{"title":"安全时间:移动对等网络中cKNN的分布式实时监控","authors":"Kihwan Kim, Ying Cai, Wallapak Tavanapong","doi":"10.1109/MDM.2008.18","DOIUrl":null,"url":null,"abstract":"A continuous k nearest neighbor (cKNN) query is a query that continuously returns a set of k nearest moving objects (mobile hosts) to a given query point. For example, report three nearest moving sensors to a given location continuously. Most existing research efforts focus on centralized solutions. In a mobile peer-to-peer network (M-P2P), a centralized approach incurs expensive communication cost. In this paper, we propose Safe-Time - a distributed solution for cKNN given a stationary query point for M-P2P. The two key features are as follows. 1) Actual execution of a cKNN query is not needed during a safe-time period since the query result is guaranteed to remain the same during this period. 2) Once the safe-time expires, execution of a cKNN query involves only objects in a circular band of width equal to the estimated distance between the kth and the k + 1th nearest neighbors. To further reduce communication cost for dense queries, we introduce Unite-Safe-Time that executes one virtual query derived from nearby queries instead of executing each of them separately. Our simulation result shows that the proposed distributed solutions outperform a centralized solution under a range of conditions. Safe-Time incurs up to 2/3 less communication cost compared to a centralized solution. Unite-Safe-Time shows up to 1/3 less communication cost than Safe-Time in our study.","PeriodicalId":365750,"journal":{"name":"The Ninth International Conference on Mobile Data Management (mdm 2008)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"12","resultStr":"{\"title\":\"Safe-Time: Distributed Real-Time Monitoring of cKNN in Mobile Peer-to-Peer Networks\",\"authors\":\"Kihwan Kim, Ying Cai, Wallapak Tavanapong\",\"doi\":\"10.1109/MDM.2008.18\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A continuous k nearest neighbor (cKNN) query is a query that continuously returns a set of k nearest moving objects (mobile hosts) to a given query point. For example, report three nearest moving sensors to a given location continuously. Most existing research efforts focus on centralized solutions. In a mobile peer-to-peer network (M-P2P), a centralized approach incurs expensive communication cost. In this paper, we propose Safe-Time - a distributed solution for cKNN given a stationary query point for M-P2P. The two key features are as follows. 1) Actual execution of a cKNN query is not needed during a safe-time period since the query result is guaranteed to remain the same during this period. 2) Once the safe-time expires, execution of a cKNN query involves only objects in a circular band of width equal to the estimated distance between the kth and the k + 1th nearest neighbors. To further reduce communication cost for dense queries, we introduce Unite-Safe-Time that executes one virtual query derived from nearby queries instead of executing each of them separately. Our simulation result shows that the proposed distributed solutions outperform a centralized solution under a range of conditions. Safe-Time incurs up to 2/3 less communication cost compared to a centralized solution. Unite-Safe-Time shows up to 1/3 less communication cost than Safe-Time in our study.\",\"PeriodicalId\":365750,\"journal\":{\"name\":\"The Ninth International Conference on Mobile Data Management (mdm 2008)\",\"volume\":\"13 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2008-04-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"12\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"The Ninth International Conference on Mobile Data Management (mdm 2008)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/MDM.2008.18\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"The Ninth International Conference on Mobile Data Management (mdm 2008)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MDM.2008.18","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Safe-Time: Distributed Real-Time Monitoring of cKNN in Mobile Peer-to-Peer Networks
A continuous k nearest neighbor (cKNN) query is a query that continuously returns a set of k nearest moving objects (mobile hosts) to a given query point. For example, report three nearest moving sensors to a given location continuously. Most existing research efforts focus on centralized solutions. In a mobile peer-to-peer network (M-P2P), a centralized approach incurs expensive communication cost. In this paper, we propose Safe-Time - a distributed solution for cKNN given a stationary query point for M-P2P. The two key features are as follows. 1) Actual execution of a cKNN query is not needed during a safe-time period since the query result is guaranteed to remain the same during this period. 2) Once the safe-time expires, execution of a cKNN query involves only objects in a circular band of width equal to the estimated distance between the kth and the k + 1th nearest neighbors. To further reduce communication cost for dense queries, we introduce Unite-Safe-Time that executes one virtual query derived from nearby queries instead of executing each of them separately. Our simulation result shows that the proposed distributed solutions outperform a centralized solution under a range of conditions. Safe-Time incurs up to 2/3 less communication cost compared to a centralized solution. Unite-Safe-Time shows up to 1/3 less communication cost than Safe-Time in our study.