Exploitation Of Known Interference In Multi-user MIMO-OFDM Downlink

There has been considerable interest in MIMO system recently, due to its potential for significant gain in channel capacity, so does OFDM in separating wide band fading channels into several narrow band flat fading channels. In multi-user MIMO downlink, multi-user interference (MUI) will degrade system performance greatly, thus should be suppressed. On the other hand, MUI can also be exploited in base station (BS) to enhance system performance, since the knowledge of MUI is known to BS. This thesis analyzes how interference exploitation based on CI (channel Inverse) could degrade system performance in MIMO-OFDM systems, and proposes two solutions. The main contributions are as follows:1. The key techniques in flat fading channel such as separating interference and interference exploitation based on CI precoding in multi-user MIMO downlink are generalized to frequency selective channel. Meanwhile, simulation model of single-link MIMO-OFDM system is set up.2. Phenomenon of performance deterioration with interference exploitation in multi-user MIMO-OFDM downlink has been discussed based on the statistical property of symbol error rate, and the main reasons are illuminated by simulations.3. Detailed knowledge of pre-detection precoding for CI precoding has been discussed here, which help us avoid the phenomenon of performance deterioration. And significant performance enhancement is verified by simulation.4. Another new solution is proposed, which is used to adjust the power allocated in different sub-channel. And two power allocation strategies are proposed such as inverse power allocation and iteration power allocation. Finally the simulation results show that we can solve the problem by power allocation perfectly.