Research On Bionic Covert Underwater Acoustic Communication By Mimicking Cetacean Whistles

Underwater acoustic communication technology utilizes the sound wave to realize the underwater long-distance communication.And it is one of the key technologies for marine environment observation and target detection.Due to the openness and unreliability of the underwater acoustic channel,the underwater acoustic communication signals are easily intercepted and even subjected to malicious attacks.As a result,the covertness and security of the underwater acoustic communication should be taken into consideration when designing the underwater acoustic communication system.With the aim to obtain the low probability of being detected by eavesdroppers,traditional covert underwater acoustic communication methods mainly utilize spread spectrum and multicarrier modulation to reduce the signal-to-noise ratio of the communication signals.But these methods also reduce the effective distance of the underwater acoustic communication.Instead,bionic covert underwater acoustic communication methods usually adopt the whistles and clicks of cetaceans to carry information.Hence,the eavesdroppers will consider the communication signals as the ocean ambient noise and filter them out,which remedies the limitation of the traditional methods.However,the clicks are not suitable for the actual transducers because of their broadband characteristics.On the contrary,the bandwidths of the whistles are narrow and the whistles usually have good time-frequency characteristics.Therefore,this thesis studies the bionic covert underwater acoustic communication on the basis of cetacean whistles.With regard to the problem that the extraction result of the whistle's time-frequency structure contains outliers,this thesis proposes a time-frequency structure smoothing algorithm based on the improved Sage-Husa adaptive Kalman filter.It is an important prerequisite for the bionic covert underwater acoustic communication research to obtain the pure whistle from the collected original signal.And the short time Fourier transform can usually be used to extract the whistle.However,when the noise energy is larger than the signal energy,the extraction result of the time-frequency structure can be abnormal.This thesis utilizes the improved Sage-Husa adaptive Kalman filter to smooth the extraction result of time-frequency structure,which solves the problem of unknown noise distribution in the extraction result.The simulation results show that the estimation performance of this method is remarkable and the smoothed time-frequency structure matches the real value well.With regard to the loss of the bionic performance during the process of constructing the time-frequency structures of the whistles and the low communication rate,this thesis proposes two bionic communication methods based on the segmented time-frequency structures of the whistles.The whistles,which usually have good time-frequency characteristics,are suitable to modulate the information directly for the aim of preventing the loss of the bionic performance.Moreover,for the purpose of improving the underwater acoustic communication rate,the original time-frequency structures need to be divided into several segments uniformly to represent different symbol signals.And the information is modulated by the time-delay or frequency-shift in each symbol signal.Simulation and experiment results show that the methods both have good concealment performance and communication performance.With regard to the problem that the whistles fail to conceal the modulation signals,this thesis proposes a bionic communication method based on the Chebyshev chaotic signals which are concealed by the whistles.In this thesis,the Chebyshev chaotic signals are used to replace the traditional modulation signals.Their power spectral density is low and the energy is evenly distributed in the whole frequency band.Thus,they show noise-like characteristics.Moreover,they are generated by the cosine function and are orthogonal with each other.To enhance the covertness of the underwater acoustic communication,the modulation signals based on the Chebyshev chaotic signals are superimposed on the timedomain waveforms of the original whistles and the communication signals will show bionic characteristics.The results of the simulation and the experiment show that this method solves the problem of the whistles failing to conceal the modulation signals.And the proposed method also has good concealment performance and communication performance.The bicnic covert underwater acoustic communication methods proposed in this thesis all have good concealment and reliability.And they have a good application prospect in the underwater acoustic sensor networks which require high concealment.