The Research And Application Of Link Quality Estimation In Wireless Sensor Networks

As the development of Wireless Communications, Electronics and Sensor Technology, WSN (Wireless Sensor Network) attracted widespread attention gradually. By sensing, data processing and wireless communication in short distance, WSN has used from the military field to anti-terrorism, explosion-proof, environment monitoring, health care, household, commercial, industrial, and other many fields.More and more scholars were studied in the characteristics of the WSN. However, after a large number of experiments, we know the wireless sensor network has also faced many problems when use it in the observation because of the complexity of the environment and the diversity of application. Such as data delived from sensor nodes will be lost in the process of transmission, gathering node to receive data will appear the phenomenon of repetition, and network systems cannot run for a long time due to uneven energy consumption caused. These problems of generation and related to the design of WSN routing protocol, and choose routing link is due to the results of evaluation from link quality, through access to relevant data, and a large number of experiments.Link quality estimation(LQE) is a basic structure module designed from a network layer to cross layer. The LQE of existing algorithms such as WMEWMA, ETX or four-bit nowadays, which research and compared in improved algorithm based on RNP(the Required Number of Packet) algorithm. We dicovered that four-bit, as well as a eastimation method which based on the back packets, and an index algorithm in LQE, has good hardware independence. Its evaluation of the success rate of 90% or more in some experiments. But it also has some defects, such as inaccurate estimates when it responds to emergencies, packet retransmission, and affected by the control factor, etc. This subject aims to improve the four-bit algorithm to implement a link quality assessment algorithm based on the existing theoretical and experiment research. Making four-bit more accuracy and less energy consumption when dealing with instantaneous link status. We have achieved an improved link quality estimation algorithm, making the data accutate and reliable transmission in the WSN, and leading the WSN a long-term and steady running. This improved estiamtion method, we can be achieved in the actual network deployment.At the same time, this article presented a link quality based code dissemination protocol. To improve the performance of code dissemination protocol in lossy networks, we leverage link quality between nodes to calculate the impact of each node. To reduce the overhead of control plane messages, we use the periodically broadcasted ADV messages to maintain the state of neighboring nodes. The results show that our protocol outperforms Deluge in terms of network completion time and message transmissions especially in the case of poor network quality. We have implemented the EER based on TinyOS and evaluated its performance using TOSSIM. Compared to Deluge(the wireless reprogramming protocol for TinyOS), the EER obtains3%-11% of performance improvement in terms of completion time and data traffic.