An improved NSGA-II-based optimization method for OFDM waveforms in integrated underwater acoustic detection and communication systems

Aiming at the integrated requirements of underwater detection and communication, this paper proposes a joint optimization method based on an improved non-dominated sorting genetic algorithm-II (NSGA-II) to address the high peak-to-average power ratio (PAPR) and high integrated sidelobe level ratio (ISLR) of the aperiodic autocorrelation of orthogonal frequency division multiplexing (OFDM) waveforms. Under the constant-modulus phase constraint, the method optimizes the phases of reserved subcarriers. Additionally, it incorporates adaptive crossover and mutation strategies and applies local gradient refinement to a small number of elite individuals, thereby achieving effective suppression of both PAPR and ISLR. Simulation results show that, compared with the conventional NSGA-II, the proposed strategy converges faster and yields higher-quality solution sets. Furthermore, compared with the Tone Reservation-Gerchberg–Saxton (TR-GS) and TR-L-norm cyclic algorithm (TR-LNCA), the proposed approach achieves lower PAPR and ISLR, while simultaneously improving communication bit error rate performance and enhancing the detection probability of weak targets. Sea trial results further verify the feasibility and effectiveness of the proposed method.