A Distributed Architecture for Reasoning about a Higher-Level Context

Dargie, Waltenegus; Hamann, T.

doi:10.1109/wimob.2006.1696351

Cited by 3 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…, M, where c(l) is the lth MFCC. Finally, to reduce the effect of very low and very high MFCC components (at both edges of the Mel spectrum), the so-called liftering process 3 is performed. Equation (7) shows a typical liftering function.…”

Section: B Aggregatormentioning

confidence: 99%

“…The context-recognition process involves modeling frequency-domain audio features and building a Bayesian network. Similarly, Dargie and Hammann [3] use Bayesian networks to estimate the whereabouts of a mobile user. The Bayesian network models stochastic features of data taken from various sensors (humidity, temperature, and light).…”

Section: Introduction Cmentioning

confidence: 99%

“…The implementation of the architecture will be demonstrated for auditory-based context recognition. In earlier reports [3], [8], a part of the architecture (i.e., without the adaptation component) was implemented to recognize the whereabouts of a mobile user (rooms, corridors, or outdoor). The raw data were obtained from temperature, humidity, and light sensors.…”

Section: Introduction Cmentioning

confidence: 99%

See 2 more Smart Citations

Adaptive Audio-Based Context Recognition

Dargie

2009

IEEE Trans. Syst., Man, Cybern. A

Self Cite

View full text Add to dashboard Cite

Abstract-Context recognition is an essential aspect of intelligent systems and environments. In most cases, the recognition of a context of interest cannot be achieved in a single step. Between measuring a physical phenomenon and the estimation or recognition of what this phenomenon represents, there are several intermediate stages which require a significant computation. Understanding the resource requirements of these steps is vital to determine the feasibility of context recognition on a given device. In this paper, we propose an adaptive context-recognition architecture that accommodates uncertain knowledge to deal with sensed data. The architecture consists of an adaptation component that monitors the capability and workload of a device and dynamically adapts recognition accuracy and processing time. The architecture is implemented for an audio-based context recognition. A detail account of the tradeoff between recognition time and recognition accuracy is provided.

show abstract

Section: B Aggregatormentioning

confidence: 99%

Section: Introduction Cmentioning

confidence: 99%

Section: Introduction Cmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Audio-Based Context Recognition

Dargie

2009

IEEE Trans. Syst., Man, Cybern. A

Self Cite

View full text Add to dashboard Cite

show abstract

“…Often a piece of context primitive is not sufficient to appropriately model a real-world situation [10]. The real world is far too complex to be captured in complete detail by a single primitive.…”

Section: Aggregatormentioning

confidence: 99%

Dynamic Generation of Context Rules

Dargie

2006

Self-Managed Networks, Systems, and Services

View full text Add to dashboard Cite

Abstract. This paper presents a scheme for the dynamic generation of context rules which are useful for modifying the behaviour of mobile devices according to the social and physical settings of their users. Existing context-aware systems employ a pool of predefined rules which will be executed whenever a context of interest is sensed and captured. Defining rules at design time, however, has the following limitations: (1) The system designer should identify the set of context primitives which describe a context of interest as accurately as possible; (2) the various states of each context primitive should be predetermined and mapped to sensory data, which often requires experience or expertise; (3) the availability of mechanisms for capturing the context primitives is presupposed; if instead of the specified context primitives other context primitives are discovered, which may equally describe a similar situation, the system does not react to them, unless of course, all possible alternatives are foreseen; and (4) the desired behaviour itself may not be a priori known, as it is specific to the user. To facilitate a flexible use of context rules, they should be generated dynamically. The scheme proposed in this paper associates decision events -signifying the activities of a user -with a set of context primitives which are acquired at the time the decision events are produced. From the decision-context associations, context rules are generated. The Context-Aware E-Pad will be introduced to illustrate the scheme proposed.

show abstract

“…The first field focuses on capturing, modeling, reasoning, and representing context information [3]; and on employing context information to provide useful services [1], [2]. Where context is understood to be the dynamic aspects of users, devices, and systems, including the dynamic aspects of communication infrastructures and application servers.…”

Section: Introductionmentioning

confidence: 99%