Research On Secure Top-k Query Protocol In Two-tiered Sensor Networks

Wireless sensor networks have been widely used in many areas,such as military,medical care and environmental monitoring.As an extension of the traditional wireless sensor networks,two-tiered sensor networks have been increasly adopted in most large-scale sensor networks since they offer good scalability,efficient query processing and energy-saving.In this two-tiered architecture,storage nodes act as an intermidle tier between sensor nodes and the sink.They store sensed data received from nearby sensors and answer queries from the sink.The important role of storage nodes in this two-tierd architecture makes them attractive to attackers.A compromised storage node can not only expose the sensitive data to attackers but also forge false query results to the sink,which poses a great threat to the whole networks.Therefore,developing a secure query protocol is of paramount importance such that the data privacy is preserved as well as the integiryt of the query results can be verified.In this thesis,we focus on challenge of desining secure top-k query protocol in two-tiered sensor networks.Our main research work is as follows:(1)We proposed a secure and verifiable top-k query protocol called SVTQ.SVTQ enables the storage node to process queries correctly while preserving the privacy of sensed data and the integrity of query results.To preserve data privacy,we proposed prime aggregation whereby storage node can compute the top-k query result without knowing the actual data values.To preserve the integiry of query result,we further proposed a novel idea of differential chain whereby data items collected by each sensor will be linked with each other just like a chain.The sink verifies the integrity of query result by checking whether the differential chain of each sensor is complete or not.Both theoretical analysis and experimental results show that SVTQ outperforms prior art in terms of power and space consumption while preserving the privacy and the integrity.(2)We proposed a privacy-preserving dynamic top-k query protocol called PPDT.To meet users' diversified query requirements,we first propsed a novel query type called dynamic top-k query which allows dynamic query based on the use-specified query weight in the query request.Besides,to process dynamic top-k query without privacy breaches,we proposed a secure query protocol called PPDT.PPDT enables the storage nodes to compute the weighted score for each sensed data without knowing exact data values.Both theoretical analysis and experimental results show that PPDT can support privacy-preserving dynamic top-k query with very low communication and space consumption.