Publications Repository - Gdańsk University of Technology

Doppler sensors are commonly used for movement detection and speed measurement. However, electromagnetic interference and imperfections in sensor construction result in degradation of the signal to noise ratio. As a result, detection of signals reflected from moving objects becomes problematic. The paper proposes an algorithm for reduction of distortions and noise in the signal received from a simple, dual-channel type of a Doppler sensor. The proposed method is based on examining phase relationship between I/Q channels of the sensor signal. A weighting function is calculated in order to suppress the distortions while preserving energy of the desired signal. Additionally, the proposed algorithm may select signals reflected by objects moving in a specific direction (e.g. towards the sensor). The processed signal may be further analyzed in order to detect signal frequency and compute the object velocity. The results of the experiments show that the proposed approach results in significant reduction of level of noise and interference, allowing for detection and tracking of signals reflected from moving objects.

Authors

• dr hab. inż. Grzegorz Szwoch link open in new tab