兵工自动化
兵工自動化
병공자동화
ORDNANCE INDUSTRY AUTOMATION
2013年
9期
68-72
,共5页
空基伪卫星%蜂窝波浪式网络%经验法%遗传算法%多指标优化%PDOP
空基偽衛星%蜂窩波浪式網絡%經驗法%遺傳算法%多指標優化%PDOP
공기위위성%봉와파랑식망락%경험법%유전산법%다지표우화%PDOP
air-borne pseudo satellite%pseudo satellite network in cellular wave-mode%empirical approach%genetic algorithm%multiple index optimization%PDOP
为能实现近地面大区域覆盖的多指标空基伪卫星网络的布局优化,提出了蜂窝波浪式伪卫星网络,并定义、分析了网络设计及优化的衡量指标。鉴于遗传算法大范围搜索能力弱、优化结果容易陷入局部最优等问题,因此对网络布局进行优化时,采用先由经验法粗略确定影响网络性能参数的取值范围,再用遗传算法在小区间内对网络性能指标进行多参数精寻优的方法,最后对优化后的网络性能进行研究分析。仿真结果表明:提出的蜂窝型波浪式网络优化后能实现广域覆盖,能保证服务区 PDOP 的可用性;通过对先用经验法后用遗传算法与直接用遗传算法2种方式网络优化性能的比较,还能得出前者能提高网络优化的效率及性能。
為能實現近地麵大區域覆蓋的多指標空基偽衛星網絡的佈跼優化,提齣瞭蜂窩波浪式偽衛星網絡,併定義、分析瞭網絡設計及優化的衡量指標。鑒于遺傳算法大範圍搜索能力弱、優化結果容易陷入跼部最優等問題,因此對網絡佈跼進行優化時,採用先由經驗法粗略確定影響網絡性能參數的取值範圍,再用遺傳算法在小區間內對網絡性能指標進行多參數精尋優的方法,最後對優化後的網絡性能進行研究分析。倣真結果錶明:提齣的蜂窩型波浪式網絡優化後能實現廣域覆蓋,能保證服務區 PDOP 的可用性;通過對先用經驗法後用遺傳算法與直接用遺傳算法2種方式網絡優化性能的比較,還能得齣前者能提高網絡優化的效率及性能。
위능실현근지면대구역복개적다지표공기위위성망락적포국우화,제출료봉와파랑식위위성망락,병정의、분석료망락설계급우화적형량지표。감우유전산법대범위수색능력약、우화결과용역함입국부최우등문제,인차대망락포국진행우화시,채용선유경험법조략학정영향망락성능삼수적취치범위,재용유전산법재소구간내대망락성능지표진행다삼수정심우적방법,최후대우화후적망락성능진행연구분석。방진결과표명:제출적봉와형파랑식망락우화후능실현엄역복개,능보증복무구 PDOP 적가용성;통과대선용경험법후용유전산법여직접용유전산법2충방식망락우화성능적비교,환능득출전자능제고망락우화적효솔급성능。
To achieve the layout optimization of multiple index air-borne pseudo satellite network of ground-level and large-scale coverage, the paper puts forward pseudo satellite network in cellular wave-mode, defines and analyses the measurable indicator of network design and optimization. Considering the weakness of wide-range search ability of genetic algorithm and the problem of the optimal result that is easy to fall into local optimum and so on, the suitable means of network layout optimization is that using empirical method to roughly determine the value range that affects network performance parameters, after that using genetic algorithm to conduct the multi-parameter optimizing on network performance indicators in small zones. Finally, research and analysis on performance of network which was optimized should be carried out. The simulation results show that the proposed cellular wave-type network can achieve wide area coverage and ensure the availability of PDOP in service area. By comparing the two approaches which are a combination of empirical approach first and genetic algorithm next and empirical approach merely and directly to optimize network performance, the conclusion can be achieved that the former approach can improve the efficiency and performance of network optimization.