Study Of Key Techniques For Iot Perceptual Data Transmission Based On Secure Multiparty Computation

The view “wisdom of the earth,the connection of objects” is proposed,and human beings gradually come into the era of the Internet of things from the Internet Era.Internet of things is another information industry revolutionary after the computer,Internet and mobile communication.Internet of things is based on the Internet,the use of RFID sensor and WSNs technology to build a the network information system covering all the people and things in the world,so that the economic and social life of human society,production operation and personal activities are fully run online and the smart objects are based on the infrastructure of the Internet of things.Internet of things is perception network,it is widely used in a variety of sensing technology.Internet of things is deployed a huge number of sensors,each sensor is a source of information.Different types of sensors capture different information content and format information,respectively.The data obtained by sensors are real-time,collecting the environment information according to certain frequency periodicity,continuously updated data.Data captured by the the terminal equipment on the Internet of things are called perception data.So how to transfer perception data safely is a problem that to be urgent solved.This thesis is focused on the research that how to transfer perception data safely over the Internet of things.The requirements of secure transmission over the Internet of things are higher than before.It is not only to keep the data from being attacked and tampered,but not disclosing the users' privacy at the same time.This thesis analyzes the intricate perception data.First we establish a trust mechanism on perception layer and access layer in Internet of things,putting forward a hierarchical trust mechanism and separation of different trust demand subjects in a heterogeneous environment.In the framework,we construct a credible circulation namely ” phenomenon trusted-node trusted-behavior trusted-mechanism trusted-authorization trusted”.It integrates well the perceptual node trust and reputation mechanism;we establish the perception data processing model and construct the mathematical model of this model to adapt to the security requirements of the Internet of things,thus we find Secure Multiparty Computation(SMC),namely: there are multiple parties involved in the calculation,a number of participants calculate the function together,the input information of participants are confidential,others may not know;the perception data are distributed and are not in a centralized node,we adopt the K-means clustering algorithm toprotect the privacy of distributed perception data,the calculation of the IoT construct upon cloud computing,so we have outsourced the calculation of perception data to cloud terminals,we continue to study the K-means clustering algorithm and design the multi-clustering privacy protection K-means clustering algorithm to realize the outsourcing computing.When the participates calculate,the computing is complete equivalence in the process of the agreement,each participate has its own random number to encrypt information for protecting the privacy during the interaction process,but couldn't deduce the privacy information of the other participate even if knowing the interaction results,thus implementing perception data security transmission on the Internet of things.IoT network is mainly composed of wireless sensor network,we study the problem of wireless sensor network data transmission congestion,the WSNs node transfering the perception data or pretreatment perception data to sensor nodes.When sensor nodes receive packets(or groups)exceed the upper limit of packet forwarding or processing,the excess packets are needed to be cached when the buffer is full,the excess packets will be discarded,which causes the buffer overflow,then node congestion happen;this thesis proposes an energy balance strategy based on feedback control(PLFC),using a feedback control algorithm to adjust the node transmission power,it realizes redistribution of node load on the network,solving the energy imbalance consumption of nodes on the wireless sensor network,so the strategy prolongs wireless network life cycle.Perception data come from the perception terminal equipments,at present it is popular that terminal equipments in the perception layer adopt the RFID technology.The RFID system is mainly composed of electronic tags,reader-write and servers.To protect the privacy of the RFID system,we must first authenticate both communicating sides before communication in order to prevent counterfeit electronic tag or illegal reader-write.Secondly,the information transmission between the communicating parties couldn't be in plaintext mode,certain encryption measures must be taken to prevent the privacy information leakage.A lot of RFID authentication protocols are based on reader-write and back-end server.Supposing the channel between the reader-write and back-end server being safety,the channel isn't security actually,this thesis investigates into the RFID technology and put forward perception data multidirectional authentication protocol and key agreement protocol based on RFID technology.The proposed protocol has performed the multidirectionalauthentication,the certification is carried out in label,reader-write and server,when the all authentication are all correct,the transmission channel are safety.The protocol simplifies the public value transfer between two communicating parties,and authenticating the server to convey the public value,and it has no effect on the safety,because even if the server has been broken,it is hard to pass through the follow up certification.It can effectively solve the problem of privacy protection in RFID system.Finally,we summary our work,pointing out the shortage and the next-phase work.