Traffic Emission Discovery and Control using Air Quality Indexing and Analytics Framework

1S.Kavitha Bharathi, N.Navee, Akhil reddy thatikonda


The air quality is very essential for human health and urban governance. The air quality monitoring systems are deployed to monitor the air quality levels of the regions. The pressure, temperature, humidity and rainfall based physical and chemical reactions affect the air quality levels. The air quality monitoring applications are established with spatial and temporal features. Spatial databases are used to maintain the explicit location and neighborhood details about the spatial objects. Spatial mining techniques are applied to discover the spatial patterns on the spatial objects. Temporal data refers the sequence of events with time stamp details. The objective of worldly information mining is to find shrouded relations among arrangements and sub-successions of occasions.Air quality monitoring systems are build to measure air mass distribution to solve the air pollution problems. The air quality systems are restructured as complex networks with reference to the location and time parameters. The complex network model is initiated to analyze the relationship between the regions. Time correlation matrix is constructed with time correlation of air quality nodes. Spatial homogeneity and heterogeneity are characterized with the spatial distance and wind parameters. The spatial interaction intensity is measured with the spatial correlation matrix. The spatial and temporal correlation information are used to construct the Air Quality Spatial Temporal Network (AQSTN) model. The community detecting methods are applied on the AQSTN with local similarity and region interactions.The Air Quality Indexing and Analytics (AQIA) framework is constructed to discover the traffic emissions. The spatial and temporal feature based window selection mechanism is used to discover time oriented changes in the air quality levels. The air pollution detection process is performed with emission level ranking model with K-Nearest Neighbor (KNN) classification method. The station and region based air quality and risk level event discovery process is connected with AQSTN.


Air quality systems, Traffic emission discovery, Pollution monitoring, Air Quality Spatial Temporal Networks and K-Nearest Neighbor classification.

Paper Details
IssueIssue 8