Raffaele Gravina’s research activity is centered on the Wireless Body Sensor Networks (a specific class of the Wireless Sensor Networks) and on the Internet-of-Things.
Wireless Sensor Networks (WSN)
Wireless Sensor Networks (WSNs) are currently emerging as one of the most disruptive technologies enabling and supporting next generation ubiquitous and pervasive computing scenarios. However, programming WSN applications is a complex task due to the hard constraints of the wireless sensor nodes in terms of limited resources (computing power, memory, and communications) and to the lack of proper and effective software abstractions. The research activity focuses on the development of new methodologies, frameworks and techniques for the rapid development of WSN applications in important application domains such as health-care, energy building monitoring, social interactions.
Domain Specific Frameworks and Systems for Heterogeneous Wireless Body Sensor Networks
The design of health care systems based on WSNs is complex, particularly due to the challenge of implementing intensive signal processing algorithms for data interpretation on highly resource limited wireless nodes, and have to meet hard wearability and battery duration requirements. Furthermore, debugging software on sensor nodes is very difficult and time consuming due to the lack of support architecture in embedded operating systems; redeploying the debugged code on the actual sensing devices takes significant amount of time as well. In such context, an important result of the research activity is the SPINE (Signal Processing In-Node Environment) framework, jointly developed with the Telecom WSN Lab in Berkeley, for the rapid prototyping of systems based on heterogeneous Wireless Body Sensor Networks (WBSNs) which require an efficient in-node processing and effective classification algorithms at the base station side. The experimentation of SPINE has been carried out both to evaluate efficiency and effectiveness of the developed frameworks and to build a real-time system based on wearable sensors for the human activity monitoring. Furthermore, A SPINE-based research prototype of a system for collaborative detection of the handshaking gesture, using a single wrist-attached body sensor node per person has been jointly developed with the TKN group of TUB (Germany) in the framework of the CONET project. Moreover, a new macroprogramming language for task-oriented distributed computing in WSNs has been defined and its related run-time architecture, both generic and optimized for nesC/TinyOS, has been implemented. Such language and architecture will form the basis for SPINE 2.0, the future under-development release of the SPINE framework.
Body Sensor Networks – Cloud integration
In the coming years, BSNs are likely to be exploited to allow for implicit social interaction among people who can exchange private/public information through their worn BSN nodes when they come into contact. BSNs of co-located people can be also utilized as a mobile sensor infrastructure to support other context-aware applications such as disaster, medical emergency and mass event management. In such contexts, management of a large number of disparate BSNs as well as cooperative BSNs to support various applications will be a crucial issue to deal with. Moreover, the huge amount of data that a BSN is able to deliver, demands a powerful, scalable storage and processing infrastructure to perform both online and offline analysis of BSN data streams.
The combination of BSN, with their huge amount of gathered sensor data and their limited processing power can be tackled by exploiting a Cloud computing infrastructure to offer an integrated platform that provides: (a) the ability to utilize heterogeneous sensors; (b) scalability of data storage; (c) scalability of processing power for different kinds of analysis; (d) global access to the processing and storage infrastructure; (e) easy sharing of results; and (f) pay-as-you-go pricing for using BSN services.
Cardiac Defence Response detection
The cardiac defense response (CDR) refers to the idea that organisms react physiologically to the presence of danger or threat. This reactivity has a protective function, as it provides the basis for adaptive behaviors such as the “fight-or-flight” response. This response is the first stage of a sequence of internal processes that prepares the aroused organism for struggle or escape, therefore to react to threats priming for fighting or fleeing. However, if the CDR is maintained for long periods, it may result in health risks, degrading the physiological response to anxiety. Excessive physiological reactivity is one of the main causes of emotional stress and other psychological disorder. Therefore, it is important to identify the CDR mechanism and provide clinicians with a valuable tool that could be used to study the psychological state of the subject.
The electrocardiogram (ECG) is the standard method for measuring the electrical and functional activity of the heart. Traditionally, the ECG is used to diagnose cardiovascular diseases and rhythm abnormalities. Recently, the ECG has been used for emotion recognition and detection of stress. The ECG signal is an ideal signal to study changes due to physiological responses when emotion or other external factors occur. The advantage of using the ECG signal for detecting basic emotions is that a person can be monitored using non-invasive wearable cardiac sensors. In contrast, facial recognition methods are more invasive because they require the placement of electrodes and cameras to detect subtle changes in the person’s face.
Agent-oriented Middleware for WSN
Mobile agents are a distributed computing paradigm based on code mobility that has already demonstrated high effectiveness and efficiency in IP-based highly dynamic distributed environments. Due to their intrinsic characteristics, mobile agents may provide more benefits in the context of WSNs than in conventional distributed environments. A research outcome is MAPS (Mobile Agent Platform for Sun SPOT), an innovative Java-based framework for wireless sensor networks based on Sun SPOT technology which enables agent-oriented programming of WSN applications. The MAPS architecture is based on components which interact through events. Each component offers a minimal set of services to mobile agents which are modeled as multi-plane state machines driven by ECA rules. In particular, the offered services include message transmission, agent creation, agent cloning, agent migration, timer handling, and easy access to the sensor node resources (sensors, actuators, input switches, flash memory, and battery). A performance evaluation of MAPS has been carried out by computing micro-benchmarks, related to agent communication, creation and migration. Moreover MAPS has been also used to build a real-time system based on wearable sensors for the human activity monitoring.