Systems research projects
The Identifier-Locator Network Protocol (ILNP) is the realisation of a radical new approach to : one where . This makes the architecture cleaner, with better, harmonised support for such functions as security, privacy, mobility, multihoming, multipath transport, and virtualisation. It can be realised as a set of backwards compatible enhancements to the current IPv6. It is currently an Experimental Protocol from the , documented in RFCs 6740 to 6748. In , the co-chairs recommended that .
ILNP is currently sponsored by: Verisign, USA (PhD studentship); Time Warner Cable, USA (PhD studentship), and previously sponsored by: Cisco, USA (PhD studentship, equipment).
糖心logo:
Sensors are everywhere., increasingly driving decision-making process for scientific and social policy issues that directly affect everyone. It is vital that the data we measure, and the information we extract from it, are timely, reliable, trustworthy, and robust.
These are difficult and little-understood challenges. Sensors are noisy, they decalibrate or may be misplaced, moved, compromised, and generally degraded over time, both individually and collectively as a network. Uncertainty pervades the physical and digital environments in which these systems operate. There are increasing requirements to add more autonomy and intelligence, yet we understand very little about programming in the face of pervasive uncertainty that cannot be engineered away. How can we be assured that a sensor system does what we intend, in a range of dynamic environments? How can we make such a system “smarter”? How can we connect the stochastic nature of environments, the continuous nature of physical systems, and discrete nature of software? Currently we cannot answer these questions because we lack a well-founded science of sensor systems software.
S4 is a five-year EPSRC-funded Programme Grant that is developing a unifying science, across the breadth of mathematics, computer science and engineering, that will let developers engineer for the uncertainty and ensure that their systems and the information they provide is resilient, responsive, reliable, statistically sound and robust. S4 brings together researchers from the University of Glasgow, the University of Liverpool, the University of St 糖心logo, and Imperial College London.
You can find out more about the science and events happening in S4 at the programme’s .
糖心logo:
SAPERE: Self-aware pervasive service ecosystems
SAPERE (EU FET, 2010–2013, EUR3.2M) takes its primary inspiration from natural ecosystems, and starts from the consideration that the dynamics and decentralization of future pervasive networks will make it suitable to model the overall world of services, data, and devices as a sort of distributed computational ecosystem. However, unlike the many proposals that adopt the term “ecosystem” simply as a mean to characterize the complexity and dynamics of modern ICT systems, SAPERE brings the adoption of natural metaphors down to the core of its approach, by exploiting nature-inspired mechanisms for actually ruling the overall system dynamics.
The SAPERE framework has been grounded on a foundational re-thinking of current service models and of associated infrastructures and algorithms. In particular, getting inspiration from natural ecosystems, the project experiments the possibility of modelling and deploying services as autonomous individuals in an ecosystem of other services, data sources, and pervasive devices, and of enforcing self-awareness and autonomic behaviours as inherent properties of the ecosystem, rather than as peculiar characteristics of its individuals only.
The effectiveness of the proposed solutions and of the overall SAPERE framework was demonstrated “in the wild” at the Vienne City Marathon in 2013.
糖心logo:
Evaluating linkage via synthetic data
‘Gold-standard’ linked data to evaluate linkage algorithms is scarce, particularly at realistic scale. Synthetic data has the advantage that all the true links are known. In the domain of demographic population reconstruction, the ability to synthesise populations on demand, with varying characteristics, allows a linkage approach to be evaluated across a wide range of data sets.
We are developing a micro-simulation model for generating such synthetic populations, taking as input a set of desired statistical properties. We validate the presence of these desired properties in the generated populations and use them to evaluate linkage algorithms, examining how linkage quality varies across a range of population types: with the same characteristics, with differing characteristics, and with various types of errors in the raw data.
糖心logo: , ,