Research On Improved Localization Algorithm Based On RSSI For Wireless Sensor Network In 3D

The appearance of Wireless Sensor Network (WSN) caused widespread concern around the world. As the research focus of current information field,it involves many disciplines and mutiple technologies,including network communication protocol, core support technology, self-organization and management, development and application. With the rich core support technology, the WSN constitutes a specific application-oriented network system. The localization technology, as one of the core support technologies,is the basis of all applications. Because, the sensor data is no sense, without the location information.According to different classification criteria, the localization algorithms in WSN have different classification results. Based on the need to get the distance or angle among the nodes information by the physical measurement, location algorithms can be divided into two categories:range-based localization algorithm and range-free localization algorithm.The research shows that the former can achieve high location precision.Therefore, in this thesis, range-based localization algorithm is focused on.Major tasks of the thesis are as follows: (1) The research status and development trends of wireless sensor networks are introdued briefly. In-depth research on the knowledge and basic technique of nodes localization of WSN is given, and also a detailed description, classification, analysis and comparison of the staple algorithms at present are made to sum up their advantages and disadvantages.(2) Ranging methods used commonly are analyzed and compared, including: infrared, ultrasonic, GPS, and RSSI, etc. The localization algorithm based on RSSI distance is focused on, without any additional hardware to support, the distance between sending node and receiving node is computed by using the strength of the wireless receiver signal.The localization algorithm based on RSSI provides the most inexpensive method.So the localization algorithm based on RSSI is used more often.However,in the practical applications, the complexity of the signal model caused by multipath and non line of sight, makes the RSSI ranging great instability,causing a wide range of positioning errors. To solve the above problem, locality principle is proposed to reduce the instability in this thesis.It is that the attenuation coefficient of anchor's location is obtained by communicatbing between anchors.(3) At present, most of the location algorithms are based on two-dimensional. In practice, the distribution of nodes is often in the three-dimensional space complexity environment. So the algorithm in two-dimensional plane will have many problems. Therefore, the localization algorithm in three-dimensional space is very necessary. In this thesis,the RSSI ranging is proposed to extend the practical application of three-dimensional environment for the problem of application scope.(4) For the estimated location, there maybe an obvious error due to objective and subjective reasons.It is proposed in this paper to filter the error location.When many estimated locations are more concentrated, the few are relatively far away from the central location.The few ones can be considered as unreliable and divided,others are considered as reliable to improve the location accuracy.(5) Finally, the localization performance of improved algorithm is shown by simulation experiments.Using Simulation tool MATLAB 7.1,RSSI-based localization algorithm in two dimensions (RSSI-2D), RSSI-based localization algorithm in three dimensions (RSSI-3D), improved RSSI-based location algorithm in three-dimensional space (IRSSI-3D) were simulated. It is shown by the experiments that the IRSSI-3D algorithm proposed in this thesis has high localization precision and wide applications.