Rapid, near‐optimal waveform adaptation: precompute and lookup via hash functions

Abstract

The authors propose a novel library-based waveform adaptation method which allows rapid, near-optimal, waveform adaptation for a broad set of applications. The authors refer to their method as precompute-and-lookup (PAL). In contrast to existing library-based methods, which must evaluate each waveform in the library in order to determine the most suitable waveform, the authors’ method employs hash functions. With the authors’ method, the computational complexity of waveform adaptation is solely determined by the chosen hash function, and does not grow with library size. The authors apply PAL to the problem of waveform adaptation for target detection in the presence of additive coloured Gaussian noise under practical constraints. Monte Carlo simulations show that PAL is capable of producing near-optimal waveforms.