Using the Genetic Algorithm to Locate Optimal Bi-Phase Waveforms for Pulse Compression Radar

The Air Force Research Laboratory is developing a spacebased sparse-array aperture system, called TechSat 21. The TechSat 21 experiment is a cluster of three satellites that will fly in formation and operate cooperatively to perform the function of a larger, single satellite. The payload chosen to demonstrate the TechSat 21 concept is radar. Doppler and range ambiguities resulting from the relatively small antennas of each satellite represent significant challenges. The characteristics of transmitted waveforms must be optimized to resolve the anticipated ambiguities. Since the waveforms are bi-phase encoded and can be represented as a binary sequence, the Genetic Algorithm, that uses a binary sequence to represent a gene, was the obvious choice to search for optimal waveforms. The GA approach was first tested against a known standard. The goal was to find waveforms similar to known optimal waveforms. The GA search was then used to locate waveforms that could potentially support the TechSat 21 experiment. One concept to resolve the range and Doppler ambiguities is to transmit and process groups of waveforms. The GA was used to locate pairs of wavetbrms suitable tbr this concept.