Multisensor information fusion (MIF) has received a widespread attention and has been applied to a lot of research fields in recent years. Its theory and methods have become an important research area of intelligent information processing. In this paper，the basic principles of MIF are discussed. The system structures of MIF are studied from the viewpoints of information manifestation levels and function differentiation respectively. Then the concrete methods adopted in MIF and their applications are introduced and analysed hierarchically. Finally the existing crucial problems are pointed out and the research trends are also emphasized.