EPSRC WINES Proposal EP/C547586/1

BiosensorNet: Autonomic Biosensor Networks for Pervasive Healthcare

Morris Sloman, Guang-Zhong Yang, Oliver Wells, Chris Toumazou, Tony Cass,
Danny O’Hare, Eric Yeatman, Ara Darzi, Magdi Yacoub

Imperial College London

Overview

Healthcare is coming under increasing pressure to improve the quality of care delivered to patients through effective prevention and post-operative care. This comes at a time when there is a need to curtail growth in healthcare spending fuelled by ageing populations, and the prevalence of obesity, diabetes, cancer and chronic heart and lung diseases. The emphasis on sensor based measurements provides site-specific information from the body and in particular dynamic and quantitative differences between vascular and tissue compartments, thus generating entirely new data sets on which clinical decisions are based. Development in miniaturised wireless biosensors is set to reshape the common practice in clinical medicine especially for the prevention of terminal illness, monitoring the progression of chronic disease, and assessing post-operative care and body reaction to complex therapeutic drug regimes. Thus far, new sensor technologies are increasingly being developed based on protein arrays and ion-sensitive FETs. These, together with established sensing technologies, are being used as critical sensing elements in the development of pervasive computing/sensing systems. These new sensors will be used within future ubiquitous healthcare systems to monitor patients as they maintain their normal everyday activities, in order to warn patients or healthcare workers of problems as well as collecting data for trend analysis and medical research. The use of continuous monitoring circumvents the drawbacks of conventional diagnostics and monitoring (generally limited to brief time points and frequently unrepresentative physiological states or artificially introduced exercise tests), allowing both transient and progressive abnormalities to be reliably captured. This will be combined with new drug therapy and targeted delivery, minimally invasive intervention, and novel vessel prostheses. The role of intelligent, context-aware sensor systems is to provide automated or semi-automated tools to support interventional management systems with detailed and co-ordinated information that can improve significantly the essential data available to the human observer, especially in accommodating problems of discontinuous data and data artefacts. This will enable better diagnosis and care of people (especially the ageing population), improved diagnostic technologies (more reliable, quicker, less restricting), and more reliable remote support of patients who are able to continue to live normal lives at home and at work. Both short and longer term applications have been identified, including the monitoring of post-operative patients and those on complex therapeutic drug regimes (e.g. chemotherapy); patients with chronic disease (e.g. diabetes, heart or lung disease); and mental health patients whose behaviour in the community is dependent on their compliance with their drug regimes.

This project will develop a new generation of intelligent biosensing networks that can
i)
integrate local analogue signal processing with ultra-low power sensor interface and wireless data path;
ii)
recognise the environment and physical context within which the signal is sensed;
iii) form an autonomic system capable of self-configuring a network of sensors to provide reliable long-term adaptive sensing by fusing error-prone signals from individual sensors.

We have assembled a multi-disciplinary team from Computing, Biomedical Engineering, Electronic Engineering and Medicine with the expertise needed to lay the foundations for a new generation of intelligent, self-managing, context-aware biosensor networks for pervasive healthcare. The project will address issues throughout the signalling pathway, to establish a practical platform for in vivo sensing, and provide a generic autonomic sensing architecture. Although this proposal relates to the healthcare context, autonomic sensing technology has a broad scope of applications ranging from environment, modern manufacturing, food processing, chemical process monitoring, condition-based maintenance of complex systems, battlefield surveillance and reconnaissance, and transport. In all these applications, an "autonomic" self-configuring network of sensors will underpin developments by recognising the environment and physical context within which the signals are sensed, and providing reliable long-term adaptive sensing by fusing error-prone signals from individual sensors with a redundant sensor array. The development of novel power scavenging and management with in situ analogue processing on the node itself will have a significant impact on future wireless sensor networks design.

Background

The UbiCare Centre: Ubiquitous Computing for Healthcare in the Community is based at Imperial College London and consists of a number of collaborative projects funded by DTI and Industry (www.ubicare.org). Other universities involved are Southampton and Lancaster. The industrial partners include Cardionetics, Central Data Control, Medtronic, Orange, Telewest, Tyco UK, Toumaz Technology, Docobo, Huntleigh Technology, Celoxica. The UbiMon project (www.ubimon.org) has been developing a novel miniature wireless Body Sensor Node (BSN) compatible with Berkeley motes. It uses the TI MSP430 16-bit ultra low power RISC processor with 64KB +256B Flash memory, 12-bit ADC and 6 analog channels (connecting up to 6 sensors). The BSN has a throughput of 250kbps and uses U.C. Berkeley’s TinyOS – a small, open source and energy efficient sensor board operating system. It has been used for real-time context aware ECG monitoring, integrated both with GPRS and WiFi wireless for remote monitoring and data interrogation.

Allied to the UbiMon project, significant effort has been invested by the research team in making the sensor nodes sufficiently adaptable and reconfigurable to support self-management, which is essential for both clinical and homecare deployment. Component and communication errors and failures must be made transparent to users so the system must be self-healing. The ANS project in UbiCare has researched into more adaptable operating systems for BSN and the EPSRC Amuse project has investigated policy-based adaptation for self-managed cells [5]. For addressing Trust, Privacy or Security issues, which are important to Healthcare applications, we have developed the Sultan Toolkit for specifying and analysing trust relations [1][4]. This provides support for human trust-based decisions and permits trust based on recommendations or vice versa. The trust specification included risk and experience as constraints. The design supports simple authorisation policies which queried the Trust management system for access control based on trust levels but the system was too heavyweight for automated trust decision support as the emphasis was more on the analysis of trust specifications. The PEACE system [6] is a policy-based framework for the establishment, evolution and management of mobile ad-hoc communities; its ideas are applicable to intelligent sensors forming a BSN. It defines the policies for setting up the community, assigning entities to roles within the community and how roles interact. We have developed the Ponder policy toolkit for security and management policy that also supports role based management and is used by many other institutions [5].

The focus of our current work on UbiMon is related to Cardiac monitoring with standard wearable sensors with integration to existing Medtronic implantable Reveal devices. Extension to implantable sensors and actuators will need to address new power generation techniques, wireless communication suitable for in-body transmission and biocompatibility issues. The UbiMon project has developed some very promising techniques both in terms of hardware design and context aware sensor fusion however, fundamental research is needed in intelligent sensor node design, micro-fabrication, power scavenging, and new biosensing paradigms closely coupled with bionics. The work on energy scavenging micro-power supplies at Imperial started in 2001 as a new collaboration between the Devices and Power Electronic Groups at the EEE Department funded by EPSRC. This collaboration has been fruitful and helped to establish Imperial’s position in the international Power MEMS community. This initial project resulted in the first comprehensive analytic framework in the field [15]. We have pioneered a nonlinear class of device which, unlike most other reported devices, is suitable for low frequency applications such as in or on the body. This has led to the first demonstration of working prototypes suitable for powering by body motion [14]. We have also investigated the efficiency of these and other architectures when driven by actual human motion [16], and have developed new electronic device designs suitable for the unique demands of the associated power electronics for this application [17]. We are now seeking to extend the basic work to examine improved devices at higher power levels, and further integration with electronics and with sensing systems more broadly [18]. In addition, we have also developed miniature fuel cell power generators at the Bioengineering Department which are about to be commercialised.

In biosensing, Toumazou’s group has been the first to demonstrate weakly inverted ISFETs operating in their nanopower region for real-time biochemical reaction monitoring [19]. This provides a platform for fast, chip-based algorithms for everyday solving of clinical problems which can be defined by the physician. In cardiology, this can be used to assess adequate perfusion from biochemical parameters for the prediction of likely deterioration in patient outcome for early physician intervention. In renal medicine, this provides a means of continuous intelligent monitoring of blood analytes for optimisation of dialysis therapy and prediction of end stage renal failure. In Intensive Care Units, the method offers easy, non-invasive or minimally-invasive methods of reliable blood chemistry monitoring on demand, particularly blood gases. At Imperial, we have also developed sensors for chemical species for the study of perfusion and oxygenation, processes which underpin a wide variety of disease processes and these have found application in studies of myocardial infarction [8], dynamic cardiomyoplasty [9][10]. In addition to tissue nutrition and metabolism, signalling molecules such as nitric oxide [11] and the neurotransmitters can now be measured routinely. Generic approaches to low cost mass manufacture, using for example conducting composites, are well advanced [12] and are the subject of two patents. The wealth of data generated by these devices and the complexity of their interactions requires novel signal processing and pattern recognition using algorithms capable of dealing with the inherent non-linearity [13] techniques under development by the research team.

Objectives and research Issues

This proposal assembles a multidisciplinary team with the overall objective to lay the foundations for a new generation of intelligent, self-managing, context-aware biosensor networks for critical control of human health. Specific objectives include:

The above objectives are very challenging and there are a number of research issues that will need to be addressed:

Approach and Work Packages

The proposed work for the development of autonomous sensing networks will be implemented in three stages:

  1. The development of power management, network protocols and system architecture. The work will be based on the on-body wireless sensor networks developed by the UbiMon project based on existing sensor technology.

  2. The incorporation of wireless elements and processing in the integrated sensors. Minimally-invasive sensors for the key chemical parameters that characterise tissue nutrition deficiencies, which underpin a large number of disease processes (stroke, myocardial-infarction, rheumatoide arthritis). Stable and robust sensor technology exists in the research team for dissolved oxygen, pH, pCO2, glucose and lactate. The key issue to be addressed is the integration of analogue processing (current to voltage conversion, filtering) and wireless elements on the sensor nodes. We will use well-established stable sensors as paradigms for the development of low-power integrated devices, which is expected to have broader applicability in wireless sensing scenarios.

  3. The development of sensor networks for both short term and long-term monitoring. We will investigate wireless sensor network issues related to managing redundancies in sensor arrays, the characterisation of tissue sensor interactions and the use of novel biocompatible materials. The use of built-in redundancy overcomes many of the difficulties in in vivo sensing and statistical sensor fusion and principal component analysis techniques will be used to identify outliers for elimination prior to pooling sensor outputs. Long-term recording can thus be achieved without requiring advances in materials science. A major advantage of developing low-power wireless sensing array protocols is that they permit simultaneous assessment of tissue heterogeneity.

To realise the strategies mentioned above and the key objectives defined for the proposed project, the following workpackages will be used.

WP1 Bio Sensing Technology and Bionics:

The key aim of WP1 is to develop sensors suitable for integration into micro-power wireless sensor nodes, particularly for continuous monitoring applications, using previously demonstrated bio-sensor types and methods. Integration with analogue signal conditioning circuits will underlie work across the various types of sensors to be investigated. Circuit design involves input stages for parameters described above; processing stages (e.g. calculation of critical ratios, use of Henderson-Hasselbach equation to assess blood buffer concentration); redundancy for failure tolerance; and compensation for drift and temperature variation. Matching biochemical reactions with non-linear physics of ISFETs will allow real-time data processing directly within the sensor element itself. Integrated circuits will be prototyped using foundry services, and encapsulated for in-vitro testing in biofluids.

pH and pCO2

Glass and polymer membranes suffer from extreme fragility, slow response and poor biocompatibility. Sensing elements with electrolytically-generated iridium oxide films offer super-Nernstian performance and compatibility with C-MOS processing. We will develop protocols for the construction of arrays of iridium oxide pH sensors with low power integrated analogue signal processing. With 2-3 additional processing steps, these devices can form the basis for Severinghaus-style pCO2 sensors.

Dissolved oxygen, Ions (K+, Na+, Ca2+, Mg2+, Cl-, PO43-)

Arrays of Clark-type oxygen sensors have been widely reported for dissolved oxygen and standard selective membranes exist for these widely assessed blood electrolytes. We will combine these approaches with front-end low-power signal processing for current to voltage conversion, signal averaging and outlier elimination for improved lifetime.

Glucose and lactate

Miniaturised enzyme electrodes for both glucose and lactate have been based on wild type proteins (either oxidases or dehydrogenases), we will improve on these by engineering the relevant enzymes to improve their immobilisation characteristics and dynamic response range. The ready availability of X-ray crystal structures and the ease of mutagenesis and expression make this approach technically feasible and experimentally tractable. The primary advantage of engineering sensing proteins is that they can be adapted to the particular sensing requirements.

 

WP 2 Intelligent Sensor Architecture & Power Management:

The aim of this WP is to develop an innovative architecture for sensor nodes which can host a wide range of sensor functions, while minimising size and energy use. This work will be carried out in close interaction with WP1. Current autonomous wireless sensor platforms are dominated in size and weight by battery packs, which also introduce an unacceptable maintenance burden. We will approach this challenge in two directions: reduction of power consumption, and development of alternative supplies. For energy supply, we will investigate energy scavenging solutions, particularly from body or organ motion and vibration. We have already published world-first demonstrations of early prototype motion scavenging devices suitable for biomedical applications. Here we will develop higher energy, more compact devices (target size 0.01 cm3) by exploiting MEMS integration techniques. Our initial devices, based on electrostatic principles, show that micron-scale dimensional control, and elimination of electrical parasitic elements and unwanted mechanical modes, are the keys to achieving high performance. Integration of these electromechanical devices with the overall node electronics will also be a main task. In addition we have in commercial development a mm-scale biofuel cell based on glucose oxidation. Power output for the as yet unoptimised device is around 1 m W. This can be used as an alternative power source in other applications.

To reduce power consumption, we will develop new solutions for data processing and transmission functions, in particular to eliminate the need for power-hungry digital microprocessors. Continuous-time current-mode algorithms based upon log-domain processing (where precision can be traded for ultra-low power) and switched current-processing (where higher precision is required) will allow data conditioning and reduction (by extracting key parameter values). Data reduction will help minimise transmitter power. Novel configurable interface algorithms will be created using the MOSFET in weak-inversion (standard MOSFETS or Ion-sensitive input devices such as Chem-FETS and ISFETs), requiring no more than a few nanowatts of power consumption. We will aim to integrate WP2 sensor platforms with ultra-wide band micro-power transmitters currently under development within our consortium.

WP3 Context Aware Body Sensor Networks and Robust Feature Extraction:

For the proposed sensing environment, effective sensor fusion and statistical feature reduction is crucial to the wireless sensor arrays for both built-in redundancies and tissue heterogeneity. When the BSN is used for the monitoring of patients under normal physiological status, the contextual information is important to the capture of clinically relevant episodes. The purpose of this workpackage is to investigate multisensor fusion, context awareness and robust feature extraction. Reliable detection of patient activity normally requires the use of a large number of sensors around the body, leading to a significant burden on the power consumption, hardware and bandwidth requirement. The lifetime of a wireless sensor depends mainly on its power source, and the majority of the energy is consumed by the wireless transceiver. Central to this workpackage is the design of an autonomous Bayesian-net context sensor that can be integrated with the sensor nodes developed in WP1 and WP2. For general long term sensing environments, reducing the transmission range and required bandwidth will greatly reduce the power consumption and prolong the life span of the sensor and in general, the transmission range of implantable sensors for the proposed project is expected to be 30-100mm. Clustering data transmission within the local neighbourhood can also alleviate the problem of data collision. For this WP, we will develop a novel concept of a distributed Bayesian network, which adopts the hierarchical nature and causal structure of Bayesian networks onto a body sensor network in order to reduce the data communication and distribute the processing power effectively in the multi-hop network environment. For robust feature extraction based on the proposed sensor array framework, we will develop a robust sensor fusion and feature reduction technique coupled with lightweight wireless technologies and protocols for sensor network communications, and architectures to provide reconfiguration and adaptation of sensors in response to context.

WP 4 Autonomic Management:

In linking with WP3, this workpackage will investigate the autonomic management issues for the deployment of body sensors in normal clinical environments. The objective is to introduce high-level knowledge combined with the awareness of the local context of each sensor to reliably determine the patient’s activity and the best deployment of different sensor nodes. This has to be optimised in relation to power management and accuracy of the overall sensing output. This strategy would adapt to a patient’s current activity - sleeping, walking, running as well as to changes in the patient’s medical conditions. By introducing self-healing and management based on the intrinsic redundancies built into the sensor networks, the system will cater for error prone sensor readings and adapt to component failures. Within this workpackage, we will also perform a trust and security threat analysis to determine the basic security mechanisms required. Although a BSN is essentially a single trust domain, it must also cater for introduction of new devices when healthcare workers treat the patient. The interactions with remote monitoring services and healthcare workers require trust-based access control and authorisation mechanisms as a healthcare worker may not be known in advance. We envisage monitored information being used for medical research, so anonymisation techniques must be incorporated for data mining to permit privacy. These have to be resilient to inference based attacks and should also be combined with role based access control. We will also investigate whether recommendations and trust adaptation based on experience are relevant and if so how they should be incorporated.

WP5 Clinical Demonstrator and Evaluation:

For this project, the clinical demonstration of the proposed sensing architecture will be focused on ubiquitous monitoring of post-operative surgical inpatients as well as patients require long-term monitoring particularly those with progressive heart failure. This currently requires much manpower and as a result it is impossible for surgical patients to occupy general ward beds. The sensors will be used to measure recognised physiological and biochemical parameters to assess patient well-being. On-body sensor networks based on wireless nodes will be used together with established commercially available wireless technologies to achieve this goal. Measurements of some of these parameters will be continuous, whereas others will be at much shorter intervals than currently possible. Once this system is established, sensor reliability will be tested, followed by validation against currently available patient monitoring techniques. The monitoring will be integrated with inferencing of adverse events and an alarm messaging system to notify medical staff of problems. Finally, implementation onto a simulated surgical ward environment will allow direct comparison with existing patient monitoring technologies to evaluate ease of configuration by non-computing staff, reliability and effectiveness. One of the important aspects of this workpackage is to investigate the use of the proposed wireless sensing technology in the hospital environment and its associated regulatory issues and potential interference to certain medical devices. We will be relying on close collaboration with our industrial partners in addressing as well as following relevant standards and guidelines. We will also integrate new sensors into the existing UbiMon ambulatory cardiac monitoring network. Note all the RAs will contribute to the demonstrators and evaluation in this WP.

WP6 Project Management:

The proposed project involves an interdisciplinary team of researchers in complementing areas of biosensing and wireless technologies. The management committee, consisting of workpackage leaders, will meet formally every 3 months to review progress and suggest revisions in research strategy. One unique advantage of the project is that all the groups involved are at the same institution which makes close ad-hoc interaction and collaboration much simpler. We will employ Mr. Oliver Wells as part time (20%) project manager. The proposed project involves a complex mixture of technical areas and effective project management and coordination is critical to its success. Mr Wells is currently project manager on the DTI funded UbiCare Centre at Imperial, and has 25 years industrial and large project management experience. He will be responsible for making sure progress reports are produced, organising monthly internal meetings (to review progress, scientific challenges and adjust work package priorities), arranging project workshops (6 monthly internal workshops and an open workshop every 12 months), liaising with industry and negotiating additional industrially funded projects. The internal workshops will also be attended by industrial collaborators.

Related Work and Relevant Projects

The EU FP6 Healthy Aims project is developing body wireless networks to support implantable cochlear, retina, electrical muscle stimulators, pressure and motion sensors (www.healthyaims.org). The focus is on shorter-term commercial products based on existing sensors rather than longer term intelligent sensors. We are discussing the possibility of associating with the project in order to collaborate more closely and avoid duplicate effort. Autonomic computing work has been focussed on large-scale utility computing and web services (www-03.ibm.com/autonomic/), although there is increasing interest in self-management for pervasive computing with some workshops being started on this topic. UPnP [7] is one of the best known small scale self-configuration approaches but it is also based on comparatively heavyweight web services protocols. Most of the existing work on Trust and privacy are in the context of large scale web-services such as P-Grid [1]or internet based peer-to-peer systems e.g. [2] which advocates the use trust service brokers. Conoise-G (www.ecs.soton.ac.uk/~sr2/ConoiseG/research.html) aims to use trust and reputation models for agent-based virtual organisations where participants must be accredited, and their activities policed. The Gold project (www.neresc.ac.uk/projects/GOLD/projectdescription.html) is developing tools and techniques for deploying dynamic virtual organisations, but the application focus is the Chemistry industry. The development of the general wireless sensor architecture is related to the work at Berkeley and EU SMART-ITs. However, our emphasis is unique in that it addresses sensing, power management/scavenging, miniaturisation, and low power processing for body sensor networks. The EU SECURE project (secure.dsg.cs.tcd.ie) has integrated trust and quantified risk with access control for pervasive environments. Imperial has one of the leading activities internationally on energy scavenging micro-generators. There is also work in the UK based at Southampton University, mainly on piezoelectric generators [20]; this group participates in the EU VIBES project on energy scavenging. Internationally, MIT [21], Berkeley [22], IMEC and the Australian National University are the other leaders. Generally, devices reported elsewhere have insufficient power levels, and high or narrow frequency ranges, making them unsuitable for biomedical use.

Relevance to Beneficiaries

The development of new sensors will significantly benefit the care of patients after surgery and those who are chronically ill. The proposed sensor networks will permit early discharge of patients and allow continuous monitoring afterwards under their normal daily routines. The monitored information can be logged, and analysed to detect trends, with possible early warning feedback to patients. For chronic disease, the analysis over wider populations can determine what changes occurs that later result in serious conditions such as diabetes and chronic heart failures. The proposed sensing architecture is also expected to transform how the efficacy of certain surgical procedures is evaluated. For example, the ability to seal a vessel or to create anastomosis is one of the central issues of surgery. Advances in surgical technique, in particular that of laparoscopic surgery, have led to the introduction of new instrumentation and one such system is the ligasure vessel sealing system that uses a unique combination of energy and pressure. The amount of thermal spread and associated tissue injury is of considerable importance in a primary anastomosis where structure or leakage may be invoked and in the case of vessel sealing ischaemia may be induced, leading to significant weakening and possible rupture. Currently, little is understood about the architecture of the union between the tissues except the mechanical bonding of different tissue layers. The miniaturised wireless sensing array with built-in redundancies and tissue heterogeneity will permit reliable assessment of the intrinsic healing process which can lead to improved application of these new surgical procedures. The BiosensorNet project is also expected to have an impact on how future clinical trial data is collected, as it provides an effective way of monitoring patient compliance and detailed changes in physiological responses.

Central to this project is integration of sensor node with in situ data processing and low power wireless data-paths. The result derived from the project is expected to be valuable to the future advances in pervasive sensing technologies. Thus far, power consumption, bandwidth requirement, and local processing power are major concerns in wireless sensing networks. The strategy adopted by this project allows overall balance of these constraints, and with the development of power scavenging techniques based on normal body motion, the framework is expected to have important value to the future design of long-term implantable wireless sensor networks. Furthermore, the development of autonomous context aware sensing and the effective incorporation of security and policy will benefit the general development of pervasive computing techniques, particularly related to those that require wearable sensor networks. The collaborators involved in this project are expected to adopt the main concepts in future developments. For example, Medtronic are already the leading suppliers of implantable cardiac monitors and defibrillators and improved design in sensing and determination of context with minimal power consumption is expected to significantly improve the accuracy as well as the robustness of their future devices. Finally, the work proposed is complimentary to that going on in the Faraday Medical devices partnership and GE are also looking for concepts and techniques for future patient monitoring systems.

Project Plan and Deliverables

 

Work

Package

Deliverable

Month

Title

WP1

D1

6

Initial sensor arrays design

 

D2

12

Prototype integration of sensor arrays with existing BSN nodes

 

D3

18

Prototype sensor signal conditioning, and redundant array management

 

D4

24

Prototype fabrication of low-power front end integrated analogue processing

 

D5

32

Revised design fabrication for final demonstrators

WP2

D6

6

Initial generic intelligent sensor architecture specification

 

D7

12

Final architectures specification

 

D8

18

Initial protoype MEMs + UWB wireless, Design for power management

 

D9

24

Initial fabrication of MEMs + UWB

 

D10

28

Prototype integration with biosensor

 

D11

32

Revised design fabrication for final demonstrator.

WP3

D12

6

Optimisation of power consumption by balancing processing and transmission

 

D13

12

Prototype context aware sensor node with distributed Bayesian context sensing

 

D14

18

Optimisation of processing and power

 

D15

24

Distributed Bayesian context sensing, sensor fusion and light weight protocols

 

D16

28

Prototype context sensor node

 

D17

32

BSN node + context sensing for demonstrators

WP4

D18

6

Trust & security threat analysis

 

D19

12

Initial self configuration

 

D20

18

Prototype support for adaptive trust and security for on body and remote interaction

 

D21

24

Prototype integrated security + self management

 

D22

32

Revised Secure, Self-configuring set of body sensors with remote logging

WP5

D23

6

Requirements analysis of post operative patient demonstrator + other scenarios

 

D24

12

Sensor platform and requirement validation with designs from WP1 and WP2 + Demo spec.

 

D25

24

Post-operative initial prototype and cardiac monitoring demonstrator

 

D26

28

Initial evaluation report

 

D27

32

Revised post-operative demonstrator

 

D28

36

Final demos, evaluation reports (covers WP1-4) + Regulatory conformance report

References

  1. Aberer K: P-Grid: A self-organising access structure for P2P information systems. Proc of the 6th Intl. Conf. on Cooperative Information Systems (CoopIS 2001), Trento, Italy 2001.
  2. Atif Y.: Building Trust in E-Commerce, IEEE Internet Computing, pp18-24. Jan-Feb 2002
  3. Grandison T. and Sloman M.: Specifying and Analysing Trust for Internet Applications, Proc. of the 2nd IFIP Conf. on E-Commerce, E-Business and E-Government, Oct 7 - 9, 2002, Lisbon, Portugal.
  4. Grandison T. and Sloman M.: Trust Management Tools for Internet Applications, Proc. 1st Int’l Conf. Trust Management, LNCS 2692, Springer-Verlag, 2003, pp. 91–107.
  5. Damianou N., Dulay N., Lupu E. and M. Sloman M.: The Ponder Policy Specification Language, Proc. Policy 2001: Workshop on Policies for Distributed Systems and Networks, Bristol, UK, pp29-31 Jan. 2001
  6. Keoh S.L., Lupu E., Sloman M.: PEACE : A Policy-based Establishment of Ad-hoc Communities, Annual Computer Security Applications Conference (ACSAC 2004), Tucson, Arizona, USA, Dec. 2004
  7. Universal Plug and Play Device Architecture http://www.upnp.org/resources/documents.asp
  8. D.S. Fluck, P.J. E. Etherington, D. O’Hare, C.P. Winlove, D.J, Sheridan ‘Myocardial tissue perfusion determined by particulate and diffusible tracers during ischaemia: what is measured?’ Cardiovasc. Res. 32 (1996) 869-878.
  9. A.R. Greenbaum, P.J.E. Etherington, S. Manek, D. O’Hare, K.H. Parker, C.J. Green, J.R. Pepper & C.P. Winlove ‘Measurements of oxygenation and perfusion in skeletal muscle using multiple microelectrodes’ , J. Muscle Research & Cell Motility, 18 (1997)149-159.
  10. Greenbaum AR, Jarvis JC, O'Hare D, Manek S, Green CJ, Pepper JR, Winlove CP, Salmons S ‘Oxygenation and perfusion of rabbit tibialis anterior muscle subjected to different patterns of electrical stimulation’ J. Muscle Research & Cell Motility 21(3) (2000) 285-291.
  11. E. Leung, P.J. Cragg, D. O’Hare & M. O’Shea ‘A novel in vivo nitric oxide sensor’ Chem. Comm. (1996) 23-24.
  12. O’Hare D., Macpherson J.V., Willows A. ‘On the microelectrode behaviour of graphite-epoxy composite electrodes’ Electrochemistry Communications 4 (2002) 245-50.
  13. M. Arundell, B.A. Patel, M.S. Yeoman, K.H. Parker & D. O’Hare "Hilbert transform of voltammetric data" Electrochem. Comm. 6 366-372 (2004)
  14. Mitcheson P.D., Miao P., Stark B.H., Yeatman E.M., Holmes A.S., Green T.C. "MEMS electrostatic micro-power generator for low frequency operation", Sensors and Actuators 115, 523-529 (2004).
  15. Mitcheson P.D., Green T.C., Yeatman E.M., Holmes A.S. "Architectures for vibration-driven micro-power generators", IEEE/ASME J. Microelectromech. Syst. 13 (3), 429-440 (2004).
  16. T. von Buren, P. D. Mitcheson, T. C. Green, E. M. Yeatman, A. S. Holmes, G. Troster, "Optimization of Inertial Micropower Generators for Human Walking Motion", IEEE Sensors Journal, accepted.
  17. Stark B., Mitcheson P, Miao P., Green T.., Yeatman E., Holmes A.S. "Power processing issues for micro-power electrostatic generators" Power Electronics Specialist Conference (PESC ’04), Aachen, Germany, June 21-24, 2004, pp. 4156-4162.
  18. E.M. Yeatman, "Advances In Power Sources For Wireless Sensor Nodes", 1st International Workshop on Body Sensor Networks, London, April 6-7, 2004
  19. Shepherd, L. and Toumazou, C., "Towards real time biochemical analysis using weakly invested MOSFETs", accepted for publication in Sensors and Actuators, Part B.
  20. P. Glynne-Jones, S.P. Beeby, E.P. James, N.M. White, "The modelling of a piezoelectric vibration powered generator for microsystems", in: Proceedings of the 11th International Conference on Solid-State Sensors and Actuators, Transducers ’01 and Eurosensors XV, Munich, Germany, 2001, pp. 46–49.
  21. S. Meninger, J.O. Mur-Miranda, R. Amirtharajah, A.P. Chandrakasan, J.H. Lang, "Vibration-to-electric energy conversion", IEEE Trans. Very Large Scale Integr. VLSI Syst. 9 (1) (2001) 64–76.
  22. S. Roundy, P.K. Wright, K.S. Pister, "Micro-electrostatic vibration-to-electricity converters", in: Proceedings of the 2002 ASME International Mechanical Engineering Congress and Exposition, New Orleans, LA, USA, 2002.