In response to the data quality assessment issues faced by the integration of seismometer, strong motion seismograph, and intensity meters, research and analysis work on background noise of regional early warning networks has been carried out. Using the noise power spectrum method, maximum probability acceleration noise peak (PGA), velocity noise peak (PGV), and displacement noise peak (PGD) standards, scientifically evaluate the background noise of the three types of sensors that make up the network, and analyze the differences in their noise levels; Quantitatively evaluate the noise level of stations using the noise power spectral area ratio method, and display the noise level of each station based on different color codes, making it convenient to visually judge the operation status of the station; Based on actual statistics, select appropriate proportions to determine the high and low baseline of various sensor noise models and the upper and lower limits of normal noise PGA, PGV, and PGD, and identify stations with suspected abnormal waveform records. The research results indicate that among the three types of seismic monitoring instruments, seismometers can fully and effectively record environmental noise in the full frequency band, with the lowest noise level; The strong seismic instrument can record environmental noise above 0.1 Hz, and the frequency band below 0.1 Hz is mainly due to instrument self noise, with a higher noise level than the seismometer; The recording of the intensity meter is basically the self noise of the instrument, which cannot record the Earth's pulsation and has the highest noise level. The noise power spectral area ratio method divides the noise level into four levels, which can effectively select stations with high-quality records. Detecting abnormal stations through frequency and time domains greatly improves the reliability of detection results, while also facilitating the detection of false alarms such as calibration and abnormal large pulses.The quality evaluation results of the observation data of the network can be used as a reference for maintenance personnel to focus on high-quality stations, ensure the Completeness of the data of high-quality stations, and record abnormal stations that should be repaired in a timely manner, providing an important guarantee for the reliability and accuracy of earthquake early warning and intensity quick report products.