Wearable Computing

6Agent‐Oriented Body Sensor Networks

6.1 Introduction

Many computing paradigms have been to date exploited to support modeling and implementation of wireless sensor networks (WSNs) and, more specifically, of body sensor networks (BSNs). As widely discussed in Chapter 2, different kinds of paradigms, from low level to high level, can be used to develop WSN‐based systems. Among such paradigms, the most notable ones are event‐driven programming [1], data‐based models [2], service‐oriented programming [3], macro‐programming [4], state‐based programming [5], and agent‐oriented programming [6]. This chapter proposes the agent‐oriented paradigm for the modeling and implementation of BSNs. After introducing background concepts on the agent‐computing paradigm and, specifically, on software agents in the WSN context, the chapter discusses motivations and challenges on the exploitation of agents for BSNs and provides a description of the related state‐of‐the‐art. We then present agent‐based modeling and implementation of BSNs. A case study is finally proposed that uses two well‐known agent‐oriented platforms (JADE and MAPS) to develop an agent‐based real‐time human activity recognition system.

6.2 Background

6.2.1 Agent‐Oriented Computing and Wireless Sensor Networks

Software agents are defined as being networked software entities or programs that can perform specific (even complex) tasks for a user and having a degree of intelligence that allows them to carry out parts of their tasks/activities ...