Detection of chi-square fluctuating targets in arbitrary clutter

In a recent paper, general expressions were derived for the density and cumulative probability functions of the amplitude of a linear matched-filter output given a nonfluctuating target in a clutter-limited environment. These expressions were based on the clutter amplitude density function. The results are extended to calculate the cumulative probability function of the output of a linear matched filter used to detect a chi-square fluctuating target in a clutter-limited environment. The resulting method is applied to a common radar clutter model, and experimental sonar data.     

Cross-correlation properties of algebraically constructed Costasarrays

The problem of determining the cross-correlation properties of signals based on algebraically constructed Costas arrays is addressed by examining the discrete cross-correlation of the algebraically constructed Costas arrays for a given construction and dimension. Finding two arrays that minimally correlate implies that the signals based on these arrays also minimally correlate. The properties of finite fields are reviewed, and the major algebraic constructions for Costas arrays are presented, i.e. the Welch construction and the Golomb construction. The discrete cross-correlation properties of the Costas arrays are derived for arrays of the same dimension derived from the same construction. The use of Costas arrays in the signal design problem is discussed, and examples are given to show the cross-correlation of the signals based on the algebraically constructed arrays     

Estimating matched filter amplitude probability functions byfailure rate analysis

Performance analysis of echolocation systems requires knowledge of the probability density function (pdf) or cumulative distribution function (cdf) of a matched filter output. A method is presented to estimate these and other probability functions from data by estimating the failure rate function, a function employed in reliability theory. The method can also be used to derive approximations to closed-form probability functions. The method is demonstrated using experimental sonar and radar clutter data and a closed-form radar clutter model     

Pade approximations to matched filter amplitude probabilityfunctions

Performance analysis of echolocation systems requires the probability density function (pdf) or survival function of a matched filter output. A method is presented to derive approximations to these functions using a Pade approximation to their associated characteristic function (CF). The method is based on the Hankel transform. It allows computation of detection probabilities when the matched filter responses to clutter and a target are separately known. Several numerical examples are presented     

Homodyned-K fluctuation model

The statistics of the matched filter output are examined for the case of a nonfluctuating target in K-type clutter. The resulting "homodyned-K" backscatter model is a generalization of the classic Rayleigh-Rice model and has applications in radar, sonar, optics, and ultrasonic medical imaging. This paper presents exact and asymptotic equations for the probability density function, survival function, moments, and parameter estimation. The exact equations are expressed as infinite series. Also provided are proof of the series' convergence and guidance for their numerical computation. Examples illustrating the model's application to radar and sonar are presented using electromagnetic and acoustic backscatter data