Quantizing and Sampling Considerations in Digital Phased-Locked Loops

Quantization and sampling effects on the digital phasedlocked loop (DPLL) structures obtained for demodulation of anglemodulated signals using extended Kalman filter algorithms are investigated for the high signal-to-noise ratio (SNR) case in this paper. First, the problem of quantization is considered. The validity of the uniform white sequence model for quantizer error in the DPLL is established independent of the sampling rate. Simulation results are presented for several quantizer word lengths. Also, an effective SNR is defined which allows prediction of quantized performance from unquantized results. Secondly, minimum sampling requirements for the DPLL are considered. The effect of sampling rate variation upon the predicted phase error covariance is examined. Again, simulation results are presented and compared to the predicted phase error covariance values. This results in an analytical method for determining minimum sampling rates for the DPLL. Minimum sampling rates for quantized DPLL have also been determined using the effective SNR previously defined.