Cognitive Radio with Partial Channel State Information at the Transmitter

TL;DR

This paper introduces a new cognitive radio design using LA-GPC with partial CSIT, offering improved rate performance over traditional DPC, and provides practical coding schemes with near-optimal results.

Contribution

It replaces DPC with LA-GPC for better utilization of limited channel knowledge and derives semi-analytical solutions for optimization under fading channels.

Findings

Proposed solutions perform close to brute-force optimal solutions.

Outperform systems based on naive DPC.

Solutions converge to full CSIT optimality as Rician K-factor increases.

Abstract

In this paper, we present the cognitive radio system design with partial channel state information known at the transmitter (CSIT).We replace the dirty paper coding (DPC) used in the cognitive radio with full CSIT by the linear assignment Gel'fand-Pinsker coding (LA-GPC), which can utilize the limited knowledge of the channel more efficiently. Based on the achievable rate derived from the LA-GPC, two optimization problems under the fast and slow fading channels are formulated. We derive semianalytical solutions to find the relaying ratios and precoding coefficients. The critical observation is that the complex rate functions in these problems are closely related to ratios of quadratic form. Simulation results show that the proposed semi-analytical solutions perform close to the optimal solutions found by brute-force search, and outperform the systems based on naive DPC. Asymptotic analysis also shows that these solutions converge to the optimal ones solved with full CSIT when the K-factor of Rician channel approaches infinity. Moreover, a new coding scheme is proposed to implement the LA-GPC in practice. Simulation results show that the proposed practical coding scheme can efficiently reach the theoretical rate performance.