The Key Technology Of ADS-B Data Organization And Analysis Based On Hadoop

With the continuous expansion of airports and the progressive increase in the demand for flights,the safety of airspace traffic is facing a huge potential risk,which will have higher requirements for airspace safety surveillance systems.Traditional airspace surveillance technologies include radio detection and radar systems,but their high cost?difficulty in maintenance and low efficiency of broadcast have prompted the booming development of Broadcast Automated Dependent Surveillance System(ADS-B)technology.ADS-B is a new type of surveillance system that has more prominent advantages in broadcast information accuracy,integrity,continuity,availability,reliability,and adaptability and scalability.However,in the context of the era of big data,the single-machine environment can no longer meet the storage,analysis,and monitoring of massive flight data.Therefore,analyzing,storing,and analyzing ADS-B data under the big data platform is an important issue for improving flight safety and improving air traffic management efficiency.First,according to the Aviation Radio Technical Committee(RCAT)standard,the ADS-B data received by the school's own ADS-B receiver is used to parse the original data block into meaningful information,then,storing the parsed data in the warehouse which based Hive,and classifying the message through the index table of Mysql and the bucket in Hive,which effectively improve the efficiency of data parsing and avoid the problem of low query efficiency caused by incomplete index in Hive.Based on this,it uses the efficient distributed programming and operation framework provided by the Map Reduce model to further optimize the analysis.Secondly,on the basis of the above research,aiming at the problem of low efficiency in anomaly detection and analysis of massive track data,an IBAT(Isolation Based Anomalous Trajectory)parallel abnormal track detection model based on Hadoop is constructed.The IBAT algorithm is improved based on the the isolated forest algorithm.The algorithm maps the trajectory data to the map grid,by selecting and isolating the grid cells,and calculating their abnormal values,they can detect the rapid abnormal trajectory.To further address the problem of operational efficiency in the anomaly detection of trajectory information,Implementing Distributed Improvement in Map Reduce Environment can further improve the efficiency of the algorithm.The comparison experiments show that the algorithm has higher efficiency in distributed environment.