Short Bio
I’m a PhD candidate at the University of Toronto supervised by Prof. Ashish Khisti. My special fields of interest include Machine learning, Probabilistic Modeling, Graph Signal Processing, and their application in fMRI Data Analysis. My previous line of research was focused on indoor localization and designing channel access codes on graph. I am currently working with the Kimel family translational Imagining-Genetics lab at CAMH with Dr. Aristotle Voineskos on applying ML to delineate the neural pathophysiology underlying impaired social cognition in people with Schizophrenia Spectrum Disorders (SSD).
I received my B.S. in Electrical Engineering in 2014, and my M.S. in Communication Systems in 2017 both from the University of Tehran, Iran. Previously, I worked at Sarveen Technologies as the head of Indoor Positioning Team, where I led the development of core algorithms to create a robust and adaptive localization solution. Prior to that, I was a research assistant at center for Wireless Multimedia Communications (WMC) at University of Tehran with Prof. Farshad Lahouti for four years, where I did my research on indoor localization, designing codes on graph and iterative random access schemes.
You can download my CV here.
Education
2018 – Present
University of Toronto
PhD Candidate - Electrical and Computer Engineering
Toronto, ON, Canada
- Advisor: Ashish Khisti, PhD.
- Average: 18.85/20 (Ranked first)
- Thesis title: Joint channel coding and medium access control in machine-to-machine communication. Defended (20/20)
- Advisors: Farshad Lahouti, PhD. Maryam Sabbaghian, PhD
- Average: 17.94/20
- Thesis title: Indoor Positioning System Using Wi-Fi Fingerprinting Method.
- Advisors: Farshad Lahouti, PhD.
Research & Work Experiences
2016 – 2018
Sarveen Technologies Inc.
Head of Indoor Positioning Team
Science and Technology Park, Tehran, Iran
Sarveen Technologies Inc. is a young but well-funded innovative company specializing in indoor positioning, activity recognition, and IoT technologies. At Sarveen, we focus on inventing the next generation of smart wearables and Internet of Things devices, starting from product concept, all the way to hardware, algorithm, software and venture design.
As the head of Indoor Positioning Team, I lead the development of core algorithms to create a robust and adaptive positioning solution used in a wide range of location-aware Sarveen products (such as smart shopping service, elderly care system, and livestock health monitoring). I design the necessary tools and software for data collection, performance assessment, and project setup automation. In addition, I have implemented numerous algorithms, from classification and clustering to decision fusion methods like particle filter, Viterbi, and BCJR. Furthermore, I actively cooperate in the venture design of InJust, a location-aware marketing service for shopping malls.
2010 – 2014
Center for Wireless Multimedia Communications (WMC)
Research Assistant
University of Tehran, Tehran, Iran
Under the supervision of Dr. Lahouti, I conducted research on two general fields: indoor positioning, and random access scheme design using factor graphs. During the Digital Venture Design course in WMC, we defined the business plan of a location-aware marketing tool for shopping malls, which later incorporated into a business product (InJust).
Publications
Time-Resolved fMRI Shared Response Model using Gaussian Process Factor Analysis
M. Ebrahimi, N. Calarco, K. Campbell, C. Hawco, A. Voineskos, A. Khisti arXiv preprint arXiv:2006.05572
Two-layer Coded Channel Access with Collision Resolution: Design and Analysis
M. Ebrahimi, F. Lahouti and V. Kostina IEEE Transactions on Wireless Communications (Volume: 19, Issue: 12, Dec. 2020)
Coded Random Access Mechanism for Communication Networks
F. Lahouti, V. Kostina, and M. Ebrahimi US Patent App. 16/362,567
Coded random access design for constrained outage
M. Ebrahimi, F. Lahouti and V. Kostina 2017 IEEE International Symposium on Information Theory (ISIT), Aachen, Germany
Presentations
2017 IEEE International Symposium on Information Theory (ISIT)
Aachen, Germany. (June 2017)
The First Toronto Workshop on Graph Spectral Machine Learning
Invited Talk, Ryerson University (August 2019)
Honors & Awards
:)
Ranked 1st among all communication system students
M.Sc degree, University of Tehran.
M.Sc. thesis nominated for the ECE school best dissertation award
University of Tehran, Tehran, Iran. (Winners TBA)
Excellent Student M.Sc. Admission Offer
B.Sc. degree, University of Tehran Entrance examination waived as an award for being among the top-10% students (Ranked 6th among 123)
Ranked 194th among 277,820 participants
In the nationwide university entrance examination in Mathematics and Physics fields for B.Sc.
Teaching Experience
CSC412: Probabilistic Learning and Reasoning
- University of Toronto, Winter 2021
- Teacher Assistant
- Instructor: Jesse Bettencourt
ECE421: Introduction to Machine Learning
- University of Toronto, Fall 2020
- Designing Programming Assn: CNN feature visualization using JAX
- Instructor: Prof. Nicolas Papernot
ECE421: Introduction to Machine Learning
- University of Toronto, Fall 2019, Winter 2020, Winter 2021
- Head Teacher Assistant
- Instructor: Prof. Ashish Khisti
ECE1504: Statistical Learning
- University of Toronto, Winter 2020
- Teacher Assistant
- Instructor: Prof. Ashish Khisti
CSC458: Computer Networking Systems
- University of Toronto, Fall 2018, Fall 2019
- Teacher Assistant
- Instructor: Dr. Sajad Shirali-Shahreza, Prof. Yashar Ganjali
Advanced Theory of Communications
- University of Tehran, Spring 2017
- Chief Teacher Assistant
- Instructor: Maryam Sabbaghian, PhD
Communication Systems II
- University of Tehran, Fall 2017
- Chief Teacher Assistant
- Instructor: Amir Masoud Rabiei, PhD
Wireless Communication
- University of Tehran, Spring 2016
- Chief Teacher Assistant
- Instructor: Ali Azam Abbasfar, PhD
Mathematics I
- University of Tehran, Fall 2012
- Teacher Assistant
- Instructor: Mohammadreza Kolahdouz, PhD
Selected Projects
01
Design and Implementation of an Adaptive Indoor Positioning System
Sarveen Technologies Inc. Using BLE/WiFi RSSI, accelerometer, magnetometer, gyroscope, and map information as inputs, the implemented algorithm derives functional features from each source and combines them together to cope with inaccuracies in the inputs. The algorithm adapts to the absence of any source and the variants between different devices and sensors. The algorithms are implemented both offline and online using Matlab and Java (Android).
03
Indoor Map Grid Generator
Using Matlab GUIDE Sarveen Technologies Inc.
This tool lets you define grid points on a map layout and generate connectivity between the points for navigation and indoor positioning purposes.
04
Design and Implement of a Wireless Setup for Text Communication
Using AVR micro-controller touchpad and RF module, Microprocessor course
05
Implementing a Verilog Code Parser Using C++
Object-Oriented Electronic Modeling course
06
Direction of Arrival Estimation
Using Maximum Likelihood method Detection and Estimation Theory course
07
Radio Access Technology Selection In Heterogeneous Networks
Using MADM method Cellular Networks course
08
Information Theoretic Analysis of Physical Layer Security in Wireless Networks
A secrecy graph approach Information Theory course (Sharif University of Technology)
09
Design, Simulation, Implementation and Measurement of 4.5 GHz Pencil Beam Microstrip Array Antenna
Antenna Lab
10
Path Finder Robot
Using AVR micro-controller, General Workshop
Software Skills
Programming Languages:
- Python(proficient)
- Matlab(proficient)
- C/C++(familiar)
- Java(familiar)
- Verilog(familiar)
Professional Software and Toolboxes:
- PyTorch
- TensorFlow v1.x v2.x
- JAX
- Numpy, Pandas, Matplotlib
- Android programming (Android Studio)
- Matlab GUIDE
- Latex
- FL Studio (music production)
