Research On Key Technologies Of Data Aggregation In Internet Of Things Environment

In recent years,with the popularization of sensor applications,the Internet of Things technology(IOT)has been widely used in industrial manufacturing,logistics,intelligent transportation,medical and health,environmental monitoring,security system,smart home and other fields.The IOT technology promotes the development of information technology in various fields such as industry,military and civilian,and sets off the fourth revolutionary wave of the information industry.In data aggregation,an intermediate node in the network does not directly forward the data from predecessor nodes,but the intermediate node aggregates the data from predecessor nodes and itself by performing a certain computation function(e.g.,average,sum,maximum,and minimum)and then only forwards the aggregated data to its successor node.Since data aggregation operation aggregates multiple data in to one data,data aggregation operation saves resources in energy,computing power,bandwidth,etc.Due to the limitation of resources,data aggregation increases the life time of network,and data aggregation is a critical operation in the IOT.In recent years,the scale of sensor nodes in the IOT has grown rapidly.Correspondingly,the scale of data in the IOT has also exploded.With the rapid growth of the number of nodes and the size of data,there are several challenging issues in the aggregation of Internet of Things data that need to be addressed:(I)The problem of maximizing the quality of aggregation(QoA)under deadline constraint.Due to the increase in the scale of sensor nodes in the IOT,more and more applications have strict requirements on the latency of data aggregation.In this context,many applications have specific constraints on the latency of data aggregation.In deadline-constrained data aggregation,some nodes are unable to participate in data aggregation due to the time constraint,which leads to a decline in the data aggregation quality.Therefore,maximizing the quality of aggregation(QoA)under deadline constraint is an urgent problem that should be solved in data aggregation of the IOT environment.(2)The problem of trusted computing in data aggregation.The wireless sensor networks(WSNs)are often deployed in unattended hostile environments.In unreliable WSNs,the data would be retransmitted while transmission failed,which increases the energy consumption and the latency of daga aggregation.Due to the limitation of enery,energy saving is a key issue that WSNs must consider.More and more applications save energy by restricting the number of retransmissions.Therefore,maximizing the QoA under deadline and energy constraints in unreliable WSNs is a challenge of data aggregation in IOT.(3)The problem of high trusted data aggregation in Internet of Things.Because sensor networks are often deployed in unattended hostile environments,they are prone to various attacks from inside and outside of the network.For external attacks,traditional security technologies such as encryption and privacy protection can be used for prevention and processing.However,traditional security technologies can not address attacks from the internal nodes of network.The trust management mechanism is a reliable solution to address attacks from the internal nodes of network.Therefore,how to achieve trusted data aggregation in the IoT environment is a challenge at present.This paper first analyzes the research status of data aggregation in the Internet of Things environment,and summarizes the existing problems and challenges of current research work.On this basis,this paper studies the following key issues:maximizing the QoA under deadline constraint;maximizing the QoA under deadline and energy constraints in unreliable WSNs;high trusted data aggregation in Internet of Things.The main research content and contributions of this paper are as follows:(1)To maximize the QoA under deadline constraint,this paper proposes a distributed and efficient heuristic algorithm.Starting from the sink,the heuristic algorithm performs the construction of data aggregation tree and the node scheduling from top to bottom.The execution of the algorithm is a two-step process.First,the algorithm schedules the nodes in the ready state.An optimal node is selected from the already scheduled neighbor nodes as the parent node of each ready state node,and time slot and channel are allocated for each ready state node.The principle of selecting the optimal parent node is:(a)the node in ready state does not conflict with scheduled nodes,(b)the node that is selected as the parent node of the ready node allocates the maximum time slot and maximizes time slot reuse.The second step of the algorithm is to resolve conflicts between ready state nodes.When there are conflicts between the ready state nodes,the best configuration nodes are preferentially scheduled,and the other nodes transfer to the"waiting to be scheduled state".The nodes with the best configurations are the nodes that maximize time slot reuse.When all conflicting ready state nodes with best configurations are scheduled,the nodes in "waiting to be scheduled state"transfer to the ready state,and they reselect the parent node and are reassigned to slots and channels.In addition,the algorithm increases the quality of data aggregation under deadline constraints by using multi-channel(?1)communication technology.In summary,the algorithm not only achieves the construction of the optimal data aggregation tree,but also completes the optimal scheduling of nodes.The experimental results show that compared with the existing research work,the algorithm not only improves the quality of data aggregation under deadline constraint,but also resolves neighbor node conflicts.Especially in the network with a complete graph structure,the performance of the algorithm reaches a theoretical maximum.(2)To maximize the QoA under deadline and energy constraints in unreliable environments.This paper proposes an algorithm for constructing the optimal aggregation tree based on Markov chain.The problem of maximizing the quality of data aggregation under deadline and energy constraints in an unreliable environment is essentially a combinatorial optimization problem.The Markov chain is a random process without memory.The state of time t+1 is only related to time t,and is independent of the state before time t.The Markov chain has been used in recent years for similar combinatorial optimization problems.Different aggregation trees are visited through Markov random walks,so that the trees with higher QoAs have greater probabilities of being visited,and thus the optimal aggregation tree is constructed.The existing work only studies the problem of maximizing the quality of data aggregation under time and energy constraints in an unreliable environment in the case of a fixed data aggregation tree.The existing work does not consider the impact of the data aggregation tree on the quality of data aggregation.The experimental results show that compared with the existing research work,the proposed method improves the quality of data aggregation under time and energy constraints in an unreliable environment.(3)For the high trusted data aggregation in Internet of Things,this paper proposes a trustworthy data aggregation algorithm based on context and data density correlation degree.In this algorithm,the nodes are firstly clustered according to the existing data density correlation degree based clustering algorithms,and then data trust is calculated based on clustering.Due to clustering,the value of data trust is more accurate than that of existing work.For the problems of node trust,data trust and link trust,different measures are adopted,which improves the throughput and robustness of system.The experimental results show that compared with the existing research work,the proposed scheme not only improves the accuracy of data trust but also improves the system's ability to resist attacks and throughput under malicious attacks.In summary,this paper focuses on the problem of maximizing the quality of data aggregation under deadline and energy constraints and the problem of trusted computing of data aggregation.A number of new research advances has been made in maximizing the quality of data aggregation and the trusted computing of data aggregation.The research results have important theoretical significance and application value for achieving highly reliable data aggregation in the Internet of Things environment.