A Improved Method Based on Enhanced Diagonal Secant Updating Frequency Domain Fourth-Order Cumulant Scheme for DSSS Signals

In this paper, an asymptotic estimating technique for detection of direct sequence spread spectrum (DSSS) signals using modified frequency domain fourth-order cumulant is proposed .The general higher-order statistics that recently are used for detection of DSSS signals may not be easily applied to signal processing because of too complex computation and huge processing time. In this paper, based on 2-D slices fourth-order cumulant, an efficient algorithm is proposed to solve the problem of complex computation and huge processing time and extended for non-stationary stochastic processes. This paper implies an asymptotic estimating technique with a title of enhanced diagonal secant updating scheme based on the steps of backtracking improved via Wolfe-like condition in the Armijo-type line search, for solving nonlinear equation of frequency domain 2-D slices fourth-order comulant method. This method only requires to store a row vector while ignoring all the low and diagonal elements and therefore the required memory is reduced strongly. In addition, as it uses two line search strategies (predictor and corrector) to obtain a new iterates point, the spectral properties of the diagonal updating scheme is improved, and rapid convergence property is gained. This method has a very good solving speed and a better performance among the Newton-like methods. Simulation results and computation analysis indicate that the processing time of the modified frequency domain fourth-order cumulant method is better than the commonly used frequency domain fourth-order cumulant methods. The validity of this method has shown by the computer simulation especially in low signal-to-noise ratio (SNR) conditions.

کلیدواژه ها:

Chip Rate ، Correlation Oscillation ، DSSS Signal ، Fourth-Order Cumulant ، Radiometer