Research On Top-K Query Processing Scheme In Wireless Sensor Networks

With the" sustainable development in wireless communications technology, microelectronic technology and micro-sensor technologies, wireless sensor networks (WSNs) have become a very hot research topic. WSNs, which has a dynamic and large-scale topology, is composed of a large number of sensor nodes, and it is self-organizing and application-related. WSNs can be used in many applications including military, environmental monitoring, health care, intelligent transportation and other fields. Query processing is one of the key technologies in application layer of WSNs, energy efficient query processing techniques can extend the network lifetime and enhance the user experience. Query delay and energy consumption are main aspects of performance evaluation for query processing. Top-K query, which returns the earlier k data items of the data set produced by WSNs under some rules of comparison, is the basis of many query styles in WSNs, and it surports many applications.With the help of query tree, the thesis proposes a novel feedback-based Top-K query processing scheme which is energy efficient. In the course of several interactions between a parent node and its children nodes, the children nodes return some selective data according to the feedback information received from parent node in order to reduce the amount of data packets to be transmitted. Simulation results, comparison and analysis show that this scheme, which can meet the requirements of single and accurate Top-K queries well, is energy efficient.This thesis also proposes a Top-K query processing scheme based on data distribution table, mainly for the application of high-frequency Top-K query, which must meet a certain delay requirements. The scheme uses a combination of PUSH and PULL in response to the query processing. In the first phase, the sensor nodes PUSH the data distribution information rather than the data itself to the base station, in the second phase, the base station pulls the data selectively according to data distribution table. Theoretical analysis and comparison show that, compared with the current research on repeated Top-K queries, this scheme reduces the amount of data packets to be transmitted and query delay, and only some but not all nodes participate in the query response.The Top-K query processing schemes return user accurate results. For single Top-K query, this thesis proposes a scheme based on feedback, it reduces the amount of data packets to be transmitted. For high-frequency Top-K query, it proposes a scheme based on data distribution table, not only reducing the data transmission and query delay, but also reducing the number of sensors participated.