Abstract: In order to suppress the parasitic current on the outer conductor of electrically coaxial cables, a structure for parasitic current reduction is designed and researched based on adaptive genetic algorithm. The structure consists of a metallic sheet which is electrically connected to the out conductor of coaxial cable, with five pairs of symmetrical I-shaped slots embedded. Lumped capacitors are loaded to miniaturize the entire size. The operation of the entire structure is similar to a series of parallel LC resonators. When the parasitic current flows on the outer conductor of the coaxial cable, the equivalent parallel LC circuit which exhibits a high impedance path can suppress the current efficiently. An adaptive genetic algorithm is presented which can be used to optimize the values of lumped capacitors, and the crossover probability and mutation probability can adjust adaptively and nonlinearly, the analysis results show its good convergence speed and strong robusticity compared with standard genetic algorithm. By means of optimal calculation of the values of lumped capacitors, numerical results and theoretical analysis are presented and discussed. The results show that the structure can suppress the parasitic current at the 5 250-5 925 MHz band for a -10 dB transmittance. The method is valuable to the design of parasitic current reduction structure, and can also show the quantitative analysis on the current reduction effect.