Belief Function Fusion based Self-calibration for Non-dispersive Infrared Gas Sensor

Non-dispersive infrared gas sensing is one of the best gas measurement method for air quality monitoring. However, sensors drift over time due to sensor aging and environmental factors, which makes regular calibration necessary. In this paper, we first propose a general belief function fusion framework for NDIR gas sensor calibration, where we focus on getting a reasonable fused belief function of the true CO
₂
level. To deal with belief functions highly conflict that may highly conflict with each other, we further propose a modified weighted average approach which utilizes the Wasserstein distance as a measure of the similarity between the belief functions. The numerical experiments show excellent initial results which confirms the belief function fusion framework for NDIR gas sensor is possible.