Objective Traditional underwater acoustic communication and underwater radio frequency communication technologies, due to their low transmission rates, limited bandwidth, and susceptibility to interference, can no longer meet the modern underwater communication requirements for high-speed and reliable information transmission. Underwater wireless optical communication (UWOC), as an emerging communication technology, has gradually become one of the key technologies in the field of underwater communication due to its advantages of high bandwidth, low latency, and strong anti-interference capability. However, oceanic turbulence has a significant impact on the transmission performance of optical signals. Caused by ocean currents, the activities of marine organisms, and temperature and salinity gradients in seawater, oceanic turbulence leads to temporal and spatial fluctuations in the refractive index of seawater, which in turn affects the quality of optical signal transmission. Therefore, studying how to achieve efficient and reliable underwater optical communication in the presence of oceanic turbulence has important theoretical and practical significance.

Methods Based on the anisotropic oceanic turbulence model and the Gamma-Gamma turbulence channel model, this paper presents an in-depth study of the communication performance of fixed-length digital pulse interval modulation (FDPIM) in anisotropic underwater turbulent environments. FDPIM encodes binary data into fixed-duration dual-pulse symbols and transmits information by varying the interval between adjacent pulses, thereby overcoming the variable symbol-length issue inherent in conventional digital pulse interval modulation.

Results and Discussion  When M=3, as SNR increases, the slot error rates of PPM, DPIM, and FDPIM all decrease significantly and become close to each other; Among them, FDPIM is closer to PPM in the high-SNR region, while MDPIM exhibits the highest slot error rate due to its lower symbol separability. When maintaining the same SNR and increasing the modulation order M of FDPIM from 3 to 6, its slot error rate decreases overall, indicating that a higher-order encoding space enhances decision robustness, but it also imposes higher requirements on implementation complexity, synchronization, and decision threshold design. Under Gamma-Gamma anisotropic ocean turbulence, increasing the anisotropy factors in the x and y directions can reduce the packet error rate at a given SNR because energy becomes more concentrated along the dominant directions, mitigating decision errors caused by multipath and scintillation. Meanwhile, with R_SN=20 dB and both anisotropy factors equal to 2, the packet error rate increases monotonically with transmission distance, primarily due to increased path loss leading to a reduced effective SNR, thereby increasing the probability of decision errors.

Conclusion The results demonstrate that FDPIM exhibits favorable communication performance in anisotropic oceanic turbulent environments. In terms of average transmit power, the average transmit power of FDPIM decreases with the increase of modulation order M, and approaches that of M-ary digital pulse interval modulation under high-order modulation. Regarding peak power, the peak power of FDPIM increases as M rises; However, compared with pulse position modulation (PPM) and digital pulse interval modulation, it requires lower peak power. In terms of bit rate per unit, the bit rate per unit of FDPIM first increases and then decreases with the increase of modulation order. It is lower than that of PPM and MDPIM under low-order modulation but gradually approaches that of PPM under high-order modulation. As for the slot error rate, the slot error rate of FDPIM decreases with the increase of signal-to-noise ratio and becomes close to that of PPM under high SNR conditions. In terms of the packet error rate, the packet error rate of FDPIM decreases with the increase of SNR and the anisotropy factor, indicating that a higher anisotropy factor helps enhance the anti-interference capability of the signal. In addition, the packet error rate rises significantly with the increase of transmission distance, which suggests that the signal suffers more energy loss and noise interference during long-distance transmission.