An event driven technique for filtering computational complexity reduction

This work is a contribution to enhance the signal processing chain required in mobile systems like satellites, cell phones, biomedical implants remote motors, etc. The system is powered by a battery therefore it must be power efficient. Filtering is a basic operation, almost required in every signal processing system. The classical filtering is time-invariant, the sampling frequency and the filter order remains unique. Therefore it can render a useless increase of the processing activity, especially in the case of sporadic signals. In this context an adaptive rate filtering technique, based on an event driven sampling is devised. It adapts the sampling frequency and the filter order by analysing the input signal characteristics. It correlates the processing activity to the signal variations. The computational complexity and the output quality of the proposed technique are compared to the classical one for a speech signal. Results show a drastic computational gain, of the proposed technique compared to the classical one, along with a comparable output quality.