Non-Supervised Learning for Spread Spectrum Signal Pseudo-Noise Sequence Acquisition

An idea of estimating the direct sequence spread spectrum（DSSS） signal pseudo-noise（PN） sequence is presented. Without the apriority knowledge about the DSSS signal in the non-cooperation condition, we propose a self-organizing feature map（SOFM） neural network algorithm to detect and identify the PN sequence. A non-supervised learning algorithm is proposed according the Kohonen rule in SOFM. The blind algorithm can also estimate the PN sequence in a low signal-to-noise（SNR） and computer simulation demonstrates that the algorithm is effective. Compared with the traditional correlation algorithm based on slip-correlation, the proposed algorithm＇s bit error rate（BER） and complexity are lower.

A Neural Network Approach to Blind Estimation of PN Spreading Sequence in DS/SS Signals

In this paper, a new approach is proposed to estimate pseudo noise(PN) sequence in the lower SNR DS/SS signals blindly. This method utilizes the characteristics of self-organization, principal components analysis and extraction of unsupervised neural networks adequately, in addition to its higher-speed operation ability, successfully solve the difficult problem about PN sequence blind estimation. The theoretic analysis and experimental results show that this approach can work very well on lower SNR input signals.

Neural network approach to blindestimation of PN spreading sequence in lower SNR DS/SS signals

An approach based on discrete Karhunen-Loeve transformation of the DS/SS signals is proposed to estimate PN sequence in lower S/N ratio DS/SS signals. Characteristics of self-organization and principle components extraction of unsupervised neural networks are exploited adequately. Theoretical analysis and experimental results are provided to show that this approach can work well on the lower S/N ratio input signals.

Approach to blind estimation of the PN sequence in DS-SS signals with residual carrier

This paper presents an approach of singular value de- composition plus digital phase lock loop to solve the difficult problem of blind pseudo-noise （PN） sequence estimation in low signal to noise ratios （SNR） direct sequence spread spectrum （DS-SS） signals with residual carrier. This approach needs some given parameters, such as the period and code rate of PN sequence. The received signal is firstly sampled and divided into non-overlapping signal vectors according to a temporal window, whose duration is two periods of PN sequence. An autocorrelation matrix is then computed and accumulated by those signal vectors one by one. The PN sequence with residual carrier can be estimated by the principal eigenvector of the autocorrelation matrix. Further more, a digital phase lock loop is used to process the estimated PN sequence, it estimates and tracks the residual carrier and removes the residual carrier in the end. Theory analysis and computer simulation results show that this approach can effectively realize the PN sequence blind estimation from the input DS-SS signals with residual carrier in lower SNR.

Subspace decomposition-based correlation matrix multiplication

The correlation matrix, which is widely used in eigenvalue decomposition （EVD） or singular value decomposition （SVD）, usually can be denoted by R = E[yiy＇i]. A novel method for constructing the correlation matrix R is proposed. The proposed algorithm can improve the resolving power of the signal eigenvalues and overcomes the shortcomings of the traditional subspace methods, which cannot be applied to low SNR. Then the proposed method is applied to the direct sequence spread spectrum （DSSS） signal＇s signature sequence estimation. The performance of the proposed algorithm is analyzed, and some illustrative simulation results are presented.