CIV: Information Volume of Complex Mass Function

Abstract

Dempster-Shafer evidence theory is an efficient theory for dealing with uncertainty which has been applied in many fields. As an extension of evidence theory, complex evidence theory is effective for handling the uncertainty of complex-valued information. However, the existing uncertainty measurements in complex evidence theory only consider the first-order uncertainty of a complex mass function (CMF) but ignore the higher-order uncertainty. Hence, in this paper, a novel information volume of complex mass function (CIV) is proposed, which is a complex-valued extension of information volume of mass function (MIV). Besides, the real part of CIV is compatible with the MIV and Shannon entropy. The properties of CIV are illustrated by examples. In addition, based on CIV, a complex evidence combination algorithm is proposed and applied in target recognition. The results show that the proposed algorithm can combine CMFs and recognize targets with correctness and high efficiency.