Research On The Equalization Technique For FRFT-OFDM Underwater Acoustic Communication System

The technology of Orthogonal Frequency Division Multiplexing (OFDM) has many advantages, such as strong resistance to multipath fading, high frequency spectrum utilization, high speed of data transmission and so on. Because of these advantages, OFDM has been widely used in the field of communication network. However, just because of the narrow working frequency band and comb amplitude frequency response characteristics of underwater acoustic channel affect the frequency selective fading of OFDM, so that the system performance of OFDM will drop. The underwater acoustic communication system of OFDM, based on fractional Fourier transform (FRFT-OFDM), uses the fractional Fourier transform to substitute the Fourier transform of the traditional OFDM system to modulate and demodulate. In order to against the frequency selective fading and improve the communication quality of the system, It uses the linear frequency modulation (LFM) signal as subcarrier. In furthermore, many key technologies of FRFT-OFDM system have to be solved, the equalization technique is among them.In the thesis, we mainly study the equalization technique of FRFT-OFDM system. During the study of the equalization technique, we do some works as follows:First, the characteristics of underwater acoustic channel and coherence of multiplicity channel model is studied, and we use the theory of ray to establish the channel model. The definition and characters of fractional Fourier transform and numerical calculation is studied, and we mainly study the basic principle of FFT-OFDM and FRFT-OFDM systems, including the orthogonality, the system model and the realization of modem. In theory, we analysis and compare the anti-noise performance, the response of multiplicity performance and the ability to resist the Doppler effect between the two systems, and further we simulate these comparisons by computer, the results show that FRFT-OFDM systems have greater improved than FFT-OFDM systems in the same channel condition.Next, the characteristics of the fractional convolution in FRFT-OFDM system are different from the characteristics of frequency domain convolutions, so that some traditional frequency domain equalization algorithm can't be directly applied in the FRFT-OFDM system. According to the above problem, the thesis, according to the order of fractional Fourier transform overlay properties, proposes an equalization module based on the frequency domain of FRFT-OFDM underwater acoustic communication system. The equilibrium module will first pass the time domain signals of underwater acoustic channel and FFT transform to frequency domain signal, after equalization processed in the frequency domain, the signal will be recovered by transform of the fractional Fourier. The thesis also studies the equalization algorithm of traditional frequency domain, including zero forcing equalization algorithm, minimum mean square error equalization algorithm, adaptive LMS equalization algorithm and adaptive RLS equalization algorithm. We also analysis the algorithms'error performance in the FRFT-OFDM underwater acoustic communication system through by the computer simulation. The results indicate that adaptive RLS equalization algorithm is the best and the zero forcing equalization algorithm is the worst.Last, the thesis studies the theory and the classification of the time reversal mirror technology. And we also study the decision feedback equalization technology. Owing to the serious time delay expansion of the underwater acoustic channel, the filter which is needed in the decision feedback equalization has high order. It increases the complexity of the system, in additional, the time reversal mirror technology can not totally eliminate inter-symbol interference during the process of the equalization. In order to overcome the shortcomings of these two equalization methods, we combine the time reversal mirror technology and the decision feedback equalization technology. By using the decision feedback equalization technology, the problem of having remnants inter-symbol interference in the time reversal processing can be solved. At the same time, the time reversal mirror technology can decrease the number of the time delay order in the decision feedback equalizer. With the action of these two methods, we can improve the communication quality of the multi-path condition. We also prove the feasibility of this means in the FRFT-OFDM system by simulation experiment. The results of BER curves and Constellation indicate that in this way can improve the error performance effectively, the phase information can be completely separated.