Optimizing the computational effort of satellite signal acquisition based on mean recognition convolutional neural network

Abstract The ability to synchronize the satellite signals affects navigation performance. Only after a procedure of acquisition, namely coarse synchronization, data information needed position can be extracted from signals. Conventional acquisition methods rely on extending signal integration time improve correlation quality signal. However, coherent results in large computational burden, which is unrealistic for civilian receivers. In this paper, we try optimize effort phase and propose an algorithm based mean value recognition convolutional neural network. matrix with high separated by network, then Doppler shift code introduced, effectively change computation 1/4338 original one. trained model network embedded into receiver directly, has important engineering applications.