PRESENTATIONNAME-智能与网络化系统研究所.ppt

1、Introduction（引言） In this paper we propose a practical approach for units localization in a typical scenario where some nodes, e.g., the one sat the network border, are used as anchors (i.e. they are placed in known positions ). All the RSSI values of the packets exchanged among nodes at different power levels are collected (RF mapping phase) and processed both to build the ranging model to be fed into a centralized Minimum Least Square (MLS) algorithm. The ranging model is calibrated from the online collected values, by selecting the optimal approximation family on the specific environment And normalizing the intensity of the received power. 本文我们为放置了一些节点的典型环境提出一种实际的节点定位方法。首先在网络的边界放置一些锚节点（它们的位置是已知的），把在不同能量水平之间的节点交换到的所有数据包的rssi值收集起来，然后建立一个带有集中最小二乘法的距离模型。距离模型根据选择最优近似family和规范接受能量的强度来实现对在线采集到的rssi值进行校准。 Our contribution can be summarized as a practical self localization system which does not Need the development of costly deployments of thenodes; a ranging model derived from calibrated RSS Information. 我们的系统可归纳为 一个不需要建立大量分布节点的实用自身定位系统 一个可以校准rssi值的距离模型 2、Localization scenario（定位场景） 1.RF mapping of the network: it is obtained by conveying short packets at different power levels through the network and by storing the average RSSI value of the Received packets in memory tables (we suggest to consider 5-10 values for estimating the RSSI ) 2.creation of the ranging model: all the tuples recorded Between two anchors are processed at the central Unit to compensate the nonlinearity and calibrate the model. 3.centralized localization algorithm: an optimization Problem is solved and provides the position of the nodes. 我们包括以下不同阶段的定位步骤 网络的RF地图：它可以通过在网络不同能量等级之间传送短的数据包和在存储表内存储接受包rssi的平均值（我们建议考虑5-10个值来对rssi估算） 距离模型的建立：对两个锚节点间以组的形式纪录在中心单元的数据的处理来补偿非线性和校准模型 集中定位算法：解决系统优化的问题和提供节点的位置 3、Generation of the RF attenuation modelRF 衰减模型的建立 1.the approximation function family. While in indoor Environments developing an attenuation model is extremely difficult and we have to rely on empiric