This paper proposes a set of adaptive environment optimization system based on digital filtering technology, feature extraction technology and as a basis for the optimization of athletes’ sports training, which adopts a multi-level feedback control architecture, with the environment perception layer responsible for data acquisition, the decision analysis layer executing intelligent algorithms, and the executive control layer, realizing intelligent recognition of sports status and quantitative assessment of training effect. The empirical results show that the athletes in the experimental group improved their specialized skills by 12.7% and their physical fitness composite index by 15.3%, which are significantly better than the control group. Physiological monitoring shows that the average heart rate in the optimized environment is reduced by 9.1%, and the recovery time is shortened by 26.2%, indicating that the system can effectively reduce the load and promote the recovery, providing an effective technical solution for the scientific and personalized sports training.
