Abstract

Discriminant analysis is a technique used in statistics and machine learning to separate two or more classes of objects or events. We introduce linear, quadratic, and mixture discriminant analysis methods into radar signal classification. However, the selection of an appropriate discriminant analysis method can be difficult and no comparison study of these discriminant analyses for radar signal classification can be found in the open literature. This article presents a theoretical analysis and practical comparison of these three discriminant analysis methods for radar signal classification. The theoretical analysis is derived from Bayes' theorem. The practical comparison study is performed on a dataset consisting of five radars emissions. The advantages and drawbacks of each discriminant analysis method are highlighted. This study demonstrates that quadratic discriminant analysis (QDA) is predominantly a better method for radar signal classification using our radar dataset. On average, it demonstrates 95%, 93%, and 86.6% classification accuracy for three, four, and five radar emitters in our radar dataset, respectively. Linear discriminant analysis (LDA) achieves on average 88.7%, 84.9%, and 79.2% classification accuracy for three, four, and five radar emitters, respectively. Mixture discriminant analysis (MDA) also achieves the same classification performance as QDA. Theoretical analysis shows that both LDA and QDA are a special case of MDA and that MDA can set up more decision boundaries than LDA and QDA if the feature distribution in the dataset is an ensemble of Gaussian distributions. Therefore, MDA is recommended for radar signal classification.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.