Adaptive Transmit Power Allocation for Layered Space-Time Architecture with Improved Iterative Parallel Interference Canceller (PIG)

Multiple antenna systems have potential to achieve a high capacity compared to a single antenna system in a Rayleigh fading environment. Layered space time (LST) architecture can attain a tight lower bound on the MIMO channel capacity. In (S. Marinkovic, et al., 2001) it is shown that for LST system, the iterative parallel interference canceller with decision statistics combiner (PIC-DSC) can achieve the performance of minimum mean square error (MMSE) receiver. Besides, once partial channel state information is available at transmitter, substantial performance improvement can be attained through allocating power across various transmitting antennas. In this paper, we address the problem of the design of a power allocation strategy for LST architecture with iterative PIC-DSC at receiver. Minimization of bit error rate is employed as the optimization criterion. Convolutional code is used as constituent code. Compared to existing method, the proposed scheme significantly outperforms the conventional LST system