Power Allocation in Cognitive Radio Based on Minimum Interference Generation

This paper considers a wireless multiuser system with N users that can transmit over K orthogonal frequency bands. In this system, interference from other users is treated as noise. In this system we assume each user has a target rate and the goal is to allow as many users as possible to achieve their target rate. We propose a new power allocation algorithm that minimizes both number of frequency bands to be used and necessary power at any user to achieve a target rate. We show that this approach reduces the overall interference power in the system, and thus increases the number of users achieving a desired target rate compared to other power allocation algorithms. This is supported by simulation results, which also shows that in a dense network, sum capacity of the system is increased compared to regular water filling.