Abstract
In downlink, multicell cooperation is a powerful method to mitigate intercell interference and to increase transmit diversity. The cooperation can especially improve the bit error rate (BER) performance under large scale fading. However, those cooperative base stations (BSs) need to share all the channel state information (CSI) among all BSs and this would lead to heavy loading. In this paper, we want to lower the CSI exchange loading under two cooperative BSs and multiuser multiple input and multiple output (MIMO) downlink communication systems. The proposed method only shares one positive value to represent the complex channel matrix for each user and optimizes the power allocation based on the value. The loading can be reduced to 1 over 2N r N t with order of 2.5 BER improvement under the 0dB SNR. Therefore, it could save more transmit power under the same BER performance or cell coverage. © 2012 IEEE.