A deep learning-based cyclic shift keying spread spectrum (CSK-SS) underwater acoustic (UWA) communication system is proposed for improving the performance of the conventional system in low signal-to-noise ratio and multipath effects. The proposed deep learning-based system involves the long- and short-term memory (LSTM) architecture-based neural network model as the receiving module of the system. The neural network is fed with the communication signals passing through known channel impulse responses in the offline stage, and then directly used to demodulate the received signal in the online stage to reduce the influence of the above factors. Numerical simulation and actual data results suggest that the deep learning-based CSK-SS UWA communication system is more reliable communication than a conventional system. In particular, the collected experimental data show that after preprocessing, when the communication rate is less than 180 bps, a bit error rate of less than 10⁻³ can be obtained at a signal-to-noise ratio of −8 dB.

In mobile underwater acoustic (UWA) communications, the Doppler effect causes severe signal distortion, which leads to carrier frequency shift and compresses/broadens the signal length. This situation has a more severe impact on communication performance in the case of low signal-to-noise ratio and variable-speed movement. This paper proposes a non-data-aided Doppler estimation method for M-ary spread spectrum UWA communication systems in mobile scenarios. The receiver uses the spread spectrum codes dedicated to transmitting signals with different frequency offsets as local reference signals. Correlation operations are performed symbol by symbol with the received signal. The decoding and Doppler estimation of the present symbol are achieved by searching the correlation maximum in the code domain and frequency domain. The length of the current symbol is corrected for the next symbol synchronization using the estimated Doppler coefficient. To optimize the process of Doppler estimation and symbol synchronization, a heuristic search method is used. By adjusting the Doppler factor search step size, setting the threshold value, and using the Doppler factor estimation of the previous symbol, the search range can be significantly reduced and the computational complexity decreased. The Fisher-Yates shuffle algorithm is used to traverse the search range to ensure reliability of the results. Simulation results show that enlarging the frequency-domain search step size in some degree does not affect the decoding accuracy. On 15 May 2021, a shallow-water mobile UWA spread spectrum communication experiment was conducted in Weihai, China. The horizontal distance between the transmitter and the receiver is 3.7–4.0 km, and the communication rate is 41.96 bits per second. The transmitting ship moves at a speed of 0–3 m/s, and the bit error rate (BER) is lower than 1e−3, which is better than that of the sliding correlation despreading method with average Doppler compensation.

This paper proposes a frequency hopping binary frequency shift keying underwater acoustic (UWA) communication system, where a denoising diffusion probabilistic model (DDPM) and a convolutional neural network (CNN) are sequentially used for signal reconstruction and signal demodulation, respectively. Unlike the deep transfer learning (DTL)-based system, this system employs a DDPM to process the received Mel-spectrogram, reconstructing the distorted Mel-spectrogram caused by UWA channel effects, and generating a spectrogram that approximates the transmitted signal, which is then demodulated by the CNN. The proposed system outperforms conventional systems and achieves performance comparable to DTL-based systems in simulation and experiment. DTL requires data samples from new scenarios to learn signal characteristics during deployment; in contrast, this method uses the generative capability of DDPM to enable direct deployment in dynamic underwater environments without additional adaptation processes, offering flexibility and suitability for complex and variable UWA propagation channels.

This paper proposes a deep learning-based code index modulation-spread spectrum (CIM-SS) underwater acoustic (UWA) communication system. The system is characterized by the variant model of the recurrent neural network at the receiver of the communication system, which can directly demodulate the received signal after the synchronization without de-carrier and de-spreading operation. To verify the performance of the deep learning-based CIM-SS UWA communication system, the channel impulse response will be used to simulate the UWA channel. The signals passing through the UWA channel will be used for offline training and online testing of the neural network model. The bit error rate performance of the system with different model structures under different signal-to-noise ratio (SNR) conditions is compared. The simulation results show that the deep learning-based system can achieve better performance than the conventional system under the conditions of low SNR and severe UWA channels.

Structural health monitoring is an important way to ensure the safe and reliable work of the structures during the service period. Surface crack is one of the important indicators for monitoring and assessing the damage of structures. However, crack detection still mainly relies on human visual observation and recording. Besides, people could only record the crack width of a few points instead of all points. To solve this problem, this paper aims to propose a method which can automatically detect crack and calculate the width of every point on the crack but not several points. Thus, this paper built a large database containing more than 100,000 pictures which were taken in different scenarios, and used this database and self-designed deep convolutional neural network to detect cracks. And the prediction accuracy of CNN reached 95%. In addition, combined with image processing technology, this paper further carried out crack extraction and width calculation. Thus, the whole process of crack detection, extraction and width calculation is fully automated, and the calculation results were proved to be close to the actual results, which meets the engineering requirements and perform better than traditional OTSU algorithm.