Low-cost air pollution wireless sensors are emerging in densely distributed networks that\nprovide more spatial resolution than typical traditional systems for monitoring ambient air quality.\nThis paper presents an air quality measurement system that is composed of a distributed sensor\nnetwork connected to a cloud system forming a wireless sensor network (WSN). Sensor nodes are\nbased on low-power ZigBee motes, and transmit field measurement data to the cloud through a\ngateway. An optimized cloud computing system has been implemented to store, monitor, process,\nand visualize the data received from the sensor network. Data processing and analysis is performed\nin the cloud by applying artificial intelligence techniques to optimize the detection of compounds and\ncontaminants. This proposed system is a low-cost, low-size, and low-power consumption method that\ncan greatly enhance the efficiency of air quality measurements, since a great number of nodes could\nbe deployed and provide relevant information for air quality distribution in different areas. Finally,\na laboratory case study demonstrates the applicability of the proposed system for the detection of\nsome common volatile organic compounds, including: benzene, toluene, ethylbenzene, and xylene.\nPrincipal component analysis, a multilayer perceptron with backpropagation learning algorithm, and\nsupport vector machine have been applied for data processing. The results obtained suggest good\nperformance in discriminating and quantifying the concentration of the volatile organic compounds.
Loading....