• Corpus ID: 10145152

Probabilistic Data Management for Pervasive Computing: The Data Furnace Project

@article{Garofalakis2006ProbabilisticDM,
  title={Probabilistic Data Management for Pervasive Computing: The Data Furnace Project},
  author={Minos N. Garofalakis and Kurt P. Brown and Michael J. Franklin and Joseph M. Hellerstein and Daisy Zhe Wang and Eirinaios Michelakis and Liviu Tancau and Eugene Wu and Shawn R. Jeffery and Ryan Aipperspach},
  journal={IEEE Data Eng. Bull.},
  year={2006},
  volume={29},
  pages={57-63}
}
The wide deployment of wireless sensor and RFID (Radio Frequency IDentification) devices is one of the key enablers for next-generation pervasive computing applications, including large-scale environmental monitoring and control, context-aware computing, and “smart digital homes”. Sensory readings are inherently unreliable and typically exhibit strong temporal and spatial correlations (within and across different sensing devices); effective reasoning over such unreliable streams introduces a… 

Figures from this paper

An adaptive RFID middleware for supporting metaphysical data independence
TLDR
This paper details MDI-SMURF, a Radio Frequency Identification (RFID) middleware system that alleviates issues associated with using RFID data through adaptive techniques based on a novel statistical framework.
PEEX : Extracting Probabilistic Events from RFID Data
TLDR
PEEX is a system that enables applications to easily define, extract, and manage meaningful probabilistic highlevel events from low-level RFID data by using a declarative query language, thus enabling applications not only to detect events but also manage them further as necessary.
Probabilistic RFID Data Management
TLDR
It is demonstrated, through experiments with real RFID traces collected on a small antenna deployment, that PEEX significantly improves event detection rates compared with deterministic techniques, and provides applications a flexible trade-off between event recall and precision.
Event Processing in Sensor Streams
TLDR
This chapter surveys state-of-the-art research on event processing in sensor networks, and provides a broad overview of major topics in sensor event streams.
Probabilistic Inference over RFID Streams in Mobile Environments
TLDR
A probabilistic model to capture the mobility of the reader, object dynamics, and noisy readings is proposed and can offer 49\% error reduction over a state-of-the-art data cleaning approach such as SMURF while also being scalable and efficient.
A probabilistic reasoning framework for smart homes
TLDR
This paper presents an ongoing effort to build a generic probabilistic reasoning framework for the networked homes that can be utilized for designing smart agents in a systematic and unified way.
Efficiently managing uncertain data in RFID sensor networks
TLDR
This paper proposes a novel framework to effectively manage the uncertainty of RFID data in large scale traceability networks and proposes a Markov-based model for tracking objects globally and a particle filter based approach for processing noisy, low-level RFIDData locally.
Probabilistic Modeling of Streaming RFID Data by Using Correlated Variable-duration HMMs
TLDR
A probabilistic model is presented, specifically Correlated Variable-Duration Hidden Markov Models (CVD-HMMs), to capture uncertainty and correlations of locations of tagged objects and can infer object locations from raw RFID streams.
Capturing Data Uncertainty in High-Volume Stream Processing
TLDR
A data stream system that captures data uncertainty from data collection to query processing to final result generation, and employs probabilistic modeling and inference to generate uncertainty description for raw data, and a suite of statistical techniques to capture changes of uncertainty as data propagates through query operators.
Managing and Mining Sensor Data
TLDR
Managing and Mining Sensor Data is a contributed volume by prominent leaders in this field, targeting advanced-level students in computer science as a secondary text book or reference, and practitioners and researchers working in the field will also find this book useful.
...
...

References

SHOWING 1-10 OF 17 REFERENCES
Approximate Data Collection in Sensor Networks using Probabilistic Models
TLDR
This paper proposes a robust approximate technique called Ken that uses replicated dynamic probabilistic models to minimize communication from sensor nodes to the network’s PC base station, and shows that Ken is well suited to anomaly- and event-detection applications.
Model-Driven Data Acquisition in Sensor Networks
A Pipelined Framework for Online Cleaning of Sensor Data Streams
TLDR
Extensible receptor Stream Processing (ESP) is presented, a declarative query-based framework designed to clean the data streams produced by sensor devices.
Querying and mining data streams: you only get one look a tutorial
TLDR
In these situations, algorithms that can summarize the data stream involved in a concise, but reasonably accurate, synopsis that can be stored in the allotted (small) amount of memory and can be used to provide approximate answers to user queries along with some reasonable guarantees on the quality of the approximation are needed.
I Sense a Disturbance in the Force: Unobtrusive Detection of Interactions with RFID-tagged Objects
TLDR
The paper catalogs the experimental results obtained, provides plausible models and explanations and highlights the promises and future challenges for the role of RFID in ubicomp applications.
Probabilistic reasoning in intelligent systems - networks of plausible inference
  • J. Pearl
  • Computer Science
    Morgan Kaufmann series in representation and reasoning
  • 1989
TLDR
The author provides a coherent explication of probability as a language for reasoning with partial belief and offers a unifying perspective on other AI approaches to uncertainty, such as the Dempster-Shafer formalism, truth maintenance systems, and nonmonotonic logic.
Relational Dynamic Bayesian Networks
TLDR
RDBNs are a generalization of dynamic probabilistic relational models (DPRMs), which are an extension of dynamic Bayesian networks (DBNs) to rst-order logic and two new forms of particle ltering are proposed.
Trio: A System for Integrated Management of Data, Accuracy, and Lineage
TLDR
This paper provides numerous motivating applications for Trio and lays out preliminary plans for the data model, query language, and prototype system.
Inside the Smart Home
Conceptions of the Home.- Inside the Smart Home: Ideas, Possibilities and Methods.- Conceptions of the Home.- Smart Homes: Past, Present and Future.- Households as Morally Ordered Communities:
Answering Queries from Statistics and Probabilistic Views
Systems integrating dozens of databases, in the scientific domain or in a large corporation, need to cope with a wide variety of imprecisions, such as: different representations of the same object in
...
...