A Low-Delay Low-Complexity EKF Design for Joint Channel and CFO Estimation in Multi-User Cognitive Communications

Abstract

Parameter estimation in cognitive communications can be formulated as a multi-user estimation problem, which is solvable under maximum likelihood solution but involves high computational complexity. This paper presents a time-sharing and interference mitigation based EKF (Extended Kalman Filter) design for joint CFO (carrier frequency offset) and channel estimation at multiple cognitive users. The key objective is to realize low implementation complexity by decomposing high-dimensional parameters into multiple separate low-dimensional estimation problems, which can be solved in a time- shared manner via pipelining operation. We first present a basic EKF design that estimates the parameters from one TX user to one RX antenna. Then such basic design is time-shared and reused to estimate parameters from multiple TX users to multiple RX antennas. Meanwhile, we use interference mitigation module to cancel the co-channel interference at each RX sample. In addition, we further propose adaptive noise variance tracking module to improve the estimation performance. The proposed design enjoys low delay and low buffer size (because of its online real-time processing), as well as low implementation complexity (because of time-sharing and pipeling design). Its estimation performance is verified to be close to Cramer-Rao bound.