Doppler Tolerant Code Optimisation Scheme for Multi-code Sonar Systems

Underwater acoustic sensing (i.e., sonar) requires probing signals to have a narrow auto-correlation peak with minimal autocorrelation sidelobe levels and minimal crosscorrelations. Another requirement is that the signals should be robust to Doppler effects. Codes achieving the ideal in all these characteristics simultaneously are not possible and compromises must be made in designing probing signals. In this research, we use a novel gradient-descent algorithm for optimising sets of multiple codes for desirable correlation properties while being robust to Doppler scaling. We constrain our optimised code-sets to follow a specific power profile while maintaining a Peak-to-Average Power Ratio (PAPR) close to 1. The codes are optimised for multiple Doppler scaling factors for different source/target speeds. We show that by slightly increasing the mainlobe width from a single sample, lower SLLs can be achieved. This makes our codes flexible for a wide range of applications. The optimised codes from the algorithm have superior characteristics when compared to codes previously published.