A multiclassifier approach for topology-based WiFi indoor localization

摘要: People localization is required for many novel applications like instance proactive care the elders or people suffering degenerative dementia such as Alzheimer's disease. This paper introduces a new system in indoor environments. It based on topology-based WiFi signal strength fingerprint approach. Accordingly, it robust, cheap, ubiquitous and non-intrusive which does require neither installation of extra hardware nor prior knowledge about structure environment under consideration. The well-known curse dimensionality critically emerges when dealing with complex task turns into high dimensional classification task. Therefore, core proposed framework fuzzy rule-based multiclassification system, using standard methodologies component classifier generation bagging random subspace, along logic to deal huge uncertainty that characteristic signals. Achieved results two real environments are encouraging, since they clearly overcome those ones provided by nearest neighbor matching algorithm, usually considered baseline localization.