A comparison of five algorithms for tracking frequency and frequency rate-of-change

An experimental investigation of five simple algorithms for tracking frequency and frequency rate-of-change is reported. The algorithms were developed in an evolutionary fashion from an αβ tracker. Two of the algorithms consist of αβ trackers followed by low-order, finite-impulse-response differentiators for frequency-rate estimation. The final two algorithms under consideration represent an attempt to incorporate the finite-impulse-response differentiators within a recursive tracking loop. The cascade connection of a low-order differentiator with an αβ tracker offers significant advantages in frequency-rate estimation, while the incorporation of these differentiators in the loop offers further benefits. The various algorithms are compared on the basis of the variance of the frequency and frequency-rate estimates, as well as the delay introduced in obtaining the derivative estimates