Xu Lianming

Research on a quasi-lossless compression algorithm based on huffman coding

A quasi-lossless compression algorithm for location information within civilian global positioning system (GPS) was proposed based on the analysis and comparison of performance between Huffman coding and arithmetic coding thus improving location data compression ratio and compression speed within civilian GPS accuracy. In the algorithm, predictive coding and Huffman coding were organically combined. Furthermore, this algorithm removed redundant information through compression preprocessing and secondary quantization. The coding efficiency improved with the use of predictive coding. Tests with an MSP430 microcontroller showed that compression ratio of the proposed algorithm was 87.1% and processing time is 31.4s when compression data amount reached 668kB, which coincided well with simulation results. Experimental results indicated that the proposed algorithm had low requirement for hardware resources after optimization and resulted in improved compression ratio and fast coding speed, which saved storage resources and communication cost for data transmission.

Recognition and localization of boundary and isolated nodes in wireless sensor networks

There are boundary nodes and a certain number of the isolated nodes in the wireless sensor network. If conducting the effective judgment and processing over them, the positioning errors of the entire network will be reduced. Based on principles and approaches of the graphic theory, the boundary nodes and isolated nodes will be judged. Through an analysis on the node density and the range and direction of the receiving anchor nodes, the judgment and positioning approaches for the boundary nodes and the isolated nodes in the wireless sensor network can then be come up with. The simulation experiment for positioning performance in different situation suggested that the positioning accuracy of the misbehavior nodes will be improved by over 17% after the modification while the one of the isolated nodes in the irregular network topology after modification will be improved by over 10%.