Detection performance can be improved by using distributed detection systems. In these systems, each detector sends either a binary decision or condensed information on the observations available at the detector level to the fusion center. Then, a global decision about the presence of a target is made using binary fusion rules.In this thesis, we have considered the problem of data fusionin the context of adaptive CFAR detection (CMAP-CFAR). In this context, we studied three different types of data fusion, such as binary integration, binary fusion and fuzzy fusion.