| Orthogonal frequency division multiplexing(OFDM)has been widely used in communication systems due to its strong ability of anti-multipath interference.Most of the wireless communication channels are not only multipath,but also time-varying.They are called doubly selective channels,which exhibit both time and frequency selective fading.The channels’ time-varying characteristic is more significant in the communication scenario where the moving speed of terminals is high.This can result in Doppler spread,which can destroy the orthogonality between subcarriers seriously,resulting in interchannel interference(ICI).ICI has a remarkable influence on the performance of the OFDM communication system,and existing detection algorithms can’t obtain ideal performance.What’s more,with the increase of subcarriers’ number,the existing detection algorithms are faced with too high complexity.This makes detection algorithm which can balance performance and complexity preferably become one of the crucial difficulty problems.This paper aims to solve the above problems and carries out research on OFDM detection algorithm in doubly selective fading channel.Firstly,this paper introduces the basic principle of OFDM,and provides the channel models for doubly selective fading channels.With the analysis of the ICI,this paper introduces several methods to suppress ICI.Then,some classical linear and nonlinear detection algorithms are introduced,such as the least squares,the minimum mean square error,the ordered successive interference cancellation,and the maximum a posteriori probability detection.Secondly,aiming to reduce the complexity of the detection algorithms,this paper focuses on the detection algorithm based on partial fast Fourier transform(PFFT)and proposes some improvement for complexity reduction.Maximum signal-to-noise ratio(SINR)window is adopted to improve the design of the PFFT algorithm,specifically speaking,to reduce the complexity of the PFFT with only marginal performance loss.PFFT is a new demodulation method in OFDM system.It divides the OFDM symbol into several intervals,then it adds zero symbols after each interval.In this way,the sub interval and the original signal have the same length.Then,the ICI influence on the useful signal can be assigned to the zero part.By this way the ICI can be effectively suppressed.The proposed algorithm in this paper can not only make full use of the advantage of PFFT for its effective ICI suppressing,but also reduce the algorithm’s complexity by a banded structure introduced by adopting maximum SINR windowing.In order to reduce the ICI’s harmful effect on the performance of OFDM system in doubly selective fading channels,we finally carry out research on iterative detection algorithm based on weighted-type fractional Fourier transform(WFRFT),and put forward some improvement measures.Due to its poor performance of linear detection scheme,we propose an optimum order WFRFT based iterative soft decision interference cancellation(ISDIC)algorithm to improve detection performance.Also,being different from the WFRFT in which a fixed optimum order is derived and used based on carrier-interference ratio,we propose to calculate the optimum order and acquire the optimum WFRFT transformation matrix in a real-time manner based on current specific SNR value.The proposed modified WFRFT can obtain a better detection performance with only moderate complexity increase. |
