| Petroleum is the industrial blood of a country and it is related to the internationalpeople’s livelihood. The rapid development of modern industry is in need of great amountof petroleum provision, but the requirement of nonrenewable resource and resourceexhaustion become into the principal contradiction that hinders the economical and socialdevelopment of countries in the whole world. Up to now, the biggest difficulty forgeophysics industry is to explore in a complicated geological condition which compoundsthe difficulty of exploring. As the boom of techniques of petroleum seismic exploration,seismic exploration technique develops towards to characteristics of high density,multiplicity, full wave field, high-resolution and multi tracks, which all require thecoordination of equipments of geophysics exploration. So far, the major feature ofpetroleum exploration equipments is the internationalization of seismic sensor, and thecommunications technology in seismic exploration equipments is one of the keytechniques of petroleum seismic exploration equipment networks and it directly decidesand constrains the function of the whole equipment system.Wired seismic data transmission is not suitable for complicated terrain for its highcost of cable, huge amount of work, high cost of equipment sustainment. Wireless seismicdata transmission doesn’t need transmission cable so that it can be deployed in variousfield and abominable terrains, and it is not limited by number of tracks, so it is suitable forpetroleum seismic exploration of large-scale complicated terrain. But the traditionalwireless data transmission of petroleum seismic exploration equipment generally adoptsthe radio-frequency communication technique without the support of networks technique.So it is not the real wireless networks transmission with its transmission node limited andtransmission control difficult.Wireless sensor networks is a new generation sensor networks technique, which is acomprehensive information process network that combines the sensor techniques, microelectronic mechanical system techniques, modern networks and wireless communicationtechniques. It is one of the hottest research topics in computer information area and iswidely used in public security, military,intelligent transportation and ecological environmental protection and more industrial and civilian-use areas. Wireless sensornetworks has such advantages as strong environmental adaptability, long network lifecircle, flexible cabling and convenient sustainment and low investment of equipment costso that it can be applied to wireless data transmission of petroleum seismic explorationequipment networks. But some technical problems, such as comparatively complicatedexistent protocol, limited embedded-node internet resource and certain error rate betweennodes, constrain the data transmission of wireless sensor networks in seismic explorationequipment networks.Hyperspectral imaging technology develops based on the multispectral imagingtechnology. As the spectral resolution of hyperspectral image improves, the hyperspectralimage data becomes the most important resource of remote sensing data. Hyperspectralimage can obtain more serial topographic spectral features and hidden topographicfeatures in narrow spectrum. That is why remote sensing technology of hyperspectral image becomes one of the hottest research directions. Comparing to multispectral image,hyperspectral image possesses not only higher spectral resolution and spatial resolutionbut also wider wave spectrum, which provide more reference data for terrain research.Hyperspectral remote sensing image can be regarded as3-D image data, and hascharacteristics of rich texture, multiband, high spectral resolution so it can be appliedwider to Geophysics exploration and other areas. But the excessive amount data ofhyperspectral image and comparatively high data dimension are the biggest problems tothe current communication systematic transmission and storage. Methods of traditionalhyperspectral image coding, mainly adopting static image compression method, aremostly used in the compression of single hyperspectral image data. It deducts the spatialredundancy of single band but doesn’t completely exploit relativity between thehyperspectral image spectrums, with limited compression function and low signal-to-noiseratio. The above questions are studied in this paper:(1) According to the data transmission requirement of seismic explorationequipment networks and characteristics of Wireless Sensor Networks, an all-wirelessseismic exploration networks data transmission proposal based on wireless networks isprovided in this paper; WSN networks is used in a sub-area of communication, andfrozen-in distribution and random distribution topological structure are adopted; WLAN(Wireless Local Area Networks) is used in crossed station to guarantee the requirement ofbandwidth of the Networks data transmission. The idea of random deployment of wirelesssensor directly solves the difficulty of arranging cables in bad or complicated terrain. Andat the same time the self-organizing and self-curing characters of WSN networks are madegood use of to ensure the reliability of seismic exploration data transmission.(2) To improve the functions of WSN networks, I did research in the coverage,connection and topology control of wireless sensor networks and proved therandom-distributed mathematical model of wireless sensor networks node; I mapped thebi-dimensional wireless sensor networks coverage and connection into questions ofpercolation theory of state mutation; by analyzing percolation mathematics theory, I set upand proved wireless sensor networks coverage and connection percolation theory modeland ascertained the mathematic expression relationship between the critical-node densitythreshold and critical probability when percolation of the wireless sensor networkscoverage and connection occurs and also proved the validity of the model throughexperimental simulation.(3) I did research to WSN broadcasting algorithm and analyzed the problem ofbroadcasting storm of flooding broadcasting algorithm. Based on the coverage andconnection model, I improved the traditional probability broadcasting algorithm ofwireless sensor networks. Directional antenna and directional transmit were used to findout the critical transmit probability to reach broadcasting percolation, which ensured thecoverage and connection ability and broadcasting efficiency of WSN. By usingexperimentalsimulation,it is proved that the improved algorithm is obviously superior tothe traditional probability broadcasting algorithm in message achievable rate and theend-to-end average delay performance. By studying the WSN route algorithm, I analyzedproblems of channel fight and signal crash resulted from the excessiveness of RREQflooding broadcasting information frame in AODV (Ad Hoc Ondemand Distance Vector) route broadcasting algorithm, and proposed improved AODV route algorithm usingcoverage and connection percolation model broadcasting, which is superior to the currentAODV algorithm. And it is further proved the function advantages of directionalbroadcasting algorithm which is proposed based on coverage and connection percolationmodel.(4) By analyzing the related characteristics of hyperspectral image, I compared thespatial related and spectral related features and came to a conclusion that the spectralrelativity is bigger than the spatial relativity. According to the spectral correlation index ofhyperspectral image band, serial bands which possess high spectral correlation index areused as video image sequence, and motion estimation method and predicted residualcoding method are used to eliminate the spectral relativity between adjacent bands; and tothe bands whose spectral relativity are low, single band predictive coding is used toeliminate internal self-relativity. By testing the standard hyperspectral image sequence andin certain condition of bpp, PSNR and compression function index are higher thanJPEG2000. |
PDF Full Text Request