PERFORMANCE COMPARISONS OF MODIFIED SMI AND GLR ALGORITHMS

When adaptive filtering is used for signal detection in colored interference of unknown statistics, it is often desirable for the system to have a constant false alarm rate (CFAR) feature. The detection performance of a modified version of the well-known sample matrix inversion (SMI) algorithm, called the modified SMI (MSMI), is compared in this paper with that of the generalized likelihood ratio (GLR) algorithm in colored Gaussian interference. The performance sensitivity of the MSMI and GLR in colored Weibull and log-normal interference is studied via simulation. It is found that there is almost no need to use the more complicated GLR algorithm in Gaussian interference, while in Weibull or log-normal interference the GLR should be preferable to the MSMI.