LSA spectrum utilization is unrestricted by the incumbent's maximum interference threshold. Furthermore, we show that with careful selection of the licensee eNodeB coverage radius, transmit power, and users number per eNodeB coverage area, one can engineer the best possible trade-off between the spectrum and energy efficiency.

AB - In licensed shared access (LSA) systems, the protection of the incumbent in the shared spectrum may degrade the spectrum and energy efficiency of the licensee. In this paper, we examine the optimization of these two performance metrics in a LSA vertical sharing scenario between an airport incumbent, and a mobile network operator licensee. Considering a restriction zone of a pre-defined radius, we derive the probability of the incumbent's interference threshold and then formulate a power allocation scheme as a multi-objective optimization of both energy and spectrum efficiency. We then adopt the weighted sum method to convert this multi-objective optimization into a single objective optimization and convert that into a quasi concave optimization problem. The optimum power allocation is then obtained using fractional programming. We further investigate the impact of various critical operational parameters in conjunction with the two performance metrics. Simulation results indicate a significantly improved energy efficiency in the licensee network as well as the spectrum efficiency comparable to even when the LSA spectrum utilization is unrestricted by the incumbent's maximum interference threshold. Furthermore, we show that with careful selection of the licensee eNodeB coverage radius, transmit power, and users number per eNodeB coverage area, one can engineer the best possible trade-off between the spectrum and energy efficiency.

U2 - 10.1109/ACCESS.2019.2952686

DO - 10.1109/ACCESS.2019.2952686

M3 - Journal article

VL - 7

SP - 164517

EP - 164532

JO - IEEE Access

JF - IEEE Access

SN - 2169-3536

M1 - 8895813

ER -