Abstract: ‍ ‍Aiming at the maximum side lobe level suppression and coverage performance optimization problems of dual-frequency localizer antenna arrays in the instrument landing system， a chaos quantum particle swarm optimization algorithm was improved to optimize the feeding parameters of the LOC antenna arrays pattern with given constraints. Based on the analysis of the formation of dual-frequency LOC coverage， the ICAO coverage requirements for LOC signals were analyzed， and the constraints that the LOC pattern should meet were obtained. In order to improve the global optimization ability and effectively avoid falling into local optimum， the idea of chaos and weighted average optimal position were introduced to improve the CQPSO algorithm， and the improved CQPSO algorithm was used to feed the 20-element equal-spaced LOC antenna arrays under the constraints. The parameters were optimized， and the directivity of the LOC antenna arrays was simulated and analyzed according to the obtained feed parameters. The simulation experiments showed the pattern formed by the feed parameters obtained by the improved CQPSO algorithm， compared with the pattern of the currently widely used 24-element equal-spaced LOC antenna arrays pattern， when the number of antenna was reduced by 16.67%， the maximum side lobe level was 3.32 dB lower with better coverage. Compared with the current widely used 20-element unequal-spaced LOC antenna arrays pattern， the maximum side lobe level of channel signal radiation was reduced by 25.55 dB. The effectiveness of the improved CQPSO algorithm is proved.