Article Preview
Top1. Introduction
Distributed systems have been used as a platform to allow the interaction between groups of individuals and a set of devices. As technology advances in sensing, computation, storage and communications become widespread, ubiquitous sensing devices will become a part of global distributed sensing systems (Campbell et al., 2008; Lane et al., 2010).
Recently, the predominance of mobile phones equipped with sensors, the explosion in social networks and the deployment of sensor networks have created an enormous digital footprint that can be harnessed (Zhang, Guo, Li, & Yu, 2010). Furthermore, developments in sensor technology, communications and semantic processing, allow the coordination of a large network of devices and large dataset processing with intelligent data analysis (Lane et al., 2010).
The sensing of people constitutes a new application domain that broadens the traditional sensor network scope of environmental and infrastructure monitoring. People become the carriers of sensing devices and both producers and consumers of events (Miluzzo et al., 2008). As a consequence, the recent interest by the industry in open programming platforms and software distribution channels is accelerating the development of people-centric sensing applications and systems (Lane et al., 2010; Miluzzo et al., 2008).
To take advantage of these emerging networks of mobile people-centric sensing devices, researchers arrived at the concept of Mobiscopes, i.e. taskable mobile sensing systems that are capable of high coverage. They represent a new type of infrastructure, where mobile sensors have the potential to logically belong to more than one network, while being physically attached to their carriers (Abdelzaher et al., 2007). By taking advantage of these systems, it will be possible to mine and run computations on enormous amounts of data from a very large number of users (Lane et al., 2010). People-centric sensing therefore enables a different approach to sensing, learning, visualizing and data sharing, not only self-centered, but especially focused on the surrounding world. The traditional view on mesh sensor networks is combined with one where people (carrying sensors) turn opportunistic coverage into a reality (Campbell et al., 2008). These sensors can reach into regions whereas static sensors cannot, proving to be especially useful for applications that occasionally require sensing (Abdelzaher et al., 2007). By employing these systems, one can aim to revolutionize the field of context-aware computing (Zhang et al., 2010).
By leveraging the behavioral patterns related to individuals, groups and society, a new multidisciplinary field is created: Social and Community Intelligence (SCI) (Zhang et al., 2010). Real-time user contributed data is invaluable to address community-level problems and provide universal access to information, contributing to the emergence of innovative services (Campbell et al., 2008; Lane et al., 2010; Zhang et al., 2010), such as the prediction and tracking of epidemic outbreaks across populations (Zhang et al., 2010). Thus, technological benefits are shifted from a restricted group of scientists to the whole society (Campbell et al., 2008).
Such systems can be applied to Healthcare, to facilitate both monitoring and sharing of automatically gathered health data (Campbell et al., 2008). Epidemics are a major public health concern and it has been shown impact can be reduced by early detection of the disease activity. For instance, it has been shown that the level of influenza-like illness in regions of the US can be estimated with a reporting lag of one day, when compared to clinical methods whose results take a week to be published (Zhang et al., 2010).
As most people possess sensing-enabled phones, the main obstacle in this area is not the lack of an infrastructure. Rather, the technical barriers are related to performing privacy and resource respecting inference, while supplying users and communities with useful feedback (Lane et al., 2010).