Clusters of low-cost, low-power atmospheric gas sensors were built into a sensor instrument to monitor NO₂ and O₃ in Beijing, alongside reference instruments, aiming to improve the reliability of sensor measurements. Clustering identical sensors and using the median sensor signal was used to minimize drift over short and medium timescales. Three different machine learning techniques were used for all the sensor data in an attempt to correct for cross-interferences, which worked to some degree.