Abstract: For the purpose of coping with the intensive deployment of eNB (evolved NodeB) and AP (access point) in LWA (LTE WLAN aggregation) networks, and a situation in which the user is unable to satisfy their needs, this paper proposes an access point selection algorithm for LWA networks based on optimal weighted multi-attribute decision (OWMAD). First, the algorithm pre-collects the access point attributes and establishes the attribute set, then designs the optimal weights of the decision attributes and overlays the decision matrix with the optimal weight vector. Finally, the optimal network access point is determined according to the results of the comprehensive performance of each access point. Simulation results show that the proposed algorithm can effectively improve the throughput of peripheral areas and the overall system, while reducing the average system delay and improving the user's quality of service.