Channel estimation optimization and lower bound of capacity analysis in MIMO system

In the paper the MIMO block fading channel is estimated and optimized with colored noise. On the basis of the optimization result of the estimated channel, we analyse the lower bound of mutual information and system ergodic capacity with optimal channel estimation error. The waterpouring algorithm is used as optimal transmitter power allocation strategy to maximize the lower bound of the system ergodic capacity over the spatial domain. Simulation results show that some capacity gain is obtained by optimal transmitter power allocation compared with average transmitter power allocation.