Talk on Approximate Computing & Its Research Potential

Knowledge Sharing Session

Date/Day : 26 March 2015/Thursday
Time       : 12.00 noon until 2.00 p.m
Venue     : The Cubes, Block A, Faculty of Computer Science & IT, University of Malaya
Speaker  :  Dr. Rehan Hafiz

Title: 
Approximate Computing & Its Research Potential

Abstract:

Power & energy is increasingly a first-order concern in computer systems. Approximate computing is an emerging computing paradigm that exploits energy-accuracy trade-offs. This talk shall discuss the various levels at which approximate computing is being explored. There are a number of challenges however when designing reliable systems based upon approximate computing. These include: Isolating & identifying parts of a program that must be precise, defining the correct matrices for accuracy and exploiting the application domain for error resilience.

BIODATA

Dr. Rehan Hafiz is an Assistant Professor (HEC Approved Supervisor)
Department of Electrical Engineering
School of Electrical Engineering and Computer Science (SEECS)
National University of Sciences and Technology (NUST)
NUST Campus H-12, Islamabad


Research Interest:

My research revolves around algorithm development and their digital implementations. Some of my research projects can be broadly classified as:

• Computer Vision / Digital Image / Video / Signal Processing
• Video Mosaics, Image registration, Video Surveillance
• Immersive Environments & Interactivity
• FPGA based system design
• Digital balanced detection for multichannel optical tomography
• VHDL system development for real time Image registration
• Algorithm development
• Game simulation for battlefield simulation
• Analysis of equation generation for GSM A5/2 encryption
• Real time noise suppression algorithm for Laser based spectroscopic and Tomographic application

Specialization:
Digital Image Processing, Computer Vision, Digital System Design