Parameter Estimation Of MC-CDMA And MC-DS-CDMA Signals

Multicarrier CDMA sufficiently keeps the respective advantages of code division multiple access and orthogonal frequency division multiplexing such as low probability of interception, high qualities of anti-interference, high spectrum efficiency, high capacity and so on. There are three kinds of combinations for Multicarrier CDMA: MC-CDMA, MC-DS-CDMA and MT-CDMA. Multicarrier CDMA technology can be widely used in mobile communications, wireless local area network(WLAN), ultra-wide bandwidth(UWB) communication system, and provides a variety of business types, what's more, MC-CDMA is a kind of B4 G technology.In this paper, MC-CDMA and MC-DS-CDMA signals are chosen as the research object, and their signal characteristics can be extracted by taking advantage of related algorithms and properties of CDMA and OFDM signals in non-cooperative communication environment. Specific studies are as follows:(1) To solve the pseudo-noise(PN) sequence' period in MC-DS-CDMA signals in this paper, the method based on reprocessing of the power spectrum has been studied. This algorithm takes advantage of the orthogonality characteristics among different carriers and among different users' PN sequences. A power spectrum of the received signal is computed, and is treated as an input signal to compute its secondly power spectrum density, then the value is squared. Through detecting the interval of the peak pulses, the period of PN sequence of MC-DS-CDMA signals can be obtained.(2) A modified cyclic autocorrelation algorithm is proposed to solve the estimation problem of the symbol duration, useful symbol duration, and chip duration in MC-CDMA signal. Firstly the autocorrelation function of the received MC-CDMA signals is computed, then through Fourier transform and accumulation in frequency-domain, parameters mentioned above can be estimated. Meanwhile, according to the characteristics of MC-CDMA signal, the accumulative average method is developed to estimate the symbol duration specially. By averaging amplitudes of the spectral lines in each column parallel to the time-delay axis in the three-dimensional diagram, the symbol duration can be obtained effectively. In addition, cyclic autocorrelation can be a method to distinguish MC-CDMA and OFDM signals.(3) The singular value decomposition(SVD) method has been used as a modulation recognition method to distinguish MC-CDMA from MC-DS-CDMA signals, and meanwhile, estimate the pseudo-noise sequence of MC-CDMA. At the receiver, after removal of the cyclic prefix, N ?M matrix can be structured on the basis of the PN's length, where N is the number of subcarrier, and M is the integer times of N, then SVD is applied to the autocorrelation matrix which is after fast Fourier transform(FFT). According to the different number of characteristic values, these two signals can be distinguished. After MC-CDMA signal has been determined, pseudo-noise sequence can be estimated by left singular when the power between different users is a little larger.