A deadline-driven epidemic data collection protocol suitable for tracking interpersonnel rendezvous

Avranil Tah, University of Texas at El Paso

Abstract

This thesis describes a peer-to-peer wireless data collection algorithm that uses epidemic communication to propagate time-sensitive sequentially sampled data records from sensors toward infrastructure connected upload stations via mobile data mules. These records are labeled with sequence numbers and delivery deadlines, and are transmitted in sequential order. Delivery deadlines enable transmission prioritization and trigger alarms warning of violations. The sequential ordering of records simplifies the protocols transmission-control and garbage collection mechanisms: only two monotonically increasing scalar sequence indices associated with a particular sensor must be exchanged between peers prior to selecting which records need to be communicated. One of these indices also serves as an anti-entropy message, indicating which records are known to already have been uploaded by an infrastructure-connected upload station, thereby indicating eligibility for garbage collection. This data collection algorithm is used to implement a prototype inter-personnel rendezvous reporting system potentially useful for tracking the spread of contagious disease. Radios used to transfer information among peers also serve as proximity sensors. Records are created whenever a peer discovers another. The data collection algorithm transports these records to infrastructure-connected upload stations responsible for their storage and on-line processing. My contributions include refinement, implementation, and initial testing of the algorithms for data collection and rendezvous reporting.

Subject Area

Health care management|Computer science

Recommended Citation

Tah, Avranil, "A deadline-driven epidemic data collection protocol suitable for tracking interpersonnel rendezvous" (2010). ETD Collection for University of Texas, El Paso. AAI1483985.
https://scholarworks.utep.edu/dissertations/AAI1483985

Share

COinS