Asynchronous Distributed Power Control under Interference Temperature Constraints

Cognitive radio makes it possible for an unlicensed user to access a spectrum opportunistically. This paper addresses the problem of power control in cognitive radio networks, to maximize the system utility in the presence of interference temperature constraint. Penalty function is applied in the problem formulation to take into account the transmit power budget and interference temperature constraints, which is solved efficiently by geometric programming. In the proposed distributed power control scheme, users exchange "price" messages which indicate the negative effect of interference at the receivers. Given this set of prices, each transmitter updates power levels on multiple channels using gradient descent method. It is proved that the proposed algorithm converges to the global optimum when operated in an totally asynchronous manner. The performance of the proposed scheme is investigated by numerical results in the end.