28866.傳感器k覆蓋博弈基于非后悔值學(xué)習(xí)的分布式優(yōu)化算法

1、扉頁(yè)：獨(dú)創(chuàng)性聲明本人聲明所呈交的論文是我個(gè)人在導(dǎo)師指導(dǎo)下進(jìn)行的研究工作及取得的研究成果。除了文中特別加以標(biāo)注和致謝的地方外，論文中不包含其他人或集體己經(jīng)發(fā)表或撰寫(xiě)過(guò)的研究成果，對(duì)本文的研究做出貢獻(xiàn)的集體和個(gè)人均己在論文中作了明確的說(shuō)明并表示了謝意。研究生簽名：耋熏日期：塑f竺』論文使用和授權(quán)說(shuō)明本人完全了解云南大學(xué)有關(guān)保留、使用學(xué)位論文的規(guī)定，即：學(xué)校有權(quán)保留并向國(guó)家有關(guān)部門(mén)或機(jī)構(gòu)送交學(xué)位論文和論文電子版；允許論文被查閱或借閱；學(xué)?？梢?/p>

2、公布論文的全部或部分內(nèi)容，可以采用影印、縮印或其他復(fù)制手段保存論文。(保密的論文在解密后應(yīng)遵循此規(guī)定)研究生簽名：整熏導(dǎo)師簽名：琿本人及導(dǎo)師同意將學(xué)位論文提交至清華大學(xué)“中國(guó)學(xué)術(shù)期刊(光盤(pán)版)電子雜志社”進(jìn)行電子和網(wǎng)絡(luò)出版，并編入CNKI系列數(shù)據(jù)庫(kù)，傳播本學(xué)位論文的全部或部分內(nèi)容，同意按《中國(guó)優(yōu)秀博碩士學(xué)位論文全文數(shù)據(jù)庫(kù)出版章程》規(guī)定享受相關(guān)權(quán)益。研究生簽名：——導(dǎo)師簽名：日期：KCoveragesensornetworkADistri

3、butedOptimizationAlgorithmbasedonNoregretlearningDongXiaAbstract：Withtherapiddevelopmentofwirelesscommunicationtechnology，low—cost，lowpowermultifunctionaldevelopmentandapplicationofwirelesssensornetworkshasbecomeahottopi

4、cnowadaysHowtoachieveoptimalcoverageareaandmakesnetworkmonitoringtoensurereliablenetworkperformancewhileextendingthesurvivaltimeisanimportantissueinwirelesssensornetworks[171Sensornetworkisdividedintegionalcoverageandtar

5、getcoverageAimingtargetcoverageproblem，acombinationofalargenumberofwirelesssensornetworkbasedtargetcoveragealgorithmliteratures，analyzesseveralcoveringalgorithmcurrentlyusedfortargetcoverageproblem，thencovetingalgorithmc

6、ombinedandcentralizedcoveragealgorithm，thispaperproposesanoptimizedwirelesssensornetworksKcoverageoptimizationalgorithmIntheareatobemonitored，accordingtotheimportantvalueofeachmonitoringpointisdifferent，combinedwithgamet

7、heorythesensornetworkintoKgroupsindifferenttimeperiodsseparatedbydifferentmonitoringsensorsInthetargetareaiscompletelyguaranteedtobemonitoredsimultaneously，UsetheLearningmethodsusingnonregretvalueadjustmentofthesensorgro

8、upingstrategythecoverageareaandtOmonitorextendeddetectiontimeFinallygetthebestcoveragesituationwhereallthebiggestgainvaluesaremonitored，andmonitoringthelongestHighlighttheadvantageofthisalgorithmisthatitimplementsacomple

9、tedistributedunderconditionsoflimitedenergyreachedahighefficiencyAfterconductingatheoreticalstudyfinallythealgorithmwasvalidatedsimulationTheresultsshowthatthevalueoflearningbasedonnonregretsensornetworkoptimizationalgor