Using stigmergy to build pervasive computing environments
Citation:
Peter Barron, 'Using stigmergy to build pervasive computing environments', [Thesis], Trinity College Dublin. School of Computer Science and Statistics. Computer Science, 2005, pp286Download Item:

Abstract:
Pervasive computing looks beyond the age of the personal computer to a time when everyday devices will be embedded with technology and connectivity. The goal of pervasive computing is to make such devices available throughout the physical environment to support people's usual activities with timely interventions without overwhelming them with inappropriate responses. So far, a number of research e_x001B_orts have investigated di_x001B_erent approaches to managing the complexities of developing these types of environments with varying degrees of success. The di_x001E_culty still remains as to how to develop a pervasive computing environment that can support the integration and organisation of devices and applications in a spontaneous and robust manner. The problem is partly attributed to the highly dynamic and unpredictable nature of these types of environments, and is often further hampered by the limited resources found on devices. The technology for pervasive computing is reaching a point where it is possible to convert may everyday environments into interactive spaces. Typically, these spaces have been designed from the ground up to support the anticipated needs of users, and are usually pre-installed and maintained over the period during which they are in use. Conceptually, these e_x001B_orts are centralised in their approach, in that, e_x001B_orts are focused around coordinating the resources of a particular geographical location in meeting the demands of users. However, in the future, the construction of pervasive computing environments is more likely to evolve accidentally from physical spaces as technology is incorporated into the space over time. This suggests there is a need to support the assembly of pervasive computing environments in a more ad-hoc fashion. To address this point this thesis presents a highly-decentralised method of organising the components of a pervasive computing environment, supporting spontaneous interaction between entities and providing robust system-wide behavior. The inspiration for this work stems from nature and the observations made by the French biologist Grass̩ on how social insects coordinate their actions using indirect communication via the environment, a phenomenon that has become known as stigmergy. In the stigmergic approach there are fewer dependences between entities allowing for the incremental construction and improvement of solutions without adversely e_x001B_ecting the rest of the pervasive computing environment. This thesis encapsulates this approach in a model that is used to underpin a framework for pervasive computing. A prototypical implementation of the model is provided by Cocoa. Cocoa supports the use of stigmergy to build self-coordinating environments that promote the autonomy of entities. Designed to both support and complement the use of stigmergy, the framework employs a distributed architecture organised in a peer-to-peer fashion. To ease the implementation and deployment of entities Cocoa supports a programming abstraction encapsulated in a high-level scripting language. The scripting language exploits the methodologies used by social insects to construct a society of autonomous entities capable of responding to the environment in a stigmergic manner. In order to validate the contribution of the thesis a select number of application scenarios from a range of di_x001B_erent domains have been implemented using the Cocoa framework. The evaluation is used to demonstrate how a model based on stigmergy can be used to provide a highly-decentralised method of organising the components of a pervasive computing environment. In addition, the evaluation shows how such an approach can support the spontaneous interactions of autonomous entities and provide robust system-wide behavior.
Author: Barron, Peter
Advisor:
Cahill, VincentQualification name:
Doctor in Philosophy (Ph.D.)Publisher:
Trinity College Dublin. School of Computer Science and Statistics. Computer ScienceType of material:
ThesisCollections:
Availability:
Full text availableKeywords:
Computer ScienceLicences: