Direct wireless communication through sea–air media is essential for constructing an integrated communication network that spans space, air, land, and sea. The amplitude of acoustically induced micromotion surface waves is much smaller than the noise interference in complex sea states, making the accurate extraction of these signals from the raw signals detected by an FMCW millimeter-wave radar a major challenge. In this paper, Butterworth filtering is used to extract underwater acoustic signals from the surface waves detected by radar. The physical processes of the channel were simulated theoretically and verified experimentally. The results demonstrate a fitting coefficient of 0.99 between the radar-detected water surface waves and the simulation outcomes, enabling the effective elimination of noise interference and the extraction of acoustically induced micromotion signals in environments with a signal-to-noise ratio (SNR) of −20 dB to −10 dB. Experiments modifying frequency and linear frequency modulation have verified that the usable frequency range for underwater acoustic signals is at least 400 Hz, meeting the frequency requirements of Binary Frequency Shift Keying (2FSK) modulation encoding methods. This research confirms the accuracy of the simulation results and the feasibility of filtering and extracting underwater acoustic signals, providing a theoretical basis and an experimental foundation for building cross-media communication links.

Implementing information encoding in laser acoustic signals by altering the medium’s color is one of the current hot research topics. Modulating the color of the medium can modulate the directionality of laser acoustic signals; however, there has been little research on the impact of water color on laser acoustic signals. This paper investigates the relationship between the directionality of laser acoustic signals and water color, innovatively proposing a conical sound source model. It points out that the ratio of the model’s radius (r) to the model’s vertical line (d) is a decisive factor affecting directionality. Through simulations and experiments, it has been verified that laser acoustic signals exhibit no significant directionality (r/d = 10) and that the energy distribution of sound signals in the vertical direction significantly decreases (r/d = 0.4). Sound signal directivity and absorption rate were studied in the environment of red, blue, and yellow water, and the time–frequency characteristics were also studied. The acoustic signals produced by laser breakdown of different colors have obvious time–frequency characteristic differences, among which the signal intensity generated by laser incident on yellow water is 180.13 dB and the signal intensity generated by laser incident on black water is 168.31 dB. The peak frequency of sound signal generated by laser breakdown of yellow water is the highest, which is 21,240 Hz, and the peak frequency of sound signal generated by laser breakdown of water is the lowest, which is 8828 Hz. There is an obvious positive correlation between the peak frequency of sound signals and the laser absorption rate, and calculation of the optimal water color corresponding to the highest detection threshold at different distances provides guidance for the application of laser acoustic communication.

Water surface micro-amplitude waves (WSMWs) of identical frequency are elicited as acoustic waves propagating through water. This displacement can be translated into an intermediate frequency (IF) phase shift through transmitting a frequency modulated continuous wave (FMCW) towards the water surface by a millimeter-wave radar, and information transmission across the water–air interface is achieved via the signal reconstruction method. In this paper, a novel mathematical model based on energy conversion from underwater acoustic to vibration (ECUAV) is presented. This method was able to obtain WSMW vibration information directly by measuring the sound source level (SL). An acoustic electromagnetic wave-based information transmission (AEIT) system was integrated within the water tank environment. The measured distribution of SL within the frequency range of 100 Hz to 300 Hz exhibited the same amplitude variation trend as predicted by the ECUAV model. Thus, the WSMW formation process at 135 Hz was simulated, and the phase information was extracted. The initial vibration information was retrieved through a combination of phase unwinding and Butterworth digital filtering. Fourier transform was applied to the vibrational data to accurately reproduce the acoustic frequency of underwater nodes. Finally, the dual-band binary frequency shift keying (BFSK) modulated underwater encoding acoustic signal was effectively recognized and reconstructed by the AEIT system.