Study On Signal Detection For OFDM Systems Over Doubly Selective Channels

The channel undergoes frequency selective fading caused by multipath propagation which results in the inter-symbol interference at the receiver. The interference significantly restricts to further increase the data transmission rate. To relieve the effect of frequency selective fading, Orthogonal frequency division multiplexing is widely adopted in many broadband wireless communication standards due to its advantages of high spectral efficiency and robustness over multipath channels. As the development of demand on the communication, OFDM system is required to provide quality of service guarantee, not only in terminals static or move slowly environments, but also in high speed mobile scenario. However, high speed mobile will lead to channel produce time selective fading while the channel has frequency selective fading. In this doubly selective channel, multipath superposition with different time delay and Doppler frequency shift will form the Doppler spreading. In particular, the effect of Doppler spreading destroys the orthogonality of the sub-carriers, resulting in inter-carrier interference (ICI), which deteriorate the systems performance severely. On the one hand, the existing detection algorithm degrade severely at larger Doppler frequency shift, on the other hand when the number of subcarriers is larger, the existing detection algorithm is difficult to implement in practical system. Focusing on this performance degradation problem of OFDM systems, the dissertation deeply analyzes and investigates practical and effective signal detection techniques in OFDM systems over doubly selective channels. The main work and contributions of this dissertation are listed as follows:1. In the analysis of ICI characteristics for OFDM systems over doubly selective channels, we first give the cyclic prefix (CP) OFDM systems models and signal models. Based on the signal models of OFDM systems, the mechanism of the multipath propagation and Doppler frequency shift forming the doubly selective channels arc analyzed. Then two classical channel model methods for doubly selective channels, tapped-delay line (TDL) and basis expansion model (BEM), are formulated. We analyze the process of how Doppler spreading producing ICI for OFDM systems over doubly selective channels. Then the ICI energy distribution on sub-carriers and the power of ICI varies with the number of subcarriers and Maximum Normalized Doppler Frequency (MNDF) are studied. From the points of view of Signal to Interference Ratio (SIR) and Signal to Interference plus Noise Ratio (SINR), we investigate the ICI effects on OFDM systems due to the Doppler spread. Finally, the simulation model of doubly selective channels adopted by following chapters is introduced, which lay a foundation for the research on signal detection in OFDM systems over doubly selective channels.2. In performance improvement of signal detection for OFDM systems over doubly selective channels, the MMSE-SIC algorithm based on whitening residual ICI plus noise and the optimal linear successive detection algorithm based on oversampling OFDM systems are proposed respectively. These two algorithms are as follows.(1) To reduce the complexity, the existing algorithms treat the residual ICI as white noise, which cause a decline in system performance. Consequently, an MMSE-SIC detection algorithm based on whitening residual ICI plus noise is proposed. Firstly, through analyzing the reason of performance losses due to band structure, we study the correlation of residual ICI after considering the dominant terms. According to the correlation, then the whiten matrix is constructed to whiten the residual ICI and noise simultaneously. Based on the whiten signal model, jointing the MMSE criterion and SIC detects the signal over each sub-carrier. The simulation results demonstrate that the proposed algorithm can improve the BER performance significantly comparing the conventional method exploits the band structure of the frequency domain channel matrix for OFDM systems over doubly selective channels.(2) Although the linear combining algorithm based on oversampling OFDM system can obtain the multipath diversity and Doppler diversity provided by doubly selective channels with lower complexity, the performance degrade. We propose an optimal linear successive detection algorithm based on oversampling OFDM system. This algorithm oversample at the receiver firstly, then based on the oversampled signal through maximum SINR to deduce the optimal linear detection matrix. To further cancel the residual ICI, the successive detection is introduced. To reduce the additional computational complexity caused by oversampling, a low complexity implementation method for this algorithm is also presented, which keep the computational complexity unchanged with the oversample factor larger. Simulations results show that the proposed algorithm can enhance the BER performance with the increase of the Doppler spread and the oversampled factor, which outperforms the linear combination method.3. In low complexity implementation of detection method for OFDM systems over doubly selective channels, the conventional algorithm is at cost of the performance loss to reduce the computational complexity. This dissertation extends firstly the channel matrix and detection matrix, then establishes the relationship between them by using recursive inversion instead of direct inversion at each detection. The first time inversion for extended matrix exploits the Greville recursive algorithm. The following inversions for extended matrix can deduce from the last time results. Both theoretical analysis and simulation results show that the improved algorithm retains the performance of the original algorithm with much lower computational complexity compared to the original MMSE-SD algorithm.4. In hybrid carrier scheme combating doubly selective fading, an optimal order selecting scheme for 4-Weighted Fractional Fourier Transform (4-WFRFT) over doubly selective channels is developed. We firstly deduce the expression of Carrier to Interference Ration (CIR) and then through maximizing the CIR to obtain the optimal order fector. Through selecting the optimal order factor of 4-WFRFT, this system can match the doubly selective channel characteristics through switching between OFDM and single carrier (SC) system according to obtained channel state information. Simulation results show that the optimal order 4-WFRFT scheme can improve the performance over doubly selective channels with respect to the traditional SC or OFDM.