Maximizing the Achievable Sum-Rate of a Busy Spectrum in a Dynamic Licensed Shared Access

Abstract

During the period when the spectrum is referred to as ‘busy’ in a dynamic licensed shared access system, there is
a notable reduction in the maximum data rate that can be achieved by the licensee’s system as a result of the
imposition of the incumbent's interference threshold constraint on the licensee’s system operation. To solve this
problem, a power allocation technique that optimizes the licensee’s sum-rate without adversely affecting the
incumbent’s operation is proposed in this paper. We begin by solving a sum-rate maximization problem subject
to the incumbent’s interference threshold constraint. Using the Lagrangian method, an optimal power allocation
model is derived from the formulated convex optimization problem. When compared to a non-optimized system,
results obtained from simulations show a remarkable improvement in the achievable sum-rate of the licensee
system under the proposed optimal power allocation model. The results further show that at low transmit power,
the proposed optimal power allocation scheme is better for smaller number of users, while the size of the cell
radius does not significantly affect the sum-rate gain of the proposed scheme.