Research On Clustering Hierarchy Routing Protocol For Wireless Sensor Networks

With the development of the Technology of sensor wirelesscommunication and data processing, more and more devices integratingsensor, communication and processing have been invented and manufactured.Such a large number of the sensor device capabilities of wirelesscommunications and data processing through a certain protocol constitutesself-organizing network, can process data collection and transmission, andprovide useful information for users, and this is wireless sensor networktechnology. Wireless sensor network technology will be one of the mostpromising technologies in the 21st century.Routing technology at the network layer is pivotal in the architecture ofwireless sensor networks. As an active branch of routing technology,cluster—based routing protocols excel in network topology management,energy minimization, data aggregation and so on. At present, all the design ofhierarchy clustering routing protocols is basically divided into three stages,cluster head selection, cluster formation and data transmission. Cluster headselection is the basis of cluster formation, cluster routing namely the datatransmission of clusters depends on the structure of clusters. They are keytechnologies for design of WSN hierarchy clustering routing protocols. AllWSN hierarchy clustering routing algorithm is considered and designedaround these three areas. Many researchers at home and abroad raised manyexcellent hierarchy clustering routing protocols. Among them, some protocolsdesigned whole algorithm which refer to algorithm from the formation of thecluster header to the data transmission. Some is only their own algorithms onone aspect, but no matter which stage of the agreement, how to save energyand extend the life cycle of the network must be the core issues which aretaken into account.In this paper,Based on analysis and comparison of the existing hierarchy clustering routing protocols, focusing on researching and improving thecluster-based wireless sensor network data collection protocols -EADEEGagreement. The thesis focuses on the three key technologies for design ofclustering algorithm, conducted an in-depth analysis for the performance ofthe EADEEG. Analysis showed that EADEEG can effectively extend the lifeof the network by minimizing network communication overhead and goodenergy load balancing methods. But EADEEG does not take into account thedata redundant and data aggregation, and there are several deficiencies in theaspects of cluster head selection, the choice of activities nodes in cluster andperformance verification of the protocol. This paper improved thesedeficiencies of EADEEG. According to the relativity with time and space ofdata in a real-life scene, raise the concept of spatial data dissimilarity degree,combine it with EADEEG, and propose the clustering protocol which is basedon data aggregation. The result of our experiment proves that the improvedprotocol can effectively balance energy consumption and prolong the networklifetime.In this paper, the main work and results are as follows:1. Introducing the background and significance of the study, summarizedsome research work on the wireless sensor networks clustering routingprotocol in recent years, on the basis of its analyze the current study situation.2. Introduce theoretical basis of the research work in detail.This part firstly introduces the key technologies of the design ofcluster—based routing protocols, including the election of the cluster header,cluster formation and data transmission. Compared and analyzed existingwireless sensor network protocols and clustering algorithms against the threeaspects. Lay the foundation for the lower in-depth study of clustering routingalgorithm. Further analyze the wireless sensor networks clustering routingmechanism, focusing on the electing of clusters header, clusters formation andcluster routing present the current typical clustering algorithm, and compare and analyze the characteristics and application of these algorithms. Finallycurrent research status quo, pointing out the pivot of study clustering routingalgorithm in the future.3. Depthly analyze the principles and performance of EADEEG, focusingon analyze its differences from other clustering algorithm and its shortage.Firstly analyze network model and the problem description of EADEEG,research the design principle and performance analysis, propose some lack oftheories verification and improved them. And represent and prove thequantitative relationship of the cluster radius and the probability ofmaintaining cluster header single, as well as nodes transmitting radius value toform minimum energy network. Then focus on the algorithm for selectingactive cluster nodes , found that although EADEEG consider energy issues inthe course of the node data collection, but did not consider the specific dataredundancy. Further aim at these issues, introduce one data aggregationtechnology based on spatial data dissimilarity degree and improve it. Firstestablish the neighbors table related to spatial data dissimilarity degree, andthen propose algorithm for first electing cluster header based on three factorsthe remaining nodes energy, the average residual energy of neighbor nodes, aswell as data identification code marked spatial data dissimilarity degree.Following propose data transmission algorithm and a theoretical analysis ofimproved algorithms and the performance.4．Use simulation tool OMNET++ to simulate EADEEG and ABDEEG,analyze the results by experiments on EADEEG and ABDEEG, and resultsshow that in the balance of network load and the extension of the network lifecycle, ABDEEG performs more superior than EADEEG.First introduce the use of simulation tools OMNET++, introduce itsframework structure and grammar. Then the network modeling for thisalgorithm in detail, as well as the realization. Finally introduce the simulationof the improved algorithm. Analyze and compare simulation results to show the superiority of the improved algorithm. The simulation shows that theimproved algorithm to reduce energy consumption about 15 percent, thenetwork will extend the life cycle of about 20 percent.ABDEEG(Aggregation—Based Distributed Energy-Efficient dataGathering protocol for wireless sensor networks) presented in this paper,combine the node residual energy, the average residual energy of neighbornodes and data dissimilarity degree three different factors together to the mainparameters of selecting a cluster head, use data critical code as the basis fordata integration, specify spatial data redundancy, make redundancy datarelated to the temporal and spatial can be more accurately be described indetail, and improve the lack of active node selection algorithm that isinformation is easily lost. The study in the past only research efficiency ofdata transmission, this paper break through the limitations, integrate datadissimilarity degree into selecting active nodes, form a new data aggregationtechnology and use the technology to Clustering protocol, lay a goodfoundation for studying and designing more energy efficient clustering routingprotocol for wireless sensor networks.The study of the paper work can be used for application scenes whichhave intensive distribution network nodes, a higher relativity temporal andspatial data and the requirement for the lower accurate data, so it have acertain theoretical significance and value.