The Research Of Synchronization Technique For MC-CDMA Systems

Recently, data transmission rate in the communication networks for now cannot satisfy the demand of users with the explosive development of wireless communication technique, so people make a flurry of research and development activities dedicated to beyond 3G mobile communication systems. MC-CDMA technology based on CDMA and OFDM technologies makes a flurry of research and development activities in beyond 3G mobile communication systems. The technology is not sensitive to time delay of channel, and can realize frequency diversity. So it may resolve the question of high speed data flow transmitted in wireless channel, and its frequency spectrum availability is high. MC-CDMA systems is sensitive to frequency offset and the start position offset of frames that are brought in by Doppler frequency expanding and mismatch between transmitter's and receiver's oscillator. In order to assure the performance of MC-CDMA systems, many researchers focus their research to the synchronization algorithms in MC-CDMA systems.The synchronization algorithms used in MC-CDMA systems are the same as those in OFDM systems because of the signal structure and spread spectrum code used in MC-CDMA systems. The synchronization algorithms for MC-CDMA can be loosely grouped into non-blind algorithms and blind algorithms. Presently a lot of synchronization schemes have been proposed, but there always is a contradiction between the estimation performance and the implementation complexity. Because of the reason above, a synchronization model applied in multi-path channel based on ML algorithm is improved. And a novel synchronization model based on CORDIC algorithm in digital field is proposed. Simulation results show that the estimation performance of the improved model that uses pulse shaping function is better than conventional model when SNR satisfies some conditions. It is more applicable to multi-path channel. And it can get over the problem of the adjustment of factor of filter in hardware implementation in conventional model. Because the CORDIC algorithm can change the arctan function and round of angle to addition and shift operation, the novel model presented in the paper can complete frequency offset estimation and correction in digital field in the same time. To improve the estimation performance of the model, qualification level is increased. The simulation result indicates that compared with the conventional model, the novel model using CORDIC algorithm can decrease the implementation complexity on the condition of the same estimation precision. The model in digital field which apply higher qualification level, then estimation precision is better.