Research On Routing And Node Localization For IoT-Oriented Wireless Network

Internet of Things refers to a kind of network which connects any things and Internet together in accordance with the agreed protocol by using the information sensor devices.The intelligent devices connect to the Internet through the communication technology,exchange information independently,and provide useful information for people to realize the intelligent identification,location,tracking,monitoring and management of objects.The Internet of Things has the potential for applications in industry,transportation,logistics,electric power,environmental protection,military affairs,public safety,daily life and so on.It has also attracted the attention of many researchers.The Internet of Things has greatly improved people's quality of life,and has brought great changes to people's lifestyles.According to China's national conditions,"Internet +","Industrial 4.0" are all production and lifestyle reforms which rely on the Internet of Things.The Internet of things is considered the world's next wave of information technology and new economic engine,and it will significantly impact the world's economy,politics,culture,military and society.The development of the Io T makes energy-limited sensing devices with battery-powered has the need of accessing to the Internet.The characteristic of energy-constrained makes the control of wireless data transmission energy become one of the focuses of applied researches.6Lo WPAN protocol stacks aim at the seamlessly connection between the IPv6 Internet and low-power wireless nodes,and the RPL routing protocol in 6Lo WPAN protocol stacks has the ability to interconnect massive nodes of the Internet of Things,is gradually become the de facto standard Internet of Things protocol.It has been supported by the new version of Bluetooth,Zig Bee And other classical wireless network protocol and is gradually applied.The Quality of Information(Qo I)sensing technology is based on the quality information of the data collected by the sensor.It can only transmit the information of the event of interest in the application,and reduce the redundancy of the transmission invalid data,and play a better effect of energy saving.Therefore,the study of the combination of the RPL and the Qo I,and constructing DODAG with the use of Qo I,is an effective way to reduce the energy consumption of wireless nodes.On the other hand,people are interested in the geographical location of theinterest event,and sometimes the only state of things which want to know is the specific location.With the help of wireless network,the nodes can learn the relationship with each other through intercommunication,and then the location of the wireless network node can be determined by the location of the reference nodes and wireless locating technology.At present,there are performance problems in the existing wireless network node locating technology that the positioning accuracy is not high,the self-adaptability is not strong and the algorithm calculating quantity is excessive.Therefore,it is of great significance to deeply study the technology of node location based on wireless network,improve the positioning accuracy and reduce the computational complexity of locating algorithm.In order to solve the routing problem and wireless node location problem for improving the network performance and prolonging the network lifetime,this paper aims at the problem of routing and wireless node location,and starts with the study of the different characteristics of the Internet of Things applications,and makes the following contributions:1.A DODAG construction mechanism for low power consumption and lossy networks is proposed,which can be used to improve the DODAG construction performance of RPL-based network for event detection.Firstly,the influence of noise on the DODAG structure process in RPL-based LLN is discussed.The optimization method of constructing DODAG in the noise background of using neighbor list is proposed.This method reduces the cost of data transmission by constructing a more efficient DODAG by increasing the overhead of less than one control message on average in the networking process.And then the quality of information is used to measure the data collected by the sensor and to construct the local DODAG for data fusion and transmission to reduce the cost of data transmission in the event detection network.The simulation results show that on the premise of ensuring the quality of information,the energy consumption of data transmission is reduced by 70%.2.A WKNN location fingerprint location method based on dispersion degree is proposed.Firstly,the off-line location fingerprint data are classified by the clustering algorithm,and the data with large difference from the measured data are excluded.It ensures that the correlation between the position fingerprint data and the sampled value are increased,and the computational complexity of the locating algorithm are decreased.Then,based on the discrete degree of the measured fingerprint data,the different weightsof the fingerprints are given with different discrete degrees,and the weighted K-nearest neighbor method is used to estimate the nodes' location.The tests of the actual scene show that the improved algorithm is better than the original WKNN algorithm with Euclidean distance as the weight coefficient,and the error fluctuation is smaller.The proposed algorithm is available for reference to the study of the wireless node location problem based on Wi-Fi.3.A distributed distance error correction MDS-MAP localization method based on residual energy clustering is proposed:(1)A clustering algorithm based on residual energy for wireless sensor networks is proposed,which makes the network clustering balanced,and the distributed MDS-MAP algorithm is used to reduce the localization error;(2)In the cluster,if the distance between nodes can not be obtained directly,the multi-hop distance correction algorithm by the adjacent node geometric relationship is adopted to correct the distance between the multi-hop nodes and reduce the localization error.(3)For the nodes that can not correct the multi-hop distance by the neighboring nodes' geometrical relations,the multi-hop distance can be approximatively obtained by the multi-hop relation;(4)After the localization in the clusters is finished,the localization result can be obtain by the fusion of the inter-cluster coordinate and the absolute coordinate transformation.This method can reduce the localization error and reduce the computational complexity of the localization algorithm,and then reduce the energy consumption.The simulation results show that the localization performance of the improved algorithm is superior to the classical MDS-MAP and MDS-MAP(P)algorithms.