传统的以数据为中心的存储方法有一个非常严重的缺点:网络中存在着明显的热点区域,主要出现在sink和home节点周围.位于热点区域附近的节点所消耗的能量远大于网络内的其他节点,从而严重缩短了传感器网络的寿命对于出现在sink周围的热点问题,采用在网络中分布多个sink节点的方式消除热点.对于出现在home节点周围的热点问题,提出基于环的负载平衡数据存储协议,并以此数据存储为基础,给出相应的查询处理算法消除热点.首先,负载平衡数据存储协议将传感器网络划分为多个环,在网络工作的某个时间段内,数据被分散存储在某个环内的多个节点上洪次,该存储结构的最大特点是,在不同时间段内,各环轮换工作进一步消除热点.最后,基于环的查询处理算法也是由环内的多个节点协作完成.基于环的负载平衡数据存储协议以及基于此存储协议的查询处理算法虽然在查询处理中所消耗的能量高于传统方法,但可以保证从数据存储到查询处理的整个过程中,网络的所有节点均匀地消耗能量,从而避免了传统方法中的热点问题,达到延长网络寿命的目的.实验表明,基于环的数据存储及查询处理算法可以解决传统上基于事件存储协议中出现的热点问题,并延长传感器网络的寿命.
The traditional data centric storage method has a serious problem named hotspot problem, which appears near the sink and home node. Hotspot problem causes the nodes near the hotspot consume much more energy than other nodes, which seriously shortens the lifetime of the sensor networks. In this paper, a method distributing multiple sink nodes around the sensor networks is proposed to solve the hotspot problem around the sink. A load balance data storage method based on ring and a query processing algorithm for such a kind of data storage method are proposed to solve the hotspot problem around the home node. First, the ring based load balance data storage method divides the sensor networks into a lot of rings and stores data among all nodes within a ring. Second, different rings alternatively work during the working time of the network, which avoids the hotspot problem further. Finally, the query processing algorithm based on the load balance data storage method makes all nodes within a ring cooperatively answer the user's query. Although the ring based load balance data storage method and the query processing algorithm consume more energy than the traditional method when answering a user's query, the method makes all nodes within the network consume energy evenly, which avoids the hotspot problem of the traditional method, so the lifetime of sensor networks can be extended. Experimental results also show that the ring based load balance data storage method and the query processing algorithm can solve the hotspot problem and prolong the lifetime of the sensor networks.