Blind Parameter Estimation Of MC-CDMA Signals

Wireless broadband access systems operate in high mobility and high data rate environments.In order to make the system applied in multipath fading channel conditions,it is crucial to improve the spectral efficiency and combat the frequency selective fading.A Multi-Carrier Code Division Multiple Access(MC-CDMA)scheme is proposed to solve these problems to a large extent.The MC-CDMA technology is one of three forms of multi-carrier CDMA technology,which combines the advantages of Code Division Multiple Access(CDMA)technology and Orthogonal Frequency Division Multiplexing(OFDM)technology.In CDMA technology,MC-CDMA technology has the characteristics of large system capacity.In OFDM technology,MC-CDMA technology can not only transmit data at a high rate,but also can resist the influence of multi-path fading channels.With the advent of MC-CDMA technology,the signal parameters blind estimation is very important.In this thesis,the parameters blind estimation of MC-CDMA signal is studied.The parameters blind estimation of MC-CDMA signal is performed under non-cooperative communication condition by using the algorithm of CDMA and OFDM signal parameters estimation.The main work of the thesis is as follows:(1)In order to solve the problem of sub-carrier parameter estimation of MC-CDMA signal under Gaussian white noise condition,the fourth-order cyclic cumulants algorithm is studied.By analyzing the fourth-order cyclic cumulants of MC-CDMA signal,if and only if the cycle frequency equals the subcarrier frequency of the signal,the fourth-order cyclic cumulants of the signal are not zero,so the subcarrier parameters can be accurately estimated under Gauss white noise condition.The single carrier frequency can be estimated accurately under Gauss white noise condition by analyzing the fourth-order cyclic cumulants of single carrier signal,and further the MC-CDMA signal and single carrier signal can be identified.(2)For the blind estimation of MC-CDMA signal spreading sequence period in multipath fading channels,two algorithms of autocorrelation second-order moment and cyclic autocorrelation are studied.By analyzing the second-order moment of the auto-correlation of the MC-CDMA signal,a peak appears when the delay is equal to a multiple of the spreading sequence period of the MC-CDMA signal,so that the distance between adjacent peaks can be searched to estimate the spreading sequence period of the MCCDMA signal.By analyzing the cyclic autocorrelation of MC-CDMA signal under multipath fading channel conditions,when the cyclic spectrum is equal to the reciprocal of the MC-CDMA signal spreading sequence period,the cyclic autocorrelation function peaks,thereby the spreading sequence period of the MC-CDMA signal is estimated.Finally,the performance of these two algorithms is compared and analyzed under different channel conditions.(3)For the problem of users' number estimation of MC-CDMA signal under Gaussian white noise,the AIC criterion,MDL criterion,HQ criterion and the Gai's circle algorithm are studied.The estimated accuracy of the number of users of MC-CDMA signal under different signal-to-noise ratio conditions,the estimated accuracy of the number of users when the bits of input information codes are different under the same signal-to-noise ratio condition,the estimated accuracy of the number of users when the number of users of the input signal is different under different signal-to-noise ratio conditions are used as indexes to analyze the performance.