Clustering is a significant technique in data mining, which can uncover the hidden correlation information and obtain deeper understanding of the inherent structure of data. However, when dealing with the data with extremely uneven density and increasingly complex structure, most current clustering algorithms only obtain results at a single granular level, resulting in a unilateral understanding of the data. Therefore, a novel multi-granularity clustering algorithm based on grid partition and fuzzy quotient space (MGCGF) is proposed in this paper. Firstly, by introducing the Gaussian kernel density function to characterize the distribution characteristics of data, a grid partition method is designed to select representative points for clustering. Secondly, based on the results of grid partition, the representative points composed of the maximum density values in each dimension of the grids are used for multi-granularity clustering to improve the clustering efficiency. Finally, a multi-granularity clustering algorithm is proposed by introducing fuzzy quotient space theory with representative points as input. MGCGF can be used to uncover the hierarchical structure of the data itself and form a multi-granularity space. And the clustering results with multi-granularity can be directly provided from multi-granularity spaces without re-clustering. By comparing with the other six clustering algorithms, the feasibility is verified in terms of both efficiency and accuracy.

Integrity and privacy are two important secure matrices in cyber security. Due to the limited resources and computing capability of the sensor nodes, it is challenging to simultaneously satisfy these two matrices for top-k querying in two-tiered sensor networks. To solve this problem, this paper proposes a weight-bind-based secure top-k query processing scheme (WBB-TQ), which utilises both the order-preserving symmetric encryption scheme (OPES) and the pairwise-key encryption technique to ensure data privacy in top-k querying. Since OPES can keep the size orders of the sensed data items unchanged before and after they are encrypted, the upper-layer storage nodes in the network can process top-k queries without knowing the exact values of the sensed data items. To guarantee the completeness of query results, we propose a novel method to establish chaining relationship among all the data items generated by each sensor node. By checking whether the relationship holds on not, Sink can find out whether adversaries drop and/or tamper with part or all of the qualified top-k data items in the query results. Theoretical analyses show that WBB-TQ can preserve data integrity and privacy of the top-k query results. Extensive simulation results further demonstrate that, WBB-TQ incurs very low computational and communication cost in securing top-k querying.