Research On Encryption Transmission Technology Based On Hash-chain And Semi-tensor Product Compressed Sensing

With the development of smart hardware and network scale,the traditional Internet has gradually entered the Internet of Things(IoT)era.The sensor is the basic unit that constitutes the Wireless Sensor Networks(WSNs).Wireless sensor networks of different sizes are the basic units that make up the Internet of Things.Increasing size of network and complexity of network environments pose great challenges for network communication.With the application of WSN in various fields,the drawbacks of sensor nodes limited by storage and energy are becoming more and more obvious.Saving storage and reducing communication energy consumption are the top priority of research of wireless sensor network communication.And in some extra applications of wireless sensor network,such as military collaborative operations,e-government,industrial collaborative design,scene monitoring,telemedicine and education and other transmission systems,another issue of great importance is the security and privacy of the transmitted data.Compressed Sensing(CS)is an emerging technology and is widely used in signal processing and wireless communication networks.In essence,compressed sensing technology is a method of compressing signals using less resources(lossy compression).Compressed sensing can achieve data compression and data encryption at the same time.In theory,compressed sensing technology can achieve storage-saving,efficient and secure communication,but it highlights a huge limitation for wireless sensor networks with large scale.Compressed sensing is a process of symmetric encryption and both parties need to save the key to complete the process of encryption and decryption.However,when a sensor node forms two-way transmission-pairs with many other nodes,it will store a lot of measurement matrices for data communication with different nodes.When the measurement is relatively large,such defect may lead to the actual performance of CS worse than the traditional solution.In addition,since the image signal is different from other signals and the requirement for image encryption is higher,compressed sensing technology does not guarantee the security of the image signal.In order to solve the problems of storage,communication energy consumption and security,new types of CS methods are presented,which are an image encryption scheme based on hash-chain compressed sensing and an improved semi-tensor product compressed sensing model.The main research results and innovations of this paper are as follows(1)A hash-chain compressed sensing model is proposed and applied to wireless sensor networks.In the hash-chain CS model,the sender and receivier do not need to save the whole measurement matrix,and only need to save the parameters of hash-chain to generate the hash-chain matrix,which can achieve the purpose of saving storage.The hash functions have the unidirectionality,the collision constraint and the high sensitivity to the initial value,so the hash itself has high sencurity and the hash-chain also has the above properties.Compressed sensing technology itself has lightweight encryption.The hash-chain compressed sensing model combines the hash-chain with the CS technology to reduce storage and energy and improve security.(2)An image encryption model based on hash-chain compressed sensing is proposed,which greatly improves the security of image signal transmission in wireless sensor networks.The image signal has obvious pixel characteristics and high pixel correlation.Therefore,the encryption requirements of image signals are also higher than that of other signals.Image encryption scheme based on hash-chain compressed sensing adopts double hash-chain matrix encryption mechanism.The hash-chain has a large key space and strong key sensitivity.Encrypted images perform well in gray histograms,adjacent pixel correlation,and image entropy.Therefore,the image encryption scheme based on hash-chain compressed sensing has significant encryption effect.(3)An improved semi-tensor product compressed sensing transmission model is proposed.The improved semi-tensor product compressed sensing transmission model is more flexible on the basis of semi-tensor product compressed sensing.The measurement matrix can not only perform tensor operations with the unit matrix,but also perform tensor operations with the random matrix and the generalized permutation matrix.The improved semi-tensor product compressed sensing can guarantee the signal recovery effect and improve the flexibility of the algorithm.