In order to conduct the gait recognition system, a wireless in-shoe wearable plantar pressure acquisition system based on ATmega16 and 8 FSR sensors was applied to data acquisition for the gaits which consist of standing, walking, jumping and going upstairs. And four volunteers (2 females and 2 males) were invited in this research to collect the pressure information. MATLAB and LIBSVM were applied to conduct all algorithms proposed by this study. Genetic Algorithm (GA) was used to set the best tuning (penalty) parameter and the best (gamma) of Gauss radial basis kernel (RBF) for C-support vector classification (C-SVC) model and the GA-based C-SVC was obtained. A dataset named ‘train-data’, containing 800 sets of pressure data was used to train the GA-based C-SVC as the algorithm of gait recognition. Finally a testing dataset containing 400 sets of pressure data was applied to test the algorithm of gait recognition called GA-based C-SVC. The accuracy of this GA-based C-SVC was 98% for standing, 91% for walking, 82% for going-upstairs and 97% for jumping. In generally speaking, a better GA-based C-SVC was obtained in this research.