Study On Joint Detection Algorithms Based On MMSE-BDFE

TD-SCDMA is one of the three main criterias of the third generation mobile communication system. It combines the technical advantages of CDMA systems and TDD systems.Using the uplink synchronization, smart antenna, joint detection and other advanced technologies. TD-SCDMA has the highest spectrum efficiency. Therefore, how to strengthen the research on TD-SCDMA System is extremely important.TD-SCDMA system is an interference-limited system which includes multiple access interference, multipath interference and inter-cell interference. The existences of these interferences reduce the frequency of utilization of the system. Joint detection technology is the key techniques of TD-SCDMA system.The basic idea of this technology is taking the interference signal as the useful one that full use of the informations which include signal spreading code, amplitude, timing, delay, etc.. Joint detection technology is available to reduce the multiple access interference and multipath interference.The academic circle and practice circle attach great importance to the joint detection algorithm, which achieves many results, but there are still many problems to be solved. First, the practical application of Zero-Forcing block linear equalizer (ZF-BLE) algorithm detection performance worse than Minimum Mean Square Error linear block equalizer (MMSE-BLE) algorithm performance, Second, the complexity of ZF-BLE and MMSE-BLE algorithm is very high. Therefore, this paper advances mean square error block decision feedback equalizer (MMSE-BDFE) joint detection algorithm, which combines with ZF-BLE, MMSE-BLE algorithm and the FFT used to block cyclic matrix decomposition to reduce the computational method. The algorithm can be used for TD-SCDMA uplink system for rapid detection of multiple users.Using the proposed new method, this paper simulation B.steiner channel estimation method which added threshold processing.And, it reveals the system performance under different users and different channels using the MMSE-BDFE joint detection algorithm. Through simulating of several joint detection algorithms and analyzing the simulation results, this paper draws the conclusion like that:First, MMSE-BDFE joint detection algorithm and the ZF-BLE. MMSE-BLE joint detection algorithms are effective joint detection methods. Second, comparing with the ZF-BLE, MMSE-BDFE algorithm have low bit error rate. Third, in the condition of same ber performance, MMSE-BDFE algorithm has the advantages of low complexity, operational efficiency, Orders of magnitude less than O(n2),double down memory than traditional methods. At last, experimental results show that the method has better detection performance, which can effectively eliminate the multiple access interference and multipath interference to improve system performance.