| Rake receiver is generally used in direct sequence code division multiple access (DS-CDMA) systems to exploit the multipath diversity. In order to achieve an acceptable quality of service, a good estimator for channel taps is usually needed. The aim of this thesis is to investigate the performance of Rake receiver with adaptive channel estimation for IMT-DS system—a 3G mobile communication standard. Detail results of channel estimation for this system are not yet available in the literature. In this thesis, a new dimension in the channel estimation for wideband CDMA i.e. chip-level adaptive channel estimation has been explored. Simple LMS algorithm is used to update the weights of the adaptive filter because of its simplicity and suitability for this problem. The expression for the optimum step-size is modified for fading channel estimation problem. Simulation results for optimum step-size LMS algorithm show better performance as compared to arbitrary selected step-size. In addition, a new method is proposed to obtain channel estimates with known pilot symbols which is found to give better results than available methods. For slow fading channels, like pedestrian channel, LMS estimator with no update mode is found to give satisfactory results. For fast fading channels, like vehicular channel, a common decision directed technique of channel estimation is modified to be used at chip-level in the downlink (DL). A novel despreader-respreader based channel estimator has been proposed to obtain uplink channel estimates at chip level which resolves the deficiencies of conventional methods. The performance of Rake receiver with proposed channel estimation schemes for IMT-DS system is evaluated in terms of BER by Monte-Carlo simulations for pedestrian and vehicular channels. |
