Sigma-delta modulation with powers-of-two quantizer

The authors study sigma-delta modulation (SDM) with a powers-of-two quantizer. The performances are obtained by simulation. It is observed that the power of the minimum quantization noise stays approximately the same when a powers-of-two quantizer replaces the uniform quantizer, which is traditionally used in SDM. However, the quantization noise of powers-of-two quantizers becomes more stable with respect to the step size, so that a nonoptimum choice of the step size has only a small effect on the performance. Since the values of the SDM signal represent powers of two, the lowpass filter in the demodulator can be implemented without using multipliers. The application of such SDM for implementing FIR (finite impulse response) filters is also investigated. The coefficients of FIR filters are encoded using SDM with powers-of-two quantizer. In addition to requiring no multiplier, results show that ripples of such FIR filters are lower than those based on SDM with uniform quantizer