Research On The Method Of Rapid Map Display In Embedded GIS System

The real-time map display in embedded GIS system is a research hotspot recent year. However, the contradiction of the gigantic map data and the limited hardware resource makes the real-time map display a intractable problem in this domain. Data model, multithreads technology, caching technology and other technologies are researched by scientists at home and abroad, and a great progress is achieved. Since the choosing of the processor is restricted and the calculate ability of the processor is limited in the military application, in the map display need of big-screen, high-precision, multi-type, multi-layer, multi-factor, the prior technology can not meet the militaryapplication.Considering the problem of message blocking and display delay caused by successive map operations in the project of a navigation device, this thesis focuses on the researches of the acceleration method for the map display. The results show that this method eliminates the delay effectively, and meets the navigation device application. The main work and contributions are as follows:1. A map display acceleration scheduler model of sequential message drive mechanism is proposed. Based on the message drive mechanism and the blocked storage mechanism, a map display method of sequential message drive mechanism is proposed. This method decomposes a time-consuming coarse-grained display process, which processes the map data of the whole screen, into a series of less time-consuming, small-grained sub-processes, and one single data block is managed in each sub-process. And the decomposed sub-processes are translated into a series of messages which are sent to FIFO message sequence built by display task. The corresponding data scheduler and cache method are also studied. And the message sequence is maintained dynamically. The controllability and the flexibility are enhanced, and the delay caused by the coarse-grained display process is reduced, and the display speed is accelerated.2. The data scheduler method based on the message priority and the Greedy policy is proposed. Firstly, in order to eliminate the map zooming delay, a message priority based data scheduler is proposed. The method assigns the priority for the corresponding messages according to the process sequence dynamically, and deletes the no-needed disposal sub-process messages, and abandons the unnecessary data processing so that the data needed to be handled and the display delay are reduced. The test results show that the new method reduces the display time effectively compared with the traditional method. Then a data scheduler method based on the Greedy strategy is proposed to alleviate the roaming delay. This method predicts the roaming direction and previously assigns the map data next processing to the void of present operations, so that the smooth handoff of the in the front and background cache is ensured. The test results show that the new method improves the smoothness of the map roaming effectively compared with the double buffer method.3. The map data caching model based on hashing is studied. Firstly, considering data prefetching, cache replacement, according to the advantage of hash table in the decrease data storage and search time, a map data caching model based on hashing is structured. Next, in order to solve the problem of reading map data Wastes memory in advance according to sheet, the grid index memory is huge, and cache fragment reduces the utilization rate of the memory, the method of building index ,the cache strategy according to needs of map display and the cache management of the map data are proposed. In the method above, four parameters which represent the map data block is mapping to a parameter and the memory of the single index decreases, reading map data block to memory by need according to the memory and the map operating enlarges the scale of the useful data caching, cleaning up cache fragment in the system void enhances the utilization rate of the memory. The test results show that it can enlarges the scale of the useful data caching in limited memory space, reduces the times of accessing low speed device, and accelerates map display with no slower map data accessing in the methods mentioned above.