Research And Implementation Of Lightweight Variable Length Block Encryption Scheme In Internet Of Things

It is difficult to use traditional block ciphers to solve the growing data security problems in Internet of Things(IoT).On the one hand,the micro computing and processing devices used in IoT,with weak computing power and limited storage capacity,cannot effectively support the computing resources required to encrypt data using traditional block ciphers.On the other hand,with the rapid development of IoT,the data generated by IoT end devices is massive and diverse.Therefore,there is a need for block ciphers that can provide flexible and optional input parameters for encrypting complex data.To address these issues,this thesis investigates lightweight variable length block cipher schemes for IoT.The main work of this thesis is as follows:A new lightweight cryptographic algorithm based on the Feistel structure,RAB,is designed.The original version of RAB encrypts plaintexts of 64-bit block length by iterating a round function for 10 rounds.To solve the problem of slow diffusion of Feistel structures,RAB uses a diffusion layer consisting of an Maximal Distance Separable(MDS)matrix,which allows the 3-round RAB to satisfy the avalanche effect.In order to handle different key lengths and provide different security strengths,RAB applies a key schedule that can receive different key lengths.In addition,this thesis performs differential and linear cryptanalysis of RAB based on mixed integer linear programming(MILP)techniques.The results show that RAB can meet the data security requirements of resource-constrained devices.The results of the efficiency analysis of the software and hardware implementation of RAB show that RAB has a high software and hardware implementation efficiency.A new lightweight encryption algorithm,LILP,based on the Lai-Massey architecture,transforms the RAB into a special block cipher characterised by the ability to accept as input plaintexts of at least twice the block length.The LILP uses a symmetric structure and an involute design approach,which allows decryption to reuse the encrypted circuit or code.In addition,LILP uses lightweight components internally to accommodate the limited computing power of embedded devices.The results of the security and performance analysis of LILP show that LILP is more efficient than its traditional counterpart in encrypting data on resource-constrained devices.Finally,a general-purpose IoT terminal encryption system is designed and implemented.The system uses the two encryption algorithms described above.The system uses a ZigBee network as the perception layer terminal in IoT.A cloud server and a mobile phone application as the application layer terminal in IoT.Sensor data is collected at the end devices in the ZigBee network and encrypted into sensor ciphertext,which is transmitted to the cloud server.The mobile application queries the cloud server for the sensor ciphertext and then decrypts and displays it.Experimental results show that the use of the two encryption algorithms proposed in this thesis can effectively protect the data security of end devices in IoT.