Traffic is the dominant noise source in urban areas, this has a direct impact on the health and productivity of the population. So far, the noise is difficult to control, either with active noise control or passive noise control. This is limited by the fluctuating characteristics of the noise signal and its varying frequency content. This study aims to analyze the characteristics of traffic noise signals based on spectral, temporal, and spatial patterns. Through the characteristics of the traffic noise signal, it can then be determined a control strategy that can be carried out using an urban soundscape perspective approach. Traffic noise signals were obtained through measurements at five locations on Jalan Soekarno-Hatta Surabaya in four different time zones. Analysis of traffic noise signal characteristics with patterns is carried out using a soundscape theoretical framework approach. The measurement results show that the Leq ranges from 61-82 dBA and the Dynamic Range is 17.6-34.8. Based on the spectral pattern, high noise levels occur at 20 Hz < PSD < 100 Hz. Meanwhile, based on the temporal pattern, the noise level in the four-time zones is relatively the same and comes from the noise of motorized vehicles and heavy vehicles. Residential sites are the noisiest locations because of the density that occurs on two highways, crowds in restaurants, and the sound of information media in traffic. The traffic noise control strategy is based on spectral mapping by combining sonic crystal windows as passive noise controllers and sound masking as active noise controllers.