NFC-based Computation Offloading

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering

Final Year Thesis Oral Presentation

Title: "NFC-based Computation Offloading"

by

Ms. SUCIPTO, Kathleen


Abstract:

In recent years, consumers have placed a huge demand in more sophisticated 
smartphone applications and services which require higher battery 
consumption. The growth of smartphones' battery capacity, however, is not 
keeping pace with increasing power demands. As a solution, we propose an 
NFC-based computation offloading. We choose NFC as its short distance and 
operational characteristics offer better security, lower interference, and 
lower battery consumption. In this work, we build an NFC/HCE-based 
computation offloading framework that requires no tapping and enables 
multiple transmissions. First, we present the design of the framework that 
implements the Reader Mode Enabling-Disabling method, where one device 
enables the reader mode before the other disables its reader mode during 
the role switching process. Then, we present real results of the energy 
consumption and the time duration required to offload two computations: the 
RSA key generation and encryption, and the N Queens computations. Based on 
our evaluation, the overall time duration of our framework to offload those 
computations is considerably close with the time required for the offloadee 
device to run the computations locally. The overall energy consumption is 
also significantly low compared to local computations.


Date                 : 27 April 2016 (Wednesday)

Time                 : 2:00pm to 2:45pm

Venue                : Room 4504 (lift 25/26)

Advisor              : Dr. Pan HUI

2nd Reader           : Dr. Raymond WONG