Topic
K-distribution
About: K-distribution is a research topic. Over the lifetime, 1281 publications have been published within this topic receiving 51774 citations.
Papers published on a yearly basis
Papers
More filters
01 Jan 2012
TL;DR: In this paper, the computable steps of sequential number-theoretic method for optimization (SNTO) were proposed to get the MLE of the parameters of K-distribution.
Abstract: The K-distribution is widely applied in synthetic aperture radar (SAR) image processing. However, the multi-peak complicated likelihood function causes much trouble to obtain the maximum likelihood estimation of K-distribution parameters. Based on the number-theoretic net (NT-net), the computable steps of sequential number-theoretic method for optimization (SNTO) were proposed to get the MLE of the parameters of K-distribution. Comparing with the non-ML estimator Y0.1, we do Monte Carlo trails with different values of shape parameter and different sample sizes. The simulation results show that the proposed method outperforms the fractional moment based technique.
14 Jul 2021
TL;DR: In this paper, the effect of frequency form on the reverberation envelope statistical characteristic is analyzed, and the parameters of different models are estimated by the method of moments (MOM) to perform the theory simulation and curve fitting, and then the correlation between data processing and model simulation results are calculated.
Abstract: Reverberation is the main background interference of active detection and identification, which is produced with active signal emission. It is traditionally assumed as a combination signal of a large amount of random scatters at the receiving hydrophone which leads to a Gaussian-distributed time-domain signal and a Rayleigh-distributed envelope following the central limit theorem. However, with the advent of high-resolution active sonar systems, the target-like scatter signals namely clutter arising from fluctuation or variety in seafloor or hydrographic would produce the non-Rayleigh reverberation with longer tail and it is a main factor restricting the active sonar long-range detection. Aiming at this phenomenon, the Rayleigh distribution model and two typical non-Rayleigh statistical models-the Weibull distribution and the K distribution model are studied using the low-frequency reverberation test data (signal center frequency at 420Hz) obtained from a typical shallow-water environment in the northern South China Sea. Based on the measured multi-sample reverberation data, the probability density distributions of different signal forms reverberation are calculated respectively. The parameters of different models are estimated by the method of moments (MOM) to perform the theory simulation and curve fitting, and then the correlation between data processing and model simulation results are calculated. Comparative analysis based on the experiments shows that the pulse length of transmitting signal has an important effect on the statistical characteristics of the reverberation in shallow water, which is reflected in the fact that the envelope distribution of short pulse-width reverberation presents stronger non-Rayleigh properties compared with long pulse-width reverberation. In addition, the long-range reverberation present stronger non-Rayleigh properties fitting with the non-Rayleigh models well relative to the short-range reverberation. The effect of frequency form on the reverberation envelope statistical characteristic is relatively insignificant.
01 Jan 1994
TL;DR: In this article, an insurance policy where the claim amount has a mixed lognormal distribution was considered and the individual distribution of claim amount, given the value of the parameter for the median of that distribution, was assumed to follow a normal, Laplace, uniform, power function and gamma distribution.
Abstract: Consider an insurance policy where the claim amount has a mixed lognormal distribution. In this paper, we assume that the individual distribution of claim amount, given the value of the parameter for the median of that distribution, follows a lognormal distribution. To model the heterogeneity in the population, we assume that the median is disbursed according to normal, Laplace, uniform, power function and gamma distributions. The exact ultimate forms for the probability density function of the portfolio claim amount are given for the former three distributions. Whereas, a difference equation representation is given for the latter two distributions.