Combining signal processing and machine learning techniques for real time measurement of raindrops

The data acquisition system for a new type of optical disdrometer is presented. As the device must measure sizes and velocities of raindrops as small as 0.1 mm diameter in real time in the presence of high noise and a variable baseline, algorithm design has been a challenge. The combining of standard signal processing techniques and machine learning methods (in this case, a neural network) has been essential to obtaining good performance.