A Framework for Real-time Context Monitoring in Sensor-rich Personal Mobile Environments

Project: Research project

Project Details

Description

A Framework for Real-time Context Monitoring in Sensor-rich Personal Mobile Environments A Framework for Real-time Context Monitoring in Sensor-rich Personal Mobile Environments The proposed work will develop enabling technologies for the emerging field of situation-aware, ambient decision making, which integrates sensing, data analysis, and response mechanisms into immediate surroundings. This will involve introduction of the novel concept of mining-workflows and implementation of a mining middleware which can support execution of multiple, complex mining-workflows. Data (whether captured in real-time through sensory measurements or processed, materialized, and stored for later use) is many times only statistically accurate. Thus, the middleware will provide mechanisms to efficiently deal with uncertain data or fuzzy fit between data sources and mining demands. Efficiently executing mining-workflows requires management of the inherent redundancy (i.e., the alternative ways of achieving a given mining goal) and imprecision of the data, observations, and processing units. Alternative ways of executing a particular mining task may result in different qualities (effectiveness) and costs (time and resource consumption). Unpredictability of the volume and the quality of the sensed data, the changes in the resource needs due to the mining-service evolution, as well as possible mobility of the data producers and consumers in the environment, necessitate the proposed system to adapt to the run-time situations and demand. Thus expected results of this work will include (a) description of sensing services consisting of adaptable or selfadaptive sensing, processing, and communicating components; (b) modular integration of workflow components and their run-time adaptation for providing end-to-end mining service guarantees; and (c) run-time evolution of mining-workflows based on varying system and application requirements. Complementary contributions in each of these domains will be brought together in an Ambient Information Mining (AIM) middleware.
StatusFinished
Effective start/end date6/1/098/31/13

Funding

  • OTHER: Foreign Other: $128,000.00

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.