In this study, 315 real-time transmission stations in Fujian province were accessed in the active air-gun source detection experiment. The noise levels of the network were estimated in three bands:1-10 Hz, 0.1-1 Hz, and 10-60 s. First, the background noise records during 240 h were counted to obtain the MODE line of the noise level at each station. Then, the area ratio of the global high-noise model line and low-noise model line to the MODE line was used to quantify the noise level of the network. Based on the different color standards, we divided the noise levels of the network into 10 levels for estimation, and thereby got the high-quality stations. The influence of background noise on the explosive effect of the air gun was further studied. It is confirmed that the receiving capability of the station is closely related to the background noise. We analyzed the environmental noise levels of different station bases and concluded that the environmental noise levels from low to high are fixed base, encrypted base, mobile base, and undersea base. The detection ability of the air-gun source signals would be efficiently improved by the noise assessment of the network, which would also provide an important reference for the maintenance of high-quality stations.