Joint Channel Estimation and Data Detection for Space-Time Block Coded System via EM Algorithm

Detection and decoding of the space-time coded signals over unknown fading channels is a challenging task. In the paper the expectation-maximization (EM) algorithm according to maximum likelihood (ML) criterion is applied to the problem of joint channel estimation and data detection of space-time block coded system over time-selective fading channels. In order to ensure fast converge of the algorithm, the initial of parameters is carefully chosen by combining the pilot symbols matrix assisted estimation and linear interpolation. At the basis of initialization, a series of iterations of the EM algorithm, in which consists of an estimation phase followed by detection/decoding phase, are utilized to realize the joint estimation and detection/decoding scheme. In decoding phase, linear decoding method is used due to the orthogonality of the space-time block code, thus its computation complexity is greatly reduced. Simulation results show that the scheme works with novel performance