Partial FFT Demodulation For Doppler Distorted OFDM Signals

Due to the severe temporal and frequency selectivity, underwater acoustic channels aregenerally considered as one of the most challenging communication media. As a modulationscheme, OFDM can decompose a time-varying frequency selective channel into several flatchannels, which enables low complexity one-tap equalization. However, OFDM is highlysensitive to Doppler distortion, wherein the time scaling of the signal will destroy theorthogonality, thus the performance of the traditional data detection scheme will deterioratebecause of the inter-carrier interference. Therefore, reducing the Doppler-induced ICI is oneof the most important problems to be solved in OFDM system. Doppler Effect in underwateracoustic communication is usually modeled as a compression or expansion of the signal. Inorder to effectively compensate for the Doppler signal, the ratio of compression andexpansion of the signal has to be accurately measured. Considering the error in measurement,a compensation are usually required after the resampling.This paper studies the residual Doppler caused by the error in resampling or slowlydrifted transmit/receive elements in stationary environments. The author’s works are listed asfollows:1. Analyze the main cause and impact of ICI in underwater acoustic channels.Qualitative analysis of the ICI caused by frequency deviation has been given. As for the ICIcaused by time-varying channel, the expression of channel frequency response and commonICI suppression method have been discussed, and then given different doppler factors, theperformance of the method has been demonstrated.2. Discuss a Partial FFT demodulation technology in a point-to-point, Doppler distortedunderwater acoustic communication link. Partial FFT modeling and estimating methods ofweighting factor with the help of adaptive filter have been introduced. Finally, take the RLSadaptive filter as an example, an integrated simulation of Partial FFT demodulation has beencarried out and an analysis of the performance has been demonstrated.3. Present several channel equalization methods that can cooperate with the Partial FFTdemodulation in time-varying OFDM system. Discuss the system performance of differentmethods and select the best channel estimation algorithm. Then combine the secondaryequalization and Partial FFT demodulation to further improve the system performance.Finally, the lake trial results verify the effectiveness of the methods described in this chapter.