Skip to main content
eScholarship
Open Access Publications from the University of California

Essence: Machine Learning Approaches to Scalable and Energy Efficient Sense-making for Internet-of-Things (IoT)

  • Author(s): Sarma, Santanu
  • Advisor(s): Dutt, Nikil
  • et al.
Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License
Abstract

This thesis presents an efficient and scalable sense-making framework using machine learning techniques for Internet-of-Things (IoTs) in order to understand users, contexts, and their environments to make meaningful decisions. The proposed sense-making IoT framework, called Essence, employs combination of participatory mobile crowdsensing along with infrastructure sensing to perform sense-making using machine learning techniques. While collaborative mobile crowdsensing enables information to be gathered and shared by users who are directly involved (participatory sensing) or integrated seamlessly as needed (opportunistic sensing) through user mobile platforms, the infrastructure sensing fabric of the Essence framework provides sense-making support for scenarios where mobile sensing platforms are inadequate. To address the scalability needs of the Essence framework, we employ dimensionality reduction techniques such as principal component analysis (PCA) based heterogeneous compressive sensing techniques for approximate gathering and processing of diverse sensor data. This requires new mechanisms for sensor data collection, tunable approximate processing and machine learning based decision making using a hierarchical networking architecture to create a compressive collaborative sense-making framework for IoTs. Essence uses our previous SenseDroid mobile sensing framework with machine learning enabled decision and actuation components for IoT paradigms. The Essence framework is build using a multi-tired hierarchical architecture for sensing spatial variations of a parameter of interest, perceive spatio-temporal fields, and enable energy efficient local mobile or infrastructure sensing with a small number of measurements. This approximate, yet tunable approach combines different sensing approaches opportunistically while trading scalability (and coverage) for data accuracy (and energy efficiency). We demonstrate the application of Essence sense-making framework in predicting the presence of West Nile Virus (WNV) in a region using both compressed sensing and Neural Networks (NN) as a case study. The thesis also discuss several challenges associated with sense-making approaches for emerging IoT applications.

Main Content
Current View