Research On Index Methods Of Waterborne Big Data With Location Information

Development of information technology, especially Internet of Things, promotes the popularity of various positioning technologies. Over time, the type and amount of each device location data grown rapidly.In the shipping system, the ECDIS can not meet the needs of the new data processing applications. The Big Data provides some new methods for processing and analyzing the massive and multi-configuration information. Spatial data index can provide high-speed data storage, query and retrieval for large amounts of data shipping information. This paper introduces spatial data indexing technology, research on high-performance data indexing technology in the field of water location services.According to the characteristics of water information data, we propose a new indexing algorithm, called thread quadtree, and applied to the prototype system of water location services. Add a clue chain to each node for the quadtree; this chain indicates the objects of all child nodes. If all child nodes of a query node are found in the query results, copy the objects directly from its chain, avoiding further recursive queries. Comparison of space area subdivided, the results of the comparison operation is divided into separate, contain and intersect. If a node is completely contained by query area, then copy the objects directly from the chain. This method is effective in reducing the count of comparisons and iterations, thereby improves the efficiency of query operations. In applications, add function to node, this function returns all data of child nodes. This can reduce the space cost of thread quadtree.Proposed quadtree coding algorithm and the algorithm is applied to the prototype system. This improves the efficiency of the water loaction data query. Quadtree coding algorithm is based on the idea of space division, the overall space is divided into four sub-space method according quadtree. Each subspace then divided in the same manner, formed a month level divided subspace. These subspaces are encoded in a specific order. According to the rules quadtree, designed an algorithm. This algorithm can quickly calculate the code for parent or child of a node. In the query data, this algorithm can directly find the corresponding node by calculating, avoiding the quadtree traversal operation. Each query operations can directly find the quad-tree node, using the fast algorithm. Time cost of the algorithm is fixed; it does not significantly increase when data size increases.This paper presents a new data index, the cache quadtree. This cache quadtree can be used for fast data index in shipping information services. It can also be used to cluster large data. The cache quadtree establish an association mirror in the mobile data terminal, this mirror corresponds to a node in server. First, the query operation is carried out in the mirror of the mobile terminal; the results of mirror query are the objects in the mirror and the regions what does not exist in the mirror. The regions are sent to the server, and the server executes the query operation. The results from the server and objects found in mirror are merged; this is the result of the query operation. The nodes in server and terminal are coded, using the same method. The same code ensures the data transmission between server and terminal. Query operations were executed in the terminal and the server-side separately, this effectively reduces the amount of query operations in the server.