The Mobile Sensing Platform: An Embedded Activity Recognition System

  • Authors:
  • Tanzeem Choudhury;Gaetano Borriello;Sunny Consolvo;Dirk Haehnel;Beverly Harrison;Bruce Hemingway;Jeffrey Hightower;Predrag "Pedja" Klasnja;Karl Koscher;Anthony LaMarca;James A. Landay;Louis LeGrand;Jonathan Lester;Ali Rahimi;Adam Rea;Danny Wyatt

  • Affiliations:
  • Dartmouth College;University of Washington;Intel Research;Stanford University;Intel Research;University of Washington;Intel Research;University of Washington;University of Washington;Intel Research;University of Washington;Intel Research;University of Washington;Intel Research;Intel Research;University of Washington

  • Venue:
  • IEEE Pervasive Computing
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

The Mobile Sensing Platform (MSP) is a small-form-factor wearable device designed for embedded activity recognition. The MSP aims broadly to support context-aware ubiquitous computing applications. It incorporates multimodal sensing, data processing and inference, storage, all-day battery life, and wireless connectivity into a single 4 oz (115 g) wearable unit. Several design iterations and real-world deployments over the last four years have identified a set of core hardware and software requirements for a mobile inference system. This article presents findings and lessons learned in the course of designing, improving and using this system. This article is part of a special issue on activity-based computing.