无线传感器网络分布式数据库加密方法研究

Research on Encryption Method of Distributed Database in Wireless Sensor Network

传统方法在加密无线传感网络分布式数据库时,受到网络异构性的影响产生了大量异常数据,导致其存在加/解密速度慢、数据查询复杂度以及解密后数据准确性低的问题。为此,提出一种无线传感网络分布式数据库加密方法。首先通过密度峰值法对数据实施聚类处理,使用机器学习中的支持向量机分类法找出数据库中的异常数据并剔除。然后通过OPES算法对数据库内数据实施转换、桶划分等操作,再通过划分结果建立数据库的加密空间,实现对无线传感网络分布式数据库的加密。仿真分析结果表明:该加密方法提高了无线传感网络分布式数据库中数据查询过程的复杂度和加/解密的速度,且应用该方法可使得解密后数据的准确性达到99.16%。

When encrypting distributed databases in wireless sensor networks,a large number of abnormal data are generated due to the heterogeneity of networks,which leads to problems such as slow encryption/decryption speed,complexity of data query and low accuracy of data after decryption.Therefore,a distributed database encryption method for wireless sensor networks is proposed.Firstly,the data are clustered with the density peak method,and support vector machine are used to find and eliminate the abnormal data in the database.Then,the data in the database is transformed and bucket divided with OPS algorithm.Finally,the encryption space of the database is established according to the division results to realize the encryption of the distributed database of wireless sensor network.Simulation results show that this encryption method improves the complexity of data query process and the speed of encryption/decryption in distributed database of wireless sensor network,and the accuracy of decrypted data can reach 99.16% by using this method.