| 摘要 |
A digital input signal to be delayed is applied to a two-point linear interpolation filter which imparts delay to the signal proportional to the value of a delay control signal. Errors in both the amplitude and the phase of the delayed signal are minimized by the addition of a correction signal to the delayed signal. The correction signal is provided by applying the input signal to a further filter and a multiplier connected in cascade. The further filter is a linear phase filter having a response zero at zero frequency and a delay equal to an add multiple of one-half of the sampling period, Ts, of the digital input signal. The multiplier is controlled so as to vary the amplitude of the compensating signal as a non-linear function of the delay control signal so as to provide maximum amplitude compensation at delays corresponding to odd multiples of Ts/2 and zero amplitude compensation at delays equal to integer multiples of Ts. |
