Cancellation for Periodic Disturbance with uncertain frequency

A new feedback periodic disturbance cancellation system and adaptive frequency identification algorithm is being developed that will find its application in many areas of signal processing. The core of the system is Internal Model. The algorithm has capability of tracking time varying frequency and canceling pseudo-periodic disturbances.

Application of Adaptive Internal Model Control Based Frequency Estimation Algorithm

In some applications of signal processing, it is desired to identify the frequency of sinusoidal signals or a narrow band noise from observed time series, such as power system protection, communications, radar signal, active noise control, etc. The often used technique is notch filter or adaptive notch filter

Alternatively frequency can be estimated by using a feedback adaptive control system. For control systems, internal model (IM) principle states that by introducing the periodic noise model in feedback loop, perfect noise rejection can be achieved. An algorithm is being developed on this principle. This algorithm is based on the state space form of an internal model controller. If we take a  sinusoidal signal of frequency 'ω', the controller can cancel all the noise at this frequency. Then a non-linear mapping function is used to update the frequency of IM adaptively. An integrator can eliminate this difference to be 0 which means the convergence to the actual frequency.

Click here for publications