Abstract: In order to solve the problems of insufficient objective evaluation and lack of evaluation basis for the classification and identification of radar emitter signals, a group evaluation method based on interval fuzzy principle, fuzzy cross entropy and multi-criteria compromise method is proposed. Firstly, the signal-noise ratio (SNR) classification evaluation model is established by the interval fuzzy principle, and the SNR weight is obtained by optimization based on the Hamming distance. Secondly, the hierarchical model is integrated with the group decision matrix by combining SNR weight and interval intuitionistic fuzzy weighted average operator, and the attribute weight is calculated by the entropy maximizing method. Finally, based on the framework of multi-criterion compromise method, the fuzzy cross entropy is adopted to realize feature scheme sequencing. The simulation results show that the proposed method is able to give the sorting identification feature evaluation ranking results consistent with the actual simulation experiment, and is better than the ideal point approach method, which verifies the feasibility and effectiveness of the proposed method.