Using Ontology and Complex Event Processing Engine for Human Activity Recognition in Ambient Assisted Living domain

As a result of the rising older people population, the Ambient Assisted Living (AAL) branch is growing up fast. Generally, the main goal of an AAL system is to help old people to live in their own houses longer and with an improved quality of life. In fact, its functionalities are based on the use of a set of different sensors interconnected by different types of communication systems to get information about the status of patients. The installed sensors network produce a set of data that shows a fine—grained nature, carrying generally their value, originating device, data type, timestamp and so on. For this reason, it is not always possible to have a clear overall view. Since it is often needed an analysis procedure which is able to take into account the semantic of records, ontologies are becoming widely used to describe the domain and to enrich the acquired data with its significance. In our research work, we propose a methodology arranged by two components integrated sequentially: an ontology [1] and a Complex Event Processing (CEP) [2] engine. The ontology has is built following a precise structure and it is able to describe the AAL domain, organize data according to their semantic meaning and select them (pre-processing phase). The main serious expressiveness limitation of OWL ontology is the lack of temporal reasoning, so in the framework it is introduced after ontology a CEP engine that is a technique concerned with timely detection of compound events within streams of simple events. Our challenge is to perform semantic queries on a data repository, whose records originate from a network of heterogeneous sources. The main goal of such queries is the pattern matching process, i.e. recognition of specific temporal sequences in fine—grained data. In our framework, benefits deriving from the implementation of a domain ontology are exploited i11 different levels of abstraction. Thereafter, reasoning techniques represent a pre—processing method that prepares data for the final temporal analysis. Our proposed approach will be applied to the ongoing AALISABETH [3], an Ambient Assisted Living project aimed to discover and manage the behavior of monitored users.