AGGREGATE QUERY METHOD AND SYSTEM FOR TRAFFIC DATA FLOWS

摘要

An aggregate query method and system for traffic data flows belong to the technical field of information processing. The method comprises obtaining spatial-temporal information of a moving object to generate a traffic data flow, partitioning a data space into sub-units, grouping the adjacent sub-units with similar frequencies into a small number of buckets, calculating Kalman gains of the buckets on the basis of the frequency of the buckets, indexing the buckets to form a BPT (binary partition tree) index of a current timestamp by means of a BPT, and forming a historic index by serializing the BPT after the current timestamp is over; performing an aggregate query, and when the frequency of the buckets changes too much, the aggregate query value is replaced by the optimal estimated value of the frequency of the buckets. The system comprises: an information collecting module, a data processing module, an index processing module, an application service module and an index storage module. The method enables effective suppression of the maximum relative error of outliers during traffic data flow query, thereby guaranteeing the availability of the aggregate query method.