Browsing by Author "Amsalu, Tewodros"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    Analysis of Blind Adaptive Equalization Techniques for Audio Broadcast System
    (Addis Ababa University, 2017-05) Amsalu, Tewodros; Hailemariam, Dereje (PhD)
    The wireless channel, in general, and the audio broadcasting range in particular is prone to time dispersion, which causes inter symbol interference (ISI). This dispersive channel must be compensated (equalized) by making a communication system adaptive to the time varying properties of the channel. Practically, in digital audio broadcasting training sequences may not be available to implement the adaptation, which makes blind adaptive equalization methods more appropriate for such application. In this thesis the performance of three blind adaptive equalization techniques; namely, constant modulus algorithm (CMA), multi modulus algorithm (MMA) and fractionally spaced constant modulus algorithm (FS CMA) are investigated for application in audio broadcasting. This investigation is part of a joint project between Information Networks Security Agency (INSA) of Ethiopia and the Addis Ababa University (AAU) that intends to implement advanced digital receiver for a certain application. To capture the broadcasting environment three channel models are used. Moreover, 16-quadrature amplitude modulation (QAM) is considered for the simulation. Symbol error rate (SER), rate of convergence, stability, complexity and audibility are used as performance metrics. The simulation result shows that FSCMA has much better SER performance than CMA & MMA which provide 1 to 1.5 SNR advantages over the other two methods. The three equalizers are able to recover the transmitted audio bit streams when the channel do not introduce phase shift; MMA being the only method capable of recovering phase shift. On convergence, once the algorithms converged they stay stable until the channel condition start to change. Finally, it was noted that the three equalizers are able to equalize the received input data faster than the sampling rate of the audio signal. From all the simulation results, FS CMA has better or comparable performance than the CMA and MMA. Therefore, the thesis recommends FS CMA for implementation by INSA with the assumption that a separate unit will handle the phase recovery implementation. Analysis