In order to improve the accuracy of signal recovery after transmitting over atmospheric turbulence channel, a deep-learning-based signal detection method is proposed for a faster-than-Nyquist (FTN) hybrid modulated optical wireless communication (OWC) system. It takes advantage of the long short-term memory (LSTM) network in the recurrent neural network (RNN) to alleviate the interdependence problem of adjacent symbols. Moreover, an LSTM attention decoder is constructed by employing the attention mechanism, which can alleviate the shortcomings in conventional LSTM. The simulation results show that the bit error rate (BER) performance of the proposed LSTM attention neural network is 1 dB better than that of the back propagation (BP) neural network and outperforms by 2.5 dB when compared with the maximum likelihood sequence estimation (MLSE) detection method.

Sand-dust weather conditions are considered the primary challenge to free-space optical (FSO) communications. It may cause severe attenuation that is malignant to FSO link performance. This study investigates the impact of sand-dust particles on a laser signal using the radiation propagation method and the small-angle approximation method. Numerical simulation shows that in sand-dust weather conditions, the multiple scattering effect is dominant and results in signal pulse delay and pulse broadening. Furthermore, the signal attenuation follows a negative exponential distribution to the laser wavelength. Superior performance can be achieved by employing a longer wavelength laser to reduce pulse delay and mutual interference.

The performance of wireless optical MIMO system with multiple pulse position modulation (MPPM) over correlated fading channel is investigated. The combined effects of atmospheric attenuation, atmospheric turbulence, and pointing error are taken into consideration. The bit error rate (BER) and the ergodic channel capacity are analyzed by utilizing the Poisson counting model and the exponential correlation model. Moreover, their approximate expressions are derived. The simulation results demonstrate that the pointing error is the most prominent influence factor over weak correlated channel. The performance degradation caused by a high channel correlation coefficient is more than that of pointing error in strong correlated channel. Therefore, the use of pointing, acquisition, and tracking (PAT) system and reasonable arrangement of the number and spacing of antennas at the transceiver are the keys to improve system performance.

To mitigate inter-symbol interference (ISI) caused by Faster-than-Nyquist (FTN) technology in a multiple input multiple output (MIMO) optical wireless communication (OWC) system, we propose an ISI cancellation algorithm that combines multi-head self-attention (MHSA), a one-dimensional convolutional neural network (1D CNN), and bi-directional long short-term memory (Bi-LSTM). This hybrid network extracts data features using 1D CNN and captures sequential information with Bi-LSTM, while incorporating MHSA to comprehensively reduce ISI. We analyze the impact of antenna numbers, acceleration factors, wavelength, and turbulence intensity on the system’s bit error rate (BER) performance. Additionally, we compare the waveform graphs and amplitude–frequency characteristics of FTN signals before and after processing, specifically comparing sampled values of four-pulse-amplitude modulation (4PAM) signals with those obtained after ISI cancellation. The simulation results demonstrate that within the Mazo limit for selecting acceleration factors, our proposal achieves a 7 dB improvement in BER compared to the conventional systems without deep learning (DL)-based ISI cancellation algorithms. Furthermore, compared to systems employing a point-by-point elimination adaptive pre-equalization algorithm, our proposal exhibits comparable BER performance to orthogonal transmission systems while reducing computational complexity by 31.15%.

Due to the uneven distribution of the underwater laser transmission channels, the transmission beam will be affected by water absorption and scattering effects, which induce inter-symbol interference and the degradation of link performance. Based on the binomial Henyey–Greenstein scattering phase function and considering the forward and backward effects, the pulse response expression of the non-line-of-sight underwater laser channel is derived. The effectiveness of the underwater photon transmission channel model established is simulated and verified by the Monte Carlo method. The fitting effects in the coastal waters between the theoretical calculation value of channel impulse response and simulation are better. Under the same simulation parameters, the pulse-broadening factor in coastal waters is higher than that in harbor waters, and the time-domain broadening effect of laser pulse transmission in harbor waters is large. When the asymmetry factor is large, the influence of front scattering on pulse broadening is greater.