Abstract:As an important part of Internet of Things (IoTs ), Wireless Sensor Networks (WSNs)can be widely deployed in many physical environments for data collection and monitoring,such as oceans,volcanoes,animal habitats and battlefields.In these applications of WSNs,k Nearest Neighbors (k-NN)query which returns the nearest k objects of the given query location or value is very important for users to get the required information of monitoring areas.In consideration of the built-in resource limitation problems of single tier sensor networks,k-NN query can be more efficient in two tiered WSNs which contain a large number of resource constrained sensor nodes and fewer resource rich management nodes.However,relying on management nodes for data storage and query processing raises significant security concerns.Frequently,the deployment environments of WSNs are hostile and unpredictable,which threaten the data security and privacy.What is worse,the open nature of wireless communication makes it easy for adversaries to eavesdrop data packets or compromise nodes of sensor networks.Management nodes are more vulnerable to be compromised as they store massive data and serve as an important role for responding queries. The compromised management nodes bring serious security risks to the network and significantly restrict the development of two tiered WSNs.Therefore,in recent years,privacy preservation in two tiered WSNs has attracted more and more attention.Accurately answer k-NN query while preserving data privacy is a challenging problem.This paper presents an efficient privacy-preserving k-NN query protocol for two tiered WSNs.To the best of our knowledge,this paper is the first to consider collusion attacks for a value-based k-NN query in two tiered WSNs while fulfilling the preservation of privacy.To improve query efficiency,the architecture of k-NN query is proposed based on the directional storage strategy for two tiered WSNs.To resist management nodes compromising attacks,this paper proposes a novel privacy-preserving data encoding mechanism to conceal sensitive information to response the accurate query result without data privacy leakage. Then,to resist nodes collusion attacks,we present a proxy-nodes-based one-way data hiding mechanism to disorder the data correlation between common sensor nodes and management nodes.In addition,a series of improve schemes are further introduced to enhance the performance of proposed privacy-preserving approaches,including group-based random storage mechanism, dynamic query mechanism,subfields partition mechanism and data code compression mechanism. Finally,theoretical analysis and experimental results of real dataset confirm the privacy and efficiency of our proposals.

Privacy-Preserving k-NN Query Protocol for Two-Tiered Wireless Sensor Networks

Content-Aware Data Dissemination for Enhancing Privacy and Availability in Wireless Sensor Networks

A secure and efficient user authentication protocol for two-tiered wireless sensor networks

Enhancing Sink-Location Privacy in Wireless Sensor Networks Through K-Anonymity

An Efficient and DoS-resistant User Authentication Scheme for Two-Tiered Wireless Sensor Networks

New Secure and Efficient Distributed Data Storage Approaches for Unattended Wireless Sensor Network

Efficient and Secure K-Nearest Neighbor Query on Outsourced Data

Secure $K$k-Nn Query on Encrypted Cloud Data with Multiple Keys

Outsourcing Two-Party Privacy Preserving K-Means Clustering Protocol in Wireless Sensor Networks

Achieving Practical and Privacy-Preserving kNN Query over Encrypted Data

Efficient Probabilistic K-NN Computation in Uncertain Sensor Networks

A General Purpose Data and Query Privacy Preserving Protocol for Wireless Sensor Networks

C-kNN Query Processing in Object Tracking Sensor Networks

Privacy Preservation in a Two-Tiered Sensor Network Through Correlation Tracking

Privacy-preserving k-NN interpolation over two encrypted databases

Privacy-preserving K Nearest Neighbor Query with Authentication on Road Networks

Preserving Privacy Against External and Internal Threats in WSN Data Aggregation

Secure Spatial Network Queries on Cloud Platform

Secure k-Nearest Neighbor Query over Encrypted Data in Outsourced Environments

Achieving Private, Scalable, and Precise Data Collection in Wireless Sensor Networks

Query on the cloud: improved privacy-preserving k-nearest neighbor classification over the outsourced database