Home

Research Programme 2 (RP2)

RP2 - Platforms
Programme Leaders: Dr. Cian O'Mathuna and Prof. Dermot Diamond

Context and Objectives

This research programme incorporates outputs from the research programmes on Devices and Materials, and Autonomic Sensor Communities in providing platforms, specifically more sophisticated sensors, miniaturised wireless nodes/motes and the production of assemblies of devices that will in turn feed into the demonstrators. Wireless sensor module platforms have, over the last 5 years, begun to proliferate with an associated maturing of software (VM, OS & Protocols). However, this maturation has not been mirrored by hardware developments, and at present a number of key challenges exist in achieving the large-scale, reliable and robust deployment of application-specific demonstrators. Furthermore, little effort has focused on the miniaturisation of motes with a view to seamlessly embedding them within objects, or the environment, or wearable fabrics. Advanced packaging and integration technologies, developed at TNI, will drive the hardware research in both WP2.1 Nodes and Networks and WP2.3 Body Sensor Networks. These will be based on bare silicon die assembly on thin flex circuits, as well as 3D stacking using flip-chip assembly and folded flex, and interconnects to embedded sensors. Sustainable node power also constitutes a major challenge that has to date not been addressed in a coherent fashion, combining both hardware, software and network issues in an holistic manner. At the hardware level, ultra low power data acquisition, processing and communications must be combined with sleep/wake-up modes to maximise power availability. The power supply system, which will typically combine energy scavenging, battery and power converters, also needs to be considered in terms of available energy, lifetime and size. On the software and networking side, it is critical to establish feedback to the node level whereby node functionality can be influenced to minimise power usage through modification of the frequency and size of data communication packets. The Body Sensor Network arena provides a significant challenge because of the need for a lightweight, unobtrusive entity that can be embedded in garments or fabrics or molded to the body shape. The ideal solution is to completely embed the sensing capability within the fabric/textile (including wearable chemo/bio-sensing capabilities) such that the BSN becomes totally innocuous to the wearer. A major interest for a number of companies within the consortium (Vodafone, Ericsson & Critical Path) is to identify the role that can be played by the existing mobile phone infrastructure in providing a high performance, widely distributed communications platform into which the wireless motes can be integrated.

Work Packages

WP 2.1: Nodes and Networks

In this work package two key strands of work will be undertaken. Firstly, novel sensors, developed within the RP1 Devices and Materials programme, will be integrated with the existing form factor of Tyndall motes informed by the needs of our key demonstrator targets. This will be achieved with a strong focus on power awareness which will encompass application-specific energy harvesting/ scavenging with microbatteries and storage capacitors. Secondly, we will investigate the miniaturization of the 3-dimensional mote form-factor by progressing from a 25mm cube developmental platform to a 1cm cube and ultimately to the ’intelligent seed’ in the range 5mm to 1mm on a side. A key challenge will be the interfacing (including signal conditioning and interconnects) of a wide variety of information sources such as chemo/biosensors and microanalytical instruments (RP1), physical transducers like accelerometers, gyroscopes, magnetometers and media capture devices (RP1) onto these platforms.

WP 2.2: Scalable Sensor Systems

Software for scalable sensor systems must be highly adaptive, as it must cope with a large and dynamically-changing population of components and services, coupled with a variety of sensors and information sources delivering partial and uncertain results. Simultaneously, it must deliver an overall experience which is adaptive to changing context but stable enough to present a predictable and scrutanisable service to users and other systems. Meeting these challenging goals involves considerable infrastructural development, and requires the designer to understand a wide range of quite subtle issues, all of which interfere with the development of application-level services. We propose to consolidate the research outputs from RP3 (Autonomic Sensor Communities) in a systems platform. This development will be driven by the personalization and retrieval work (at the application level) and device characteristics and power awareness (at the hardware level). This platform will differ from other sensor systems platforms in a number of key respects. For example, it will be completely standards-based, using RDF as its data exchange model and ZeroConf for resource discovery. It will support a knowledge-centric model of interaction where clients’ actions are driven by queries and triggers about the context of the system. It will employ gossiping to maintain a consistent state across an ad-hoc distributed data structure, which maximises robustness and scalability, and avoids many problems with hot-spots and hot-paths in communications. Finally, it will treat all information sources uniformly as sensors acting as inputs to uncertain reasoning algorithms.

WP 2.3: Body Sensor Networks (BSNs)

In this work package, a platform will be developed that incorporates on-body sensors for monitoring a variety of physical parameters including location, movement, posture, temperature, and personal health parameters such as respiration, heart rate, along with a system for harvesting and wirelessly transmitting this information that integrates motes and mobile phones. The aim, wherever possible, will be to completely embed the sensing capability within the fabric/textile (including wearable chemo/bio-sensing capabilities for targets like pH, lactate in sweat during exercise, and detection of external threats like the presence of toxic/hazardous gases such as CO) so that the BSN becomes totally innocuous to the wearer. For example, sensors will be integrated with vests, wearable pads and runners/shoes. A specific early goal will be to integrate on-body sensed information generated via a standardized wearable vest (Foster-Miller) with mobile phone platforms (Vodafone) to provide an platform for integration of new sensing modalities throughout the research programme. These outputs will feed into Demonstrator 1 (Personalized Health), which will be focused on the applications of wearable sensors for monitoring personal health status. The evolution of the BSN infrastructure will draw on the outputs of RP1 Devices and Materials and in particular, the Functional Molecular Materials and Sensing Devices work packages. The outputs of this workpackage will provide an important primary information source for RP3, RP4, RP5, and RP6 in that we will, through our industry partners, be in a position to organise experiments with significant scale (10’s to 100’s of sensored vests) early in the research timeline (by month 12). We will therefore have a powerful testbed generating real data for rigorous evaluation of primary research ongoing in the information focused research programmes (RP3, RP4, RP5 & RP6).

Novelty

The Body Sensor Network (BSN) research will focus where possible on adding sensing capabilities through modification of the fabric structure e.g. coating with conducting polymers or doped elastomers that convert the base fabric into a ’soft’ strain gauge while still retaining the inherent properties of the fabric. Biochemical measurements will be restricted to ex-vivo scenarios, targeting for example sweat during physical exercise, given the difficulties of performing in-vivo measurements in an innocuous wearable format [1]. We have already developed and filed a patent on a flexible, all-polymeric ultra-low power, fluidic handling manifold that can be integrated with fabrics and wearable biochemical sensors [2] to provide sophisticated capabilities like calibration, which is critical, but very difficult to integrate in an innocuous manner into a wearable format. In addition, through close cooperation with our industry partners (Humana, Foster-Millar), our goal is to have relatively large scale trials happening with groups of up to 100 or more people wearing pre-commercial sensored vests in exercise physiology trials (Demonstrator 1), which to our knowledge has not happened previously. Integration of these novel sensing and fluid handling capabilities with flexible electronics and scaled-down mote configurations, will provide an integrated sensing and communications environment that is, to our knowledge, unique in a global context.

Reference:

[1] B. A. Patel, C. A. Anastassiou, and D. O’Hare, “Biosensor design and Interfacing,” in Body Sensor Networks, G.-Z. Yang, Ed. Springer, 2006.
[2] K. Lau, S. Coyle, Y. Wu, G. Wallace, and D. Diamond, “An All-Fabric Flow Analysis System (Irish Patent Application filed January 2007),” 2007.