铁道学报
鐵道學報
철도학보
2015年
2期
39-44,45
,共7页
张雁鹏%党建武%刘晓娟%闵永智
張雁鵬%黨建武%劉曉娟%閔永智
장안붕%당건무%류효연%민영지
基于通信的列车控制%车地通信%Stackelberg博弈%越区切换%协作分集
基于通信的列車控製%車地通信%Stackelberg博弈%越區切換%協作分集
기우통신적열차공제%차지통신%Stackelberg박혁%월구절환%협작분집
Communications Based Train Control (CBTC )%train-to-wayside communications%Stackelberg games%handoff%cooperative diversity
针对城市轨道交通CBTC系统无线局域网越区切换问题,结合具有竞争机制的Stackelberg博弈模型,提出一种采用协作分集技术的列车越区切换算法,满足CBTC系统可靠连续通信的要求。本文分析列车在无线局域网环境下的中继触发和越区切换过程,中继节点根据列车请求的带宽确定价格策略,设计一种基于价格和收益的列车效用函数,证明纳什均衡的存在性。为获取列车的最优带宽策略和网络收益,提出一种分布式迭代学习法求解纳什均衡,并进一步分析列车越区切换过程。仿真结果表明,本文提出的博弈策略能够激励中继节点参与协作,合理分配网络资源;列车动态越区切换时间小于50ms,切换成功概率明显高于传统切换方法。
針對城市軌道交通CBTC繫統無線跼域網越區切換問題,結閤具有競爭機製的Stackelberg博弈模型,提齣一種採用協作分集技術的列車越區切換算法,滿足CBTC繫統可靠連續通信的要求。本文分析列車在無線跼域網環境下的中繼觸髮和越區切換過程,中繼節點根據列車請求的帶寬確定價格策略,設計一種基于價格和收益的列車效用函數,證明納什均衡的存在性。為穫取列車的最優帶寬策略和網絡收益,提齣一種分佈式迭代學習法求解納什均衡,併進一步分析列車越區切換過程。倣真結果錶明,本文提齣的博弈策略能夠激勵中繼節點參與協作,閤理分配網絡資源;列車動態越區切換時間小于50ms,切換成功概率明顯高于傳統切換方法。
침대성시궤도교통CBTC계통무선국역망월구절환문제,결합구유경쟁궤제적Stackelberg박혁모형,제출일충채용협작분집기술적열차월구절환산법,만족CBTC계통가고련속통신적요구。본문분석열차재무선국역망배경하적중계촉발화월구절환과정,중계절점근거열차청구적대관학정개격책략,설계일충기우개격화수익적열차효용함수,증명납십균형적존재성。위획취열차적최우대관책략화망락수익,제출일충분포식질대학습법구해납십균형,병진일보분석열차월구절환과정。방진결과표명,본문제출적박혁책략능구격려중계절점삼여협작,합리분배망락자원;열차동태월구절환시간소우50ms,절환성공개솔명현고우전통절환방법。
To solve the issue of handoff management of WLAN in CBTC systems for urban rail transit ,a hand‐off algorithm using cooperative diversity technology for trains has been proposed , w hich has been combined with Stackelberg game model with competitive mechanism , to meet the needs for continuous communications for CBTC systems with high reliability . After the analysis of relay triggering and the handoff process of a run‐ning train under the WLAN environment , the price strategy has been determined by relay nodes in the light of the bandwidth requested by the train . Then ,a utility function of train ,based on price and revenue ,has been designed , and the existence of Nash equilibrium has been proved . In order to achieve the optimized bandwidth strategy of train and network revenue ,a distributed iterative learning method has been proposed to figure out the Nash equilibrium and the handoff process of train has been further analyzed . The simulation experiments show that the game strategy proposed in this paper can motivate relay nodes to participate in the cooperation for reasonable resource allocation of networks , the dynamic handoff time of the train is less than 50 ms and the handoff success probability is significantly higher than that of the traditional method .