Browsing by Author "Dida, Degebassa"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    Performance of Beamforming Techniques Using Uplink Channel Characteristics in Cellular CDMA Communication Systems
    (Addis Ababa University, 2007-09) Dida, Degebassa; Ayele, Hailu (PhD)
    The demand of code division multiple access (CDMA) cellular communication has grown remarkably in the past few years. Due to this reason mobile communication networks should use the limited resources of the system in efficient and convenient manner. One of the techniques is digital beamforming, in which energy is dynamically adjusted to the desired direction. In beamforming, we multiply a complex-valued weighting vector by the outputs of antenna array and sum these outputs to generate a signal for each user. As each user will have a unique weight, we can select the weights to greatly decrease interference from other users, and therefore increase system capacity. In this thesis, the uplink CDMA cellular systems is modeled and investigated to show how different beamforming algorithms are effective in improving system performance and capacity using adaptive antenna arrays at base station. It was initially thought that only the uplink would be the capacity limiting link in CDMA cellular systems but this assumption turned out to be incorrect and the capacity in actual fact is limited by the both links. To determining uplink CDMA system capacity and the effect of inter-element antenna array spacing, step size and number of antenna elements on the overall system capacity using adaptive antenna array elements. Adaptive beamforming is more complex, but highly efficient where the radiation pattern is constructed dynamically in which interferers are blocked and beam is formed in the direction of users. The adaptive algorithms such as, LMS and CMA are investigated to estimate error and to compute array factor for CDMA system, because the spreading codes can be used as a reference for beamforming. Results of comparison show that the LMS algorithm performs better than the CMA algorithm under the same condition. The performance obtained from using beamforming techniques using six (6) adaptive antenna array element results in an improvement by about 30% on average over the non beamforming techniques from the simulation result obtained.