Range-Based Localization in Wireless Networks Using Density-Based Outlier Detection

• 作者: Aaron Gulliver Khalid K. Almuzaini
• 来源期刊: 无线传感网络(英文)
• 学科: 医学
• 关键词: LOCALIZATION POSITIONING Ad HOC Networks Range-Based Wireless Sensor Network OUTLIER Detection CLUSTERING
• 年，卷（期）: 2010,（11）
• 页码范围: 807-814
• 页数: 8页
• 分类号: R73

摘要

Node localization is commonly employed in wireless networks. For example, it is used to improve routing and enhance security. Localization algorithms can be classified as range-free or range-based. Range-based algorithms use location metrics such as ToA, TDoA, RSS, and AoA to estimate the distance between two nodes. Proximity sensing between nodes is typically the basis for range-free algorithms. A tradeoff exists since range-based algorithms are more accurate but also more complex. However, in applications such as target tracking, localization accuracy is very important. In this paper, we propose a new range-based algorithm which is based on the density-based outlier detection algorithm (DBOD) from data mining. It requires selection of the K-nearest neighbours (KNN). DBOD assigns density values to each point used in the location estimation. The mean of these densities is calculated and those points having a density larger than the mean are kept as candidate points. Different performance measures are used to compare our approach with the linear least squares (LLS) and weighted linear least squares based on singular value decomposition (WLS-SVD) algorithms. It is shown that the proposed algorithm performs better than these algorithms even when the anchor geometry about an unlocalized node is poor.