Abstract: Aiming at the performance limitations of the cell-averaged constant false alarm detector (CA-CFAR) in the presence of multi-target interference and clutter fringe effects, a constant false alarm detector (MCA-CFAR) based on the minimum selection cell in the sub-reference window is proposed. The performance of the detector is significantly improved by selecting the smallest cell in the sub-reference cell, and the detection probability, false alarm rate and detection threshold are derived in detail in the context of Rayleigh distribution. To further optimize the performance, a fuzzy logic fusion detector (FUMCA-CFAR) is designed, which uses two sensors to compute the spatial affiliation function values and fuses them by four rules, namely, algebraic sum, algebraic product, MAX, and MIN, to achieve a smooth output and reduce the loss of target information. Simulation experiments show that the FUMCA-CFAR detector based on algebraic sum fusion exhibits excellent detection performance and anti-jamming ability in both uniform and non-uniform backgrounds.