Research On Underwater Current Field Communication Technology Based On Duffing Oscillator

With the development of human society,the important value of the ocean to national development has received more and more attention.China has approximately 4.7 million square kilometers of territorial waters and nearly 20,000 kilometers of continental coastline,which not only contains abundant underwater resources,but also brings many maritime security issues.Underwater wireless communication Technology is one of the necessary technologies for the exploration,development and protection of the underwater environment.At present,underwater wireless communication technology generally includes underwater optical communication,underwater acoustic communication,electromagnetic-wave communication and current-field communication.Among them,current-field communication has attracted much attention because of its high flexibility and less impact on environmental factors.The principle of current field communication is that the transmitting end excites a changing electric field in the transmission medium,and the receiving end uses an electric dipole plate to induce the change of the electric field to obtain information.However,the voltage received by the receiving end is very weak.In order to improve the communication distance,this paper proposes to use Duffing oscillator to receive and demodulate weak signals.The Duffing oscillator system has excellent ability to detect weak signals at specific frequencies and filter out-of-band signals.In underwater communications,high-frequency noise is easily absorbed by seawater,so background noise is relatively cleaner.The environment is suitable to apply the parameter-sensitive Duffing oscillator system.Based on the analysis of the communication characteristics of the underwater current field,this paper makes a comprehensive simulation analysis of the dynamic characteristics and parameter characteristics of the Holmes-type Duffing equation,and focuses on the system's resistance to Gaussian white noise,which reflects the excellent anti-noise characteristics of the system.According to the characteristics of the output signal of the vibrator,a rectangular-domain segmentation method based on the idea of phase diagram segmentation is proposed.Compared with other methods,the calculation amount is smaller and the discrimination speed is faster.In this paper,a system structure for receiving and demodulating low SNR digital signals with Duffing equation as the core is constructed,and the performance of the demodulation structure is simply analyzed to verify its excellent ability to receive and detect low SNR signals.Then,this paper proposes to use the BP(Back Propagation)neural network to form a decision device,which uses the neural network to learn the characteristics of the signal to make a binary decision on the output of the rectangular-domain-separator,and confirms the advantage of this method by computer simulation.