Multi-data source position information fusion method based on Spark cluster and system

Abstract

The invention provides a multi-data source position information fusion method based on a Spark cluster. The method comprises the following steps that mass position information data is received through an interface, wherein the position information data is from a signaling collection system of provincial-level operators and an Internet user behavior big data platform; collected text files are encrypted and cleaned through a preprocessing program to remove invalid data, and the processed data is put into a Kafka in real time; the real-time data is processed through Spark Streaming, the processing comprises real-time position calculation, fusion, counting and trajectory generation operations, total-amount position information is obtained, and the processed total-amount position information is stored in a Redis database; the total-amount position information is sliced and exported according to a set cycle in the Redis database, the exported data is stored in an HDFS and supplied to an upper-layer application for calling and querying historical data through the Spark SQL interface. The invention further provides a system of the multi-data source position information fusion method based on the Spark cluster.

Description

technical field [0001] The invention belongs to the technical field of big data processing, and in particular relates to a location information fusion method and system of multiple data sources based on a Spark cluster. Background technique [0002] With the promotion of location information-based applications in various industries, the application prospects are broad and the needs of various industries are strong. Whether it is the tourism industry, government agencies such as transportation, civil affairs, statistics bureaus, or retail industries, there are very urgent needs. Big data applications provide very good landing scenarios. [0003] The location information is based on the user's telecommunications 2, 3, 4G network signaling location, wireless side bill data, WIFI Internet access records, GPS data extracted from user Internet behavior records, and user-related attribute data. Provide specific area population density monitoring, crowd attribution analysis, crowd ...

AI Technical Summary

Problems solved by technology

Other technologies cannot achieve real-time or quasi-real-time processing in terms of timeliness, but in some scenarios, such as algorithmic trading, fraud detection, real-time processing, etc., it is necessary to discover potential patterns in a large amount of information in a timely manner. , it is necessary to use the stream processing technology - Stream

Images