Determination of Angle of Arrival using Nonlinear Support Vector Machine Regressors

Abstract

A combination of optimization theory, statistical learning, and kernel theory labeled as "support vector machines" (SVMs) can be applied to electromagnetic problems. Recently, popular machine learning algorithms have successfully been applied to wireless communication problems, notably spread spectrum receiver design, channel equalization, and adaptive beam forming with direction of arrival estimation (DOA). The capacity of communication systems has limitations due to co channel interference. In code division multiple access (CDMA), users can share the same frequency at the same time, but the number of users is limited by the multi-user interference (MUI). This paper presents an implementation of determination of angle of arrival (AOA) estimation based on nonlinear SVM regressors (SVR), an important component of CDMA communication systems