General Unbiased FIR Filter With Applications to GPS-Based Steering of Oscillator Frequency

The general unbiased finite-impulse response (UFIR) filter proposed in this brief has important structural advantages against its basic predecessor. It can be applied to systems with or without the control input. We derive this filter in a batch form and then design its fast iterative Kalman-like algorithm using recursions. The iterative UFIR algorithm proposed is applied to the three-state polynomial model which is basic in clock synchronization. We test it by the global positioning system-based frequency steering of an oven-controlled crystal oscillator. Better robustness and higher accuracy of the UFIR filter against the Kalman filter are shown experimentally.