High-resolution active sonar is developed and widely applied in sonar engineering area recently. The sonar systems could improve signal-to-noise ratio within the resolution cell and increase the detection performance.

However, it may lead to heavier-tailed reverberation envelope distribution contrast to the classical Rayleigh model. Non-Rayleigh statistical model research of backscattered amplitude attracts researchers and engineers working on the active sonar application.

K-distribution model, as one of the non-Rayleigh statistical models was introduced from radar clutter signal processing to sonar reverberation signal processing. It has robust property, especially for the physical interpretation of its parameters. Beyond the detection purpose, K-distribution model might be helpful for seafloor classification and environment detection.

Recently, researchers ZHANG Chi, MA Xiaochuan, ZHANG Shuhao etc. from the Institute of Acoustics of the Chinese Academy of Sciences introduce and declare a series of statistical procedures by using pseudo-random variables of K-distribution. The distribution fitting, parameter estimation of distribution and evaluation of fitting goodness could be made on the basis of these steps. Thus, the accurate and reliable result of fitting and estimating of K-distribution reverberation could be obtained.

The procedures of handling sonar data are illustrated through both the simulated random data and the real underwater sonar data. Meanwhile, the differences and relationship between Rayleigh and K-distribution are declared. K-distribution is introduced from radar processing area to sonar engineering area because the heavy-tailed distribution of reverberation envelope is similar with the one of sea clutter of radar echo.

Supported by an underwater experiment, researchers sample a set of real data by a planar array. The approach is illustrated using the real data, and verified that the reverberation envelope of shallow water fits K-distribution model well.

The research on the reliable target detection in reverberation background is significantly important for the active sonar application. Statistical characterization of reverberation envelope is essential for the constant false alarm rates detection, and reliable data processing methods are the basis of the data analysis.

After the steps presented in this research are performed, the distribution fitting and the parameter estimation could be more accurate and reliable.

Reference:

ZHANG Chi, MA Xiaochuan, ZHANG Shuhao, ZHAN Fei. K-distribution Reverberation and Reliable Statistical Approach Study of Sonar Data. OCEANS 2016- Shanghai. DOI: 10.1109/OCEANSAP.2016.7485424

Contact：

ZHANG Chi

Institute of Acoustics, Chinese Academy of Sciences, 100190 Beijing, China

E-mail: zc@mail.ioa.ac.cn