Abstract The paper proposes a statistical classification method devoted to multisensor embedded systems. It is based on estimation of multidimensional probability density function (pdf) which describes statistical properties of multisensor… Click to show full abstract
Abstract The paper proposes a statistical classification method devoted to multisensor embedded systems. It is based on estimation of multidimensional probability density function (pdf) which describes statistical properties of multisensor measurement data. Various methods of pdf estimation are compared. The performance of the algorithm was checked through implementation on digital signal processor. The results of the tests confirmed high efficiency of the proposed method and its low numerical load. The paper also presents an example of application to breast cancer diagnosis.
               
Click one of the above tabs to view related content.