Detection and separation in space time block coding using noisy compound PCA - ICA model
Abdelkrim Hamza, Salim Chitroub, Moncef Benmimoune, R. Touhami · 2009
For increasing the capacity of the wireless channel, the Space -- Time Block Coding (STBC) has been proposed in the literature. The problem with such scheme is that the accurate channel state information is required. The channel is then estimated by transmitting the training sequences. Such channel estimation causes the spectral efficiency problem, i.e. the useful data rate is reduced. Likewise, the noise presence in the data affects the STBC performances. In this paper, we try to overcome these drawbacks by detecting and separating the transmitted symbols without channel estimation and by including the noise in the global model. So, the noisy compound PCA - ICA model for STBC is proposed here. Using the Bit Error Rate (BER) and Signal to Noise Ration (SNR) as criteria, the obtained simulation results show that these methods are very suitable for transmitted symbols detection and separation in the STBC context.