Large-scale public buildings are dense with crowds, and building noise is easily affected by crowds and other dynamic and complex factors. As the numerical simulation of building noise pollution is time-varying, complex, and nonlinear, a numerical simulation method of assessing building noise pollution based on adaptive transfer function is proposed. A building noise pollution monitoring system based on the Internet of Things is built to monitor the environmental elements of building noise in real time, analyze the frequency spectrum of building noise pollution sources by adaptive functions, and use the effective value of acceleration, vibration acceleration level and vibration level to measure building noise pollution according to urban environmental noise standards. Noise pollution intensity and level are determined to complete the building noise pollution numerical simulation calculation. The effectiveness of the proposed method is verified by experiments. The results show that the simulation calculation method designed this time is more accurate than the traditional method and is closer to the actual measured value. It has higher applicability and provides reference for numerical simulation of building noise pollution.