Design of pulse detection system for Measurement While Drilling based on GRNN

Measurement While Drilling (MWD) is an important method to collect downhole information. Mud pulse signal carrying a lot of information parameters of the MWD system is disturbed by varieties of noises in transmission. In this paper pulse detection system, which is based on General Regression Neural Network (GRNN), is designed. In this system it not only eliminates the baseline drift but also avoids the limit of the constant threshold.