Adaptive signal processing in subbands using sigma-delta modulation technique.

In this thesis, the use of subbanding and sigma-delta modulation in interference/noise cancellation is intensively studied and a sigma-delta modulated subbanded adaptive interference/noise cancellation system is proposed. The filter bank is fully sigma-delta modulated. The output signal from the filter bank is then used to produce the input to the adaptive filter. The adaptive filter is partially sigma-delta modulated. The output is demodulated at the final stage. Maintaining the sigma-delta modulated signal representation throughout the system results in considerable savings in complexity. The performance of the proposed system is studied and compared to the regular non sigma-delta modulated case regarding complexity, convergence speed and steady state error. The effect of the oversampling rate used in the sigma-delta modulation as well as the quality of the demodulator is also considered. It is shown that in the case of interference cancellation a comb filter is sufficient, while in the case of noise canceller a good quality demodulator is essential. The thesis concludes by highlighting the tradeoffs between the hardware complexity reduction and the overall system performance