On the capacity of cognitive radio MIMO Gaussian broadcast channels

Abstract

Fig.I. Cognitive radio MIMO Gaussian Broadcast Channels II. SYSTEM MODEL We consider a cognitive radio MIMO Gaussian channel in Fig.I. Let np,r and nc,r denote the number of receiver antennas for the licensed and cognitive users, respectively."Joint source" transmitter has n t multiple antennas. The transmit-receiver signal model from "joint source" transmitters to the licensed and cognitive users, respectively. spectrum when the licensed users are sensed to be absent in that band. However, control and coordination between the cognitive transmitter receiver pair is required to ensure the same spectrum is used.In this paper, we assume that the "joint source" transmitters do know the message of the licensed and cognitive receivers a priori. Such a model is interesting that it allows us to simple complicated problems with many constraints, and understands the ultimate limits on the cognitive receivers by giving them maximum information and allowing them to change theirs transmission and coding strategy based on all the information available at the licensed users. In essence, it enlarges the possible schemes that can be implemented at the cognitive radio.The goal of this paper is to study the capacity of cognitive radio MIMO Gaussian broadcast channels. Along the lines of [8], we consider the model depicted in Fig.I. Based on the model presented, the cognitive radio channel is an interference channel with degraded message sets in which the receiver with a single message is called the licensed user while the receiver with both message sets is called the cognitive user.In this paper, we use boldface letters to denote vectors and matrices. Trt-) , deu-) and E[·] denote the trace, determinant and expectation of a matrix, respectively. For any general matrix H, H+ denotes the conjugate transpose.In is the n x n identity matrix and H~0 denotes a positive semidefinite matrix. X" denotes the row vector (X(I), X(2),.··, X(n)) yn . A -t -t ---. P licensed Receivers P +aP p c Joint SourceAbstract-The cognitive radio MIMO Gaussian broadcast channels model is proposed, and the capacity is also studied under both the cognitive users' transmit power constraint and a set of interference power constraints at the licensed receivers. A duality technique and dirty paper coding (DPC) are used to simplify channel model where multiple antennas are available for both licensed and cognitive users in a spectrum sharing environment. Subgradient iterative power allocation (SIP A) algorithm is also used to solve the parameters of constraint in order to obtain the optimal values. Our analysis and simulation results show that the proposed channel model can be suitable for multiple cognitive users and a few of licensed users with multiple antennas.