DEPARTMENT OF ELECTRICAL AND COMPUTER ENGINEERING · Electrical and Computer Engineering 704 Advanced Engineering Mathematics (MEng or MASc only)Dr. T. Field Email: [email protected]

DEPARTMENT OF ELECTRICALAND

COMPUTER ENGINEERING

McMASTER UNIVERSITY

GRADUATE COURSE DESCRIPTIONS!

i

I

2018/2019

:

;

Electrical and Computer Engineering 704

Advanced Engineering Mathematics(MEng or MASc only)

Dr. T. FieldEmail: [email protected]

Instructor:

This course is on the survey of a number of mathematical methods of importance inengineering modeling and analysis. The course covers basic set theory, relations andmapping, math logic, algebraic structures, linear mappings and matrices, metrics andtopological properties, Banach and Hilbert spaces, orthonormal bases and Fourier series.

Recommended textbooks:1. Modem Advanced Mathematics for Engineers, by V. V. Mitin, D. A. Romanov,

and M. P. Polis, Wiley, New York, 2001..

Grading based on the Project:

. Mid-term test: 20%

. Assignments: 20%

. Final: 60%

Term II

Electrical and Computer Engineering 710Engineering Optimization

Dr. Tim Davidson,ITB-A111, Ext. 27352.e-mail: davidson@,mcmaster.ca

Instructor:

http://www.ece.mcmaster.ca/~davidson/ECE710Course web page:

Boyd and Vandenberghe, Convex Optimization, Cambridge UniversityPress, Cambridge, 2004 (Book web page can be found at:http://www.stanford.edu/~bovd/cvxbook.html

Recommended Text:

Recommended Reading: Bertsekas, with Nedic and Ozdaglar, Convex Analysis andOptimization, Athena Scientific, Belmont, MA, 2003.Nocedal and Wright, Numerical Optimization, Springer, New York,1999.Bertsekas, Nonlinear Programming, 2nd edition, Athena Scientific,Belmont, MA, 1999.Gill, Murry and Wright, Practical Optimization, Academic Press,London, 1986.

A solid background in linear algebra. Exposure to numericalcomputing, programming, optimization and engineering design will behelpful, but is not required.

Prerequisite:

Principles of engineering optimization: modelling, formulation,solution and verificationA taxonomy of optimization problems and solution methodsConvex sets, convex functions and convex optimizationDualityUnconstrained optimizationConstrained optimization, including interior point methodsComputational complexity and NP-completenessApplications to engineering design

Course Outline:

There will be two lectures a week, each of about 90 minutes induration.

Lectures:

Midterm Test: 20%Final Exam: 35%Design Project: 45%

Assessment:

IITerm:


SILICON PHOTONICS - FUNDAMENTALS AND DEVICES

Dr. Jamal Deen, email: [email protected]:

M.J. Deen and P.K. Basu, "Silicon Photonics - Fundamentals andDevices", Wiley Series in Materials for Electronic &Optoelectronic Applications, ISBN-13: 978-0-470-51750-5 - JohnWiley & Sons, 2012.

Text:

The creation of affordable high speed optical communications usingstandard semiconductor manufacturing technology is a principal aimof silicon photonics research. This would involve replacing copperconnections with optical fibers or waveguides, and electrons withphotons. With applications such as telecommunications andinformation processing, light detection, spectroscopy, holography androbotics, silicon photonics has the potential to revolutionize electronic-only systems. This course will provide an overview of the physics,technology and device operation of exclusively silicon and relatedalloys.

Description:

Basic Properties of Silicon; Quantum Wells, Wires, Dots andSuperlattices; Absorption Processes in Semiconductors; LightEmitters in Silicon; Photodetectors , Photodiodes and Phototransistors;Raman Lasers including Raman Scattering; Guided Lightwaves;Planar Waveguide Devices and Fabrication Techniques and MaterialSystems.

Course Outline:

Project: The project can be a detailed review or investigation of any ofthe topics covered in the course. It should include a discussion of thekey papers in the topic as well as the most recent results. Students areexpected to demonstrate a mastery of their chosen topic in the projectreport and presentation.

Grading: Assignments - 50%Project - 30%Presentations - 20%

Term: I


Matrix Computations in Signal Processing

Dr. J.P. ReillyInstructor:

www.ece.mcmaster.ca/-reillvWeb page:

"Matrix Computations", 3rd edition, Golub and Van Loan, JohnsHopkins University Press

Text:

"Linear Algebra and Its Applications", 3rd edition, G. Strang"Applied Numerical Linear Algebra", James W. Demmel

References:

1. Review of fundamental concepts of linear algebra2. Covariance matrices and the Karhunen-Loeve expansion,

applications3. Singular value decomposition (svd), eigendecomposition (ed)4. Gaussian elimination, condition number, and error analysis5. Cholesky decomposition and applications6. Linear Least Squares Estimation: background, normal equations,

variance of solution, full-rank and rank-deficient solution using thesvd.

7. The QR decomposition: Householder, Givens, fast Givens, andmodified Gram-Schmit techniques, systolic arrays.

8. Solving least-squares using the QR decomposition: the full-rankand rank-deficient case.

9. Toeplitz systems

Course Outline:

Assignments - 2 @ 20% each = 40%Final Exam - 60%

Grading:

ITerm:

ECE 714

MIMO Communications

Instructor:Dr. Jian-Kang Zhang

ITB-A 217, ext. 27599

http://www.ece.mcmaster.ca/~ikzhang/Tentative Course Outline:Part I: Fundamentals of Digital Communications

1. Signal space2. Modulation and demodulation3. Information theoretic limits on communications4. Equalizations5. Channel coding

Part II: MIMO and Cooperative Communications1. MIMO channels and block Transmission2. MIMO channel capacity3. Transceiver design

4. Space-time block coding MIMO Rayleigh flat fading channels5. Distributed space-time block coding for cooperative relay networks6. Massive and large scale distributed MIMO channels

Recommended textbooks for Part I:Upamanyu Madhow, Fundamentals of Digital Communication,

Cambridge University Press, 2008Robert G. Gallager, Principles of Digital Communication

Cambridge University Press, 2008

Grading: Two Projects 40% (each 20%)Final Exam 60%

Term: II

ECE 715 Optimal Control of Dynamical Systems

Summary: This is an advanced graduate-level course intended for students with interest inmodern control theory and its applications. It will provide an overview of the fundamentalsof the theory of optimal control, focusing on its various formulations and solution strategiesusing variational approaches and dynamic programming. A wide range of topics will becovered. This includes optimal control of discrete-time systems, calculus of variations andoptimal control in the continuous-time domain, optimal control based on dynamicprogramming, classical linear quadratic regulators, and application of the Pontryagin’sminimum principal to optimal control of dynamical systems with input and state constraints.

LECTURES: 3hours/week

INSTRUCTOR: Shahin Sirouspour, ITBA319, Ext. 26238, [email protected]

RECOMMENDED TEXT- Instructor’s lecture notes.RECOMMENDED READING

- Lewis, Frank L., Draguna Vrabie, and Vassilis L. Syrmos. Optimal control. 3rd

edition, John Wiley & Sons, 2012.Kirk, Donald E. Optimal control theory: an introduction. Courier Dover Publications,2012.

PREREQUISITEAn undergraduate or graduate course in state-space control (e.g. ELEC ENG 4CL4);background in optimization would be helpful but not required. Please consult the courseinstructor for further information.

COURSE CONTENT• Introduction and preliminary materials on static optimization• Optimal control of discrete-time systems

o General problem formulation and necessary conditions for solutiono Discrete-time Linear Quadratic Regulators (LQR)o Steady-state suboptimal LQR

• Variational approach to continuous-time optimal controlo Fundamental concepts in calculus of variationso General problem formulation and necessary conditions for solutiono Continuous-time LQRo Steady-state suboptimal LQR

• Final free-time and constrained optimal controlo Final free-time problemso Pontryagin’s minimum principleo Optimal control with constraints on inputs and stateso Minimum time optimal control problems

• Dynamic programing and optimal controlo Bellman*s principle of optimalityo Discrete-time optimal controlo Continuous-time optimal controlo Relation to the minimum principle

• Output feedback and structured controlo LQR with output feedbacko Tracking problemo Model reference controlo Decentralized control

• Multi-variable controlo Robust output feedback controlo State observers and Kalman filterso Linear Quadratic Regulator (LQG)/ Loop Transfer Recovery (LTR)

Final Exam: 50%Assignments: 50%

EVALUATION:

Term II

1

ECE 717: Cloud Communications

Instructor: Professor Ted SzymanskiDepartment of Electrical and Computer Engineering

Office: Room ITB Building, RM A314Email: teds at mcmaster. ca

(In all emails to the prof, please start the subject with '731-',so that the email gets by the Spam filter.)

Course Web Site: On Avenue-to-Learnhttp://avenue.mcmaster.ca/

“Cloud Communications”: This course addresses the topics of communications, 'Quality ofService' (QoS), and efficiency in networks, including the Internet and wireless networks. Thephrase 'Cloud' has emerged as a metaphor for the ' Internet’, and most services are nowmigrating to the ‘Cloud’. The ‘Cloud’ can be viewed as the “central market-place” of the digitaleconomy. New Cloud-oriented services include Video-on-Demand services (i.e., Netflix), Video-rental and music purchase services (Apple’s iTunes), Music-streaming services (i.e., GooglePlay), Cloud-Computing services (i.e., Amazon or Google’s Cloud-Computing services),Distance-Education (i.e., Avenue-to-Learn), and many more. Our economies and way-of-livingare transitioning to use the ' Internet-of-Everything” (IoE), which will change how we live. TheIoE will interconnect ‘Smart-Things”, such as smart-appliances, smart-TVs, smart-homes, smart-buildings, smart-cities, smart-healthcare-systems, and smart-transportation-systems. In 2014,Cisco has estimated that the IoE represents a $14 Trillion dollar opportunity over the nextdecade, and its starting to happen right now. The next decade should witness a “sea of changes”.

The Internet network is enabling the transition to the new economy, and the need for betternetworks with improved energy-efficiency, resource-utilization and Quality of Service (QoS)guarantees is growing. The "Internet Engineering Task Force” (IETF) is the global entity thatmanages the technology of the Internet. It has defined concepts such as the “Differentiated-Services” (DiffServ), “Weighted-Fair-Queueing” (WFQ) and MPLS technologies. Equipmentmanufacturers such as Cisco, Juniper and Avaya often implement the IETFrecommendations/standards in their Internet routers and switches. The need for better networksspans all technologies, including fiber-optic networks, all-optical networks and wirelessnetworks.

The first part of the course will focus on an introduction to basic networks, basic switches androuters, network routing algorithms and scheduling algorithms. We will briefly look at basicCMOS technologies, all-optical technologies, control-planes and forwarding-planes. The secondpart of the course will focus on basic queueing systems and traffic models. We use these toestimate the network performance and queueing latencies. The third part of the course will lookat several classic network optimization problems, including 'Network Maximum Flow’ problem,

The primary source of material will be a collection of textbooks and recent IEEE journal andconference papers. Matlab programming will be used to illustrate key algorithms. A tentative listof topics includes;

Introduction

2

1. Introduction to 799 (1.5 hours)

Part 1: Basic Routing. Switching and Scheduling

1. Basic Networks and Network Routing (Open Shortest Path First (OSPF):la. Basic Network Terminology - The Internet network, the ‘Inter-Cloud” network, the “Internet-of-Everything”, Wireless Cell Networks, Data-Center Networks, Supercomputer Networks,Networics-on-a-Chip, the IETFlb. The Control-planes and Forwarding-planes, Software Defined Networking (SDN)lc. A Matlab demo of the basic OSPF routing algorithm used in the Internet.

2. Switching Architectures: Crossbar switches, complexity measures, 3 stage CLOS switches,Time and Space domain switches. Queueing in Crossbar switches: Input Queueing, OutputQueueing, Shared Queueing, Combined Input-Output Queueing. Internally buffered crossbars. (3hours). We’ll look briefly at CMOS technologies, All-Optical Technologies, CMOS/'VCSELtechnologies, Silicon-Photonics technologies.

3. Switch Scheduling: Some topics from Graph Theory, as applied to the Switch Schedulingproblem: Maximum Size Matchings (MSM), Maximum Weight Matchings (MWM), BipartiteGraph Matchings, Iterative matching algorithms: Parallel Iterative Matching (PIM), iSLIP,MUCS. Theoretical models: Stability, Oldest Cell First, Longest Queue First, Weighted FairQueueing (WFQ).(approx. 9 hours.)

4. Student Presentations : Topics selected by students, i.e., state-of-the-art routers or switches,all-optical switches or networks, silicon-photonics switches or networks, data-center networks,supercomputer networks, energy efficiency or resource utilization for Cloud services, routing forcloud services, scheduling for cloud services, opportunistic scheduling in wireless networks,routing or scheduling in optical packet networks, network coding, etc (approx. 6 hours,depending upon class size, and time-permitting).

Part 2: Queueing Systems and Traffic Models

1. Markov Chains: fonnulations, continuous time, discrete time, solution techniques. Stability,drift. Discrete event simulation techniques, confidence intervals, method of sub-runs. Presentedin the context of a classic problem - The Camber's Ruin problem. (A simple H.264 video trafficmarkov model-tentative, depending upon time) (approx. 3 hours)

2. Traffic Models: continuous & discrete models and processes:2a. Some real video traffic models: (i) H.264 “High Definition” (HD) Video-traffic (ie 1080p,with 1920x1080 pixels), (ii) H.265 “Ultra-High-Definition” (UHD) Video, 4K and 8K UIIDVideo traffic (ie 2160p, with 3840 x 2160 pixels)2b. Some Basic Mathematical models: Poisson, Bernoulli, Memoryless, Geometric, Uniform,Erlang, Hyperexponential, 2-state Bursty Markov-Modulated models, Self-Similar traffic.Statistics on traffic models.

3

2c. Traffic Modifiers/Shapers: Weighted Fair Queueing, Differentiated Services, Leaky-BucketTraffic Policers, Token-Bucket Traffic Regulators, (approx. 6 hours.)

3. Basic Queueing Systems and analysis techniques: M/M/1 and M/M/N queues, Burke’slaw, Little’s law, reversibility, Kleinrock’s independence model, Jackson’s Theorem. Thismaterial will be presented in the framework of the ONA tool from Bell Labs, (approx. 6 hours.)

4. Intermediate Queueing: M/G/l, G/M/l, and G/G/l queues. Approximations for the waitingtime of the G/G/l queue. This material will be presented in the framework of the QNA tool fromBell Labs. (Optional - The 'Diffusion Approximation' analysis for the GG1 queue.)(3 hours)

Part 3. Optimization Problems (Time Permitting):

1 . Network Flow Optimization: Minimum-Cut Maximum-Flow algorithms, (approx. 3-6hours.)

Summary: The topic of Network QoS, resource-utilization and energy-efficiency for the ‘Cloud’is a rapidly evolving subject area. New results appear each year in IEEE conference and journalpapers. It is difficult for any one textbook to cover the rapidly evolving topics in the field.Students will therefore have to review IEEE conference and journal papers in reasonable depthas a primary source of material, and will be required to perform an in-depth project on a selectedtopic.

Tentative Marking Scheme

1 Matlab routing exercise & paper review 10% (approx.)1 Class Presentation and Paper ReviewFinal Test (end of term)Term Project & PresentationClass participation

15 % (approx.) (this depends upon class enrollment)25 % (approx.)50 % (approx.)up to 5 % (approximate, at instructors discretion)

Grades for each component may be reallocated + /- 10 % to reflect relative difficulties.

Presentations: The professor will present the lectures. Students will complete a classpresentation/paper review) on a selected topic (the length will depend upon the class size.) Eachpresentation / paper review will be graded as part of the class work. A presentation shouldinclude both qualitative and quantitative aspects: a qualitative summary of the problem, areview of prior work in the area, a technical presentation of a proof or algorithm or mathematicalmodel or analysis. The technical work may include the generation of Matlab code to illustrate analgorithm or analysis (or part thereof). Students will be able to consult with the professor whilethey prepare their presentations. In addition to their presentations, students will be graded onclass participation, i.e., contributions to make the course better.

4

Term Projects: The term project will be similar to the class presentations, but will be morethorough. The student will identify a topic for further study, perform a literature search andidentify one (or more) key references (typically IEEE journal papers), review the paper(s),undertake technical work (using some software tools) in the area, potentially regenerateresults/graphs in a published paper, potentially test a Matlab model under varying assumptionsnot considered in the original paper, potentially propose modifications of the algorithm oranalysis, and summarize the work. The term projects will be due approximately aroundDecember 5, 2012 (final date to be announced).

Hardware-Based Term Projects: I may purchase some NETFPGA development boards, whichallow for a simple router to be implemented in hardware. Students interested in working on theNetFPGA board should have skills in hardware design, FPGAs and VHDL or Verilog.

Software-based Term projects: With the permission of the professor, some term projects candevelop software. An example project might be the development of a Kazza-like Peer-to-Peersystem to exchange video-clips, images or files over the web, using the JAVA or C programminglanguages and the Berkeley socket layer software library. The technical work here may consist ofdeveloping algorithms for distributed peer-to-peer systems and demonstrating the algorithms inC, C# or Java. Another project may be to outline an 802.1In wireless network, find a Linuxdevice driver for a commercially available wireless card (or USB wireless device), and compilethe device deriver for different settings, attempting to affect the QoS. Another project could bethe development of ‘Apps’ for the iPAD or another tablet-computer.

Some Reference Texts:

[1] "Data Networks", 2nd Edition, D. Bertsekas and R. Gallager, Prentice-Hall, 1992[2] "Queueing Systems Vol. I and IF, Leonard Kleinrock, Wiley, 1978[3] "Queueing Networks and Markov Chains" , G. Bolch, S. Greiner, H. de Meer, K.S. Trivedi,Wiley, 2006[4] "Teletraffic Engineering", International Telephony Union-(ITU), Geneva, January 2005. 336pages, Available free on the web: http://www.tele.dtu.dlc/teletraffic/[5] " Modelling Foundations - Basic Mathematics" , International Telephony Union-(ITU),Geneva, January 2005. 53 pages, Available free on the web: http://www.tele.dtu.dk/teletraffic/

A good textbook which reviews the basics of networking is used in our 4th year networkingcourses:

[6] Communication Networks: Fundamental Concepts and Key Architectures, A. Leon-Garcia &I. Widjaja, 2nd Ed., ISBN 0-07-246-352-X, McGraw Hill, 2004

TERM IU (May - August 2018)

ELECTRICAL AND COMPUTER ENGINEERING

ECE 720 Power Converter Systems

Dr. Mehdi NarimaniITB-A320Email: [email protected]

Instructor:

Bin Wu and Mehdi Narimani “High-Power Converters and AC Drives,”Wiley - IEEE Press, 2017http://ca.wilev.com/WilevCDA/WilevTitle/productCd-l119156033.html

Recommended Textbook:

A course on the analysis, simulation and design of power convertersystems. Main topics include: high-power multi-pulse rectifiers, multilevelvoltage and current source converters, pulse width modulation, harmonicreduction techniques, modeling and simulation techniques, and industrialapplications. Important concepts are illustrated with design projects usingMatlab/Simulink.

Course Description:

Course Outline: IntroductionHigh-Power Semiconductor DevicesMultipulse Diode RectifiersMultipulse SCR RectifiersTwo-level Voltage Source InverterMultilevel Cascaded H-Bridge ConvertersMultilevel Diode-Clamped InverterOther Multilevel Voltage Source ConvertersCurrent Source Inverters

Assignments (5Assignments)Final ProjectTotal

Course Evaluation 75%25%100%

Term I


Information Theory and Coding

Dr. Steve Hranilovice-mail: [email protected]: http://www.ece.mcmaster.ca/~hranilovic

Instructor:

This is an introductory course in information theory and coding theory.As stated in the course text:

Information theory answers two fundamental questions incommunication theory: what is the ultimate data compression?(Answer: the entropy H) And what is the ultimate transmission rateof communication? (Answer: the channel capacity C).

In later stages of the course, coding techniques will be discussed whichapproach these ultimate limits.

Objectives:

Thomas M. Cover and Joy A. Thomas, Elements of Information Theory,John Wiley & Sons, 1991, ISBN 0-471-06259-6.

Required Text:

Stephen B. Wicker, Error Control Systems for Digital Communicationand Storage, Prentice-Hall, 1995, ISBN 0-13-200809-2

Reference Text:

1. Entropy, relative entropy, mutual information, chain rules, dataprocessing inequality, the asymptotic equipartition.

2. The Kraft inequality, Shannon-Fano codes, Huffman codes,arithmetic coding.

3. Discrete channels, random coding bound and converse, Gaussianchannels, coloured Gaussian noise and optimal “water-pouring”power allocation.

4. Linear block codes and their properties, hard-decision coding, cycliccodes, convolutional codes, soft-decision decoding, Viterbi decodingalgorithm.

5. Lattice codes, trellis coded modulation, coset-codes, multi-levelcodes/multi-stage decoding, iterative decoding.

Tentative Outline:

Mini-Project I - 15%, Mid-term - 30%Mini-Project II - 15%, Final Exam - 40%

Grading:

Term: I

ECE 724 MODELING, CONTROL AND DESIGN OF ELECTRIFIED VEHICLES

Instructor: Dr. Jennifer Bauman

Contact: Jennifer.bauman(6)mcmaster.caITB A217, 905-525-9140 X27784

Course Description:

This course covers the modeling, control,and design of electrified vehicles, including hybrid, plug-inhybrid, and pure electric vehicles. The course will use the textbook "Hybrid Electric Vehicle SystemModeling and Control, 2nd Edition" by Wei Liu.Course content will be supplemented with academicpapers and numerous real-world examples, including the analysis of real hybrid and electric vehicles.The high-level goal of this course is to understand the vehicle model as a testbed for evaluating futuredesign and control ideas. By the end of the course, students will be able to create accurate vehiclemodels validated to real-world data,and use these models to evaluate new ideas. The course content isapproximately divided as follows:

Textbook Chapter(if applicable)

Week Topic

Introduction,motivation, powertrain architectures, model types1 1Fuel consumption measurement and standards2 10Modeling of vehicle body, final drive, wheel,driver, regenerative braking3 3Batteries and battery modeling4 3, 5Engine, fuel cell,and motor modeling5 3Transmission modeling and control6 3Power electronics modeling and control7 4Ultracapacitors, hybrid energy storage systems, control8

Hybrid Control Strategies- Rule-based, fuzzy-logic control,costfunction, predictive, PHEV control

9 6,7.7 - 7.9

Electrified vehicle design10 10EV Charging-Characteristics and impact on power distribution system,Level1, 2, and 3 charging, smart-charging

11 8

Student project presentations12

Course Delivery and Assessment:

The course will have one 3-hour lecture per week. The course will be heavily project-based, with thegoal of the main project being to perform a small research study in the electrified vehicle space thatutilizes vehicle modeling to investigate design and/or control ideas. Note that there will be noextensions on any deadlines. The course assessment is as follows:

% ofDeliverable Description Total

MarkDue Date

Perforin a short literature review and choose a journalpaper to study. Write a 4-6 page (double-spaced)critique of the paper. Discuss strong/well-supportedareas of the paper and areas that could be improved(experimental approach,analysis, etc.). Is thecontribution novel? Is it useful? What are ideas forfuture directions of this research?

Jan. 19, 2018(5pm)

(submit thejournal paper

and yourcritique)

Assignment #1-Article Critique 10%

Use Ford Focus EV data to create a vehicle model inMATLAB/Simulink. Compare the simulated batterystate-of-charge to the real logged vehicle data for atleast 10 drive cycles, Write a 2-3 page (double-spaced)technical report to describe your results, andcomment on any discrepancies in your results.

Feb. 16, 2018(5pm)

(submitMATLAB/

Simulink files andtechnical report)

Assignment #2-EV Model 25%

See Project Breakdown belowProject 65% See below

Project Breakdown:

% of ProjectMark

Project Component Due Date

Literature review and research project proposal (3-4 pages double-spaced). This should include a discussion of data required. Relevantparts of this literature review can be used in your final paper.

Jan. 29, 2018(5pm)

15%

Vehicle model. This will be assessed during a 20-minute one-on-onemeeting with Dr. Bauman where the student will present the model,run the model,and show simulation results as needed. The studentshould be prepared to describe how the model was validated andhow data was used to perform this validation, By this meeting, themodel should be 95% complete,but can still be fine-tunedafterwards. Dr. Bauman will provide feedback as needed that can beincorporated into the final paper.

On or afterMarch 15,

201825%

In-class presentation. This will be a conference-style presentation ofapproximately 10-15 minutes.

April 4, 2018(last day of

class)15%

IEEE Conference-style final paper April 30, 2018(5pm)

45%

Term tl

Electrical and Computer Engineering 727Wireless Communication Networks

Dr. T. ToddOffice: ITB/A324, 529-7070 ext. 24343Email: [email protected] website: http://owl.mcmaster.ca/~todd/

Instructor:

Prerequisites: Comp Eng 4DK4 (Computer Communication Networks) or equivalent,ECE 739 (Resource Management and Performance Analysis In WirelessCommunication Networks) or equivalent, CAS 736 (Analysis OfStochastic Networks) or equivalent, or permission of the instructor. Cprogramming experience required. Access to a Unix/Linux/Windowsworkstation and a C compiler.

This is an advanced course on wireless networking which focuses onvarious topics relating to cellular and wireless mesh networks. Much ofthe course will be run using student presentations and discussion, andeach student will do a project containing a significant researchcomponent.

Course Objective:

Textbook/Reading: Various books, papers, articles and lecture notes.

Lab/AssignmentsProjectClass Presentations and/or Final Exam

Grading: 20%60%20%

IITerm:

ECE729 Resource Management in Mobile Computation Offloading

Dr. Dongmei ZhaoDepartment of Electrical and Computer Engineering

McMaster UniversityOffice: ITB-A323, ext. 26127Email: [email protected]

Course description:

This course will start with an introduction to mobile cloud computing (MCC), relatedbackground and main issues as the first part; the second part will focus on current researchprogress on binary-decision mobile computation offloading (MCO); and the third part willdiscuss recently proposed methods on offloading decisions based on fine-grain job partitioning.The course will end with discussions on some open issues and research topics that researchersin wireless networking on most likely interested in.Part of the class time will be devoted to student presentations. The focus for thesepresentations will be on MCC for vehicular networks. Students will search the literature,present recent proposals, designs, and developments on MCO for vehicular networks, andcomplete a research project at the end of the semester.

The main purpose of the course is to provide students with an overview of the state-of-the-artresearch on mobile computation offloading, and help them formulate and solve specificresearch problems,Active discussions will be important to achieve this objective.Outline

Introduction:mobile cloud computingmobile edge computing

MCO based on full migration:-Framework-formulationsRecent research work

Partitioning-based MCOOffloading for a single userSimultaneous offloading from multiple users

Open issues and research topics

References:

Course notes and a collection of journal papers from the recent literature.Prerequisites:

Computer Communications Networks,and Wireless Communication Networks,Probability and Random Processes,and Elementary Queueing Theory.

Grading:

Assignments: 20%;class presentations: 20%;projects: 60%.

Term I


NETWORK INFORMATION THEORY

Prof. Jim ChenITB-A221, X20163, [email protected]

Instructor:

Prerequisites: Undergraduate courses in linear algebra, signals and systems,probability and digital communication. Prior Exposure to informationtheory is preferred, but not required.

General Description: Network information theory deals with the fundamental limits oninformation flow in networks and optimal coding techniques andprotocols that achieve these limits. It extends Shannon’s point-to-point information theory to networks with multiple sources anddestinations. Although a complete theory is yet to be developed,several beautiful results and techniques have been developed over thepast forty years with applications in wireless communication, theinternet, and other networked systems. This course aims to provide abroad coverage of key results, techniques, and open problems innetwork information theory.

Abbas El Gamal and Young-Han Kim, Lecture Notes on NetworkInformation Theory, {online} http://circuit.ucsd.edu/ vhk/Init.html.

Textbook:

Tentative Outline(Time Permitting: 1. Entropy, Mutual Information, and Typicality

2. Point-to-Point Communication3. Multiple Access Channels4. Degraded Broadcast Channels5. Interference Channels6. Channels with State7. General Broadcast Channels8. Distributed Lossless Source Coding9. Source Coding with Side Information10. Distributed Lossy Source Coding11. Multiple Descriptions12. Joint Source-Channel Coding

Grading: Lecture Report 50%, Presentation 50%, Project (optional) 20%

Term: I

ECE 7363D Image Processing and Computer Vision

Instructor: Shahram Shirani, ITB-A320Email: [email protected]

Description:

.Central to computer vision are the mathematical models governing image formation and methodsfor processing and recovering information based on the model and the image data. In fact, thegoal of computer vision is to use observed image data to infer something about the world. In thiscourse we concentrate on statistical and geometrical models of visual data. Assuming a statisticalmodel for the visual data, we talk about learning and inference. We cover modeling of the datadensities, regression and classification methods and how we can use graphical models (e.g.,Vitterbi, belief propagation) to solve learning and inference problems.In the other half of this course we take a geometrical approach to image formation and look atproblems such as image blending and stitching and 3D reconstruction. In our discussion of 3Dcomputer vision, we focus on how to make use of the spatial and temporal coherence imposed bycamera geometry to reconstruct a 3D imagefrom a moving video camera, stereo camera rig ormultiple views from a still camera.

Learning Objectives:• Model image formation in single camera and multi-camera setups• Mathematically understand the relation between the 3D world and it's projection in 2D

images and learn how to use these to reconstruct a 3D scene model from several 2Dimages

• Find appropriate models for complex data densities• Choose the right regression model for a vision problem• Be able to use graphical models to simplify complex data models• Be able to apply computational photography techniques in order to solve image

processing and computer vision problems

Content:• Image formation

o Geometric primitives and transformationso Photometric image formationo Digital camera pipelineo Pinhole camera modelo Multiple cameras model

• Dense motion estimation• Computational photography

o Image blending and compositingo Image retargetingo Texture synthesis and transfer

o Image completion / inpaintingo Super-resolution, deblurring, and denoisingo High dynamic range imagingo Depth and defocuso Coded aperture photographyo Image stitching

• 3D reconstructiono Stereo correspondenceo Image-based renderingo 3D-TV signal formation and compression

• Learning and inference in visiono Modeling complex data densitieso Regression modelso Classification modelso Application of graphical models for learning and inference in vision

Textbooks:• Richard Szeliski, “Computer Vision: Algorithms and Applications”, Springer; 2011.

Chapters 6-12 and 14-16 will be covered.

• Simon J.D. Prince, “ Computer vision: models, learning and inference”, CambridgeUniversity Press, 2012.Chapters 8-13 will be covered.

Reference Books:• Richard Hartley, Andrew Zisserman, “Multiple View Geometry in Computer Vision”

Second Edition, Cambridge University Press, 2004.• BoguslawCyganelc, J. Paul Siebert, “An Introduction to 3D Computer Vision Techniques”

Wiley, 2009

Students will attempt a number of small implementation projects, which often build on oneanother, in order to get used to working with real-world images and the challenges that thesepresent. The students are then asked to choose an individual topic for each of their final projects.

Prerequisite: undergraduate level DSP, undergraduate level probability, undergraduate levelimage processing

Assessment:• Homework:• Four assignments having equal weight• Project:

45%

55%Term1

Electrical and Computer Engineering

737 Radar Systems

To provide a solid conceptual and quantitative background in themodelling of biological neurons and how they function as computationaldevices. Practical experience will be gained in modelling neurons from anumber of perspectives, including equivalent electrical circuits, nonlineardynamical systems, and random point-processes, and an introduction tothe mathematics required to understand and implement these differentengineering methodologies will be given.

Objective:

Instructors: Dr. T. Kirubarajan

Summary:

This course will provide an in-depth coverage of modem radar systems with application tosurveillance, automotive and biomedical systems. Starting from the fundamental radar equations,the course will address real-world issues like clutter modeling, propagation effects, objecttracking and countermeasures. The course will cover short and long-range radars, phased arrayradars, over-the-horizon radars and software-controlled radars. In addition to learning theory, thestudents will also get hands-on experience through multiple projects on processing real and/orrealistically simulated radar data.

Course Outline:

1.Radar Equations2.Radar Transmitters and Receivers3.Radar Clutter4.Detection in Noise5.Propagation Effects6.Doppler Processing7.Tracking Radar8. Phased Array Radar9. Over-the-horizon Radar10. Electronic Countermeasures11. Electronic Counter-Countermeasures12. Processing real and/or realistically simulated radar data

Suggested Reading:

1.Mark Richards, James Scheer and William Holm, “Principles of Modem Radar: BasicPrinciples”, SciTech Publishing, 2010.

2.BassemMahafza, “Radar Systems Analysis and Design Using MATLAB”, Third Edition,Chapman and Hall/CRC, 2013.

Grading:

Assignments: 20%Midterm project: 30%Final project: 50%

Term I

Instructor:Instructor:Lectures:Office Hours:

ECE 740 - Semiconductor Device Theory and ModelingProf. Y, Haddara - E-mail: [email protected]

Three hours per week

One hour immediately after each class or by appointment.

Text:Course Description:

D.A. Neaman - Semiconductor Physics and Devices, 3rd Ed., McGraw Hill (2002).This course provides a fundamental in-depth knowledge of the theory of operation,modeling, parameter extraction, scaling issues, and higher order effects of active andpassive semiconductor devices that are used in mainstream semiconductor technology andemerging devices of practical interest. There will be a comprehensive review of the theoriesand latest models for the devices that are valid out to very high frequencies and the use ofphysical device modeling. A review of the latest device technologies and architectures willbe presented. The course will be a prerequisite to the other applied courses inmicroelectronics and photonics.

Course Outline1. Review of semiconductor fundamentals.2. Homo- and hetero-junction devices - theory; modeling; parameter extraction.3. MOS capacitors and transistors - theory; modelling; parameter extraction; scaling issues; reliability.4. Bipolar transistors - theory; modeling; parameter extraction; scaling issues; reliability.5. Photodetectors-theory; modelling; parameter extraction; and scaling issues.6. Transport and modeling of disordered semiconductors (organic and polymeric) devices.Project Description: The project can be a detailed review or investigation of a specific part of the course. Examples

are Nano-scale MOS architectures and performance; Advanced silicon-based photodetectors;SiGe HBTs or Nanowire silicon-based transistors; Transistor design and performance forspecific (e.g low-noise) applications; Device (MOS, BJT or HBT) parameter extractiontechniques; Modeling issues of silicon diodes at high frequencies; Carrier transport in nano-scale MOS transistors; Conductivity of organic devices; Carrier scattering in nano-MOStransistors; Modeling issues of passive components in silicon technology at microwavefrequencies; etc.

Grading: Assignments - 35% Project - 35% Final Exam - 30%

Selected ReferencesIEEE Transactions on Electron Devices, Solid-State Electronics, Journal of Applied Physics etc.ECS, ICMTS, IEDM, ESSDERC, DRC Proceedings.Device simulators and manuals-Synopsis, Silvaco, TMA etc..Y.P.Tsividis - Operation and Modelling of the MOS Transistor, 2nd Ed., McGraw Hill (1999), (TK 7871.99.M44.T77)D.J. Roulston - Bipolar Semiconductor Devices, McGraw Hill (1990), TK 7871.86.R68.D. Ferry, L. Akers and E. Greeneich - Ultra Large Scale Integrated Microelectronics, Prentice Hall (1988).C.T. Sah - Fundamentals of Solid-State Electronics, World Scientific, Singapore (1991), TK 7871.85.S23M. Shur, Physics of Semiconductor Devices, Prentice Hall (1990), QC 611.S563.S.M. Sze - Physics of Semiconductor Devices, John Wiley & Sons (1981), TK 7871.85.S988.S.M. Sze (Ed.)- Modem Semiconductor Device Physics, John Wiley & Sons (1998), QC 611.M674.M.S. Tyagi - Introduction to Semiconductor Materials and Devices, John Wiley (1991), TK7871.85.T93.S. Wang- Fundamentals of Semiconductor Theory and Device Physics, Prentice Hall (1989), QC 611.W32.R. Warner & B. Grung - Semiconductor Device Electronics, Holt Rinehart & Winston (1991), ISBN 0-03-009559-X.

Term I


Analysis and Design of RF ICs for Communications

Dr. C.H. (James) ChenInstructor:

Bosco Leung, VLSI for Wireless Communications, Prentice-Hall,TK7874.75.L48, 2002

Texts:

1. B. Razavi, RF Microelectronics, Prentice-Hall Inc., 1998.2. T.H. Lee, The Design of CMOS Radio-Frequency Integrated Circuits,

Cambridge University Press, 1998.3. G. Gonzalez, Microwave Transistor Amplifiers: Analysis and Design,

2nd ed., Prentice-Hall Inc., 1997.4. Lawrence P. Huelsman, Active and Passive Analog Filter Design: An

Introduction, McGraw-Hill, 1993.5. D.A. Johns and K. Martin, Analog Integrated Circuit Design, John

Wiley & Sons, Inc., New York, 1997.6. P.E. Allan and D.R. Holberg, CMOS Analog Circuit Design,2nd ed.,

Oxford Press, 2002.7. B. Razavi, Design of Analog CMOS Integrated Circuits, McGraw-

Hill, 2001.8. Clarke and Hess, Communication Circuits: Analysis and Design,

Krieger, Reprint, 1994.9. H.L. Krauss, C.W. Bostian, F.H. Raab, Solid State Radio

Engineering, Wiley, 1980.

Reference Texts:

This course provides a fundamental and in-depth knowledge of theanalysis and design of radio-frequency (RF) integrated circuits (IC) in theCMOS technology for wireless communications. The topics include themodelling of active and passive components for AC and noise analysis,design examples of amplifiers, filters, oscillators, PLL and frequencysynthesizers. Circuit performance will be evaluated by both handcalculations and computer simulations. A good understanding of circuitanalysis and CAD tools (e.g., HSPICE or SpectreRF) is required.

Course Description:

1. Passive and active components at RF2. Design of low-noise amplifiers3. Active and passive filters4. Operation and design of mixers5. Oscillators6. Phase locked loop7. Frequency synthesizers

Course Outline:

Project will be the design and detailed analysis, including both handcalculations and computer simulation of an RF integrated circuit for aspecific purpose/application. Possible projects will be discussed at thebeginning of class.

Project:

Grading: Assignments - 50%Term Project - 50%

Term: I


Advanced Topics in Multimedia Coding and CommunicationsInstructor: Sorina Dumitrescu, ITB/A222, Ext: 26486, [email protected]

The aim of this course is to familiarize the students with recent results in several modem research topicsin multimedia coding and communications. The four main topics (joint source-channel coding/decoding,multiple description coding, distributed source coding, network coding) are primarily motivated by newchallenges in multimedia transmission over modern communications networks. The presentation willinclude the theoretical foundations (asymptotic rate-distortion results) as well as practical aspects(practical code design, decoding strategies), applications, and open problems.Research in joint source-channel coding/decoding for point-to-point communications is motivated by theidea that practical constraints in communications systems (eg. constraints in delay and/or complexity)may rend suboptimal the separate design of compression algorithm (source code) and error protectionscheme (channel code), thus leading to severe degradation in performance. Therefore, joint approaches inthe design of the two components have the potential of improving the system's performance. Multipledescription coding (MDC) refers to generating several separate descriptions of a common source, suchthat it can be reconstructed to some fidelity from any subset of received descriptions. By allowing agraceful degradation of the signal fidelity in adverse channel conditions, MDC becomes an importantparadigm for reliable multimedia transmission over packet lossy networks. Distributed source codingrefers to separate compression of multiple correlated sources, It finds applications in sensor networks, butalso in other scenarios like layered coding for video streaming. Network coding has at its core theprinciple of allowing each intermediate network node to encode the data it receives. This leads to anincrease of the information flow through the network versus the case when nodes perform only routing,

Intended Outline

1) Brief review of relevant elements of information theory and of some results for point-to-pointcommunication.

2) Joint source-channel coding/decoding - a) Optimal index assignment (IA) problem inquantization. Channel optimized quantizer design. Generalized Lloyd algorithm, b) Decodingstrategies which exploit the redundancy left in the source code. MAP sequence and/or MAPsymbol estimation algorithms for joint source-channel decoding (JSCD) of fixed/variable lengthcodes.

3) Multiple description coding (MDC)- Rate-distortion bounds on MDC performance. GeneralizedLloyd algorithm for optimal MD scalar quantization; the index assignment problem. Optimalindex assignment for MD lattice vector quantization. MDC via unequal erasure protection andsuccessively refinable codes. Application of MDC in multimedia streaming.

4) Distributed source coding (DSC) - Slepian-Wolf and Wyner-Ziv theorems. Practical design ofasymmetric and symmetric DSC. Applications in sensor networks. Wyner-Ziv video coding.Robust distributed source coding.

5) Network coding: The concept of network information flow. Max-flow Min-cut Theorem and theasymptotic optimality of linear network codes for single information source. Algorithm for linearnetwork code construction, Layered multicast with inter-layer network coding for multimediastreaming.

Assessment: assignments: 70%; final exam 30%.

References: The following bibliographical list will be complemented with other recent publications inIEEE Transactions on Information Theory, Communications, Image Processing, Multimedia and others.[1] K. Zeger, A. Gersho, "Pseudo-Gray coding", IEEE Transactions on Communications, vol. 38, no. 12,pp. 2147 - 2158, Dec. 1990.[2] N. Farvardin and V. Vaishampayan, "On the Performance and Complexity of Channel OptimizedVector Quantizers," IEEE Transactions on Information Theory, vol. 37, pp. 155-160, Jan. 1991.[3] K. Sayood and J. C. Borlcenhagen, "Use of Residual Redundancy in the Design of JointSource/Channel Coders", IEEE Transactions on Communications, vol. 39, no, 6, pp. 835-846, June 1991.[4] N. Phamdo and N. Farvardin, "Optimal Detection of Discrete Markov Sources over DiscreteMemoryless Channels - Applications to Combined Source-Channel Coding", IEEE Transactions onInformation Theory, vol. 40, no. l , pp. 186-193, Jan. 1994.[5] A. A. El Gamal, and T. M. Cover, "Achievable Rates for Multiple Descriptions", IEEE Transactionson Information Theory, vol. 28, no. 6, pp. 851-857, Nov. 1982.[6] W.H.R., Equitz, T.M., Cover, "Successive refinement of information", IEEE Transactions onInformation Theory, vol. 37, no. 2, pp. 269 - 275, March 1991.[7] V. A. Vaishampayan, "Design of Multiple Description Scalar Quantizers", IEEE Transactions onInformation Theory, vol. 39, no. 3, pp, 821-834, May 1993.[8] V. K. Goyal, "Multiple description coding: compression meets the network", IEEE Signal ProcessingMagazine, vol. 18, pp. 74-93, Sept. 2001.[9] V. A. Vaishampayan, N. J. A. Sloane, and S. Servetto, "Multiple-Description vector quantization withlattice codebooks: design and analysis", IEEE Transactions on Information Theory, vol. 47, no. 5, pp.1718-1734, July 2001.[10] D. Slepian and J.K. Wolf, Noiseless coding of correlated information sources, IEEE Transactionson Information Theory, vol. IT-19, pp. 471-480, July 1973.[11] A.D. Wyner and J. Ziv, "The rate distortion function for source coding with side information at thedecoder," IEEE Transactions on Information Theory, vol. IT-22, pp.1-10, January 1976.[12] Z. Xiong, A. Liveris, and S. Cheng, "Distributed source coding for sensor networks", IEEE SignalProcessing Magazine, vol, 21, pp. 80-94, Sept. 2004.[13] V. Stankovic, A. D. Liveris, Z. Xiong, C. N. Georghiades, "On code design for the Slepian-WolfProblem and Lossless Multiterminal Networks", IEEE Transactions on Information Theory, vol. 52, no. 4,pp. 821-834, April 2006.[14] P. A. Chou and Y. Wu, "Network coding for the Internet and Wireless Networks", IEEE SignalProcessing magazine, pp. 77-85, Sept. 2007,

[15] P. Sanders, S. Egner, P. Chou, M. Effros, S, Egner, K. Jain and L. Tolhuizen, "Polynomial timealgorithms for multicast network code construction," IEEE Transactions on Information Theory, vol. 51,no. 6, pp. 1973-1982, June 2005.[16] R. W. Yeung, S. -Y. R. Li, N. Cai, and Z. Zhang, Network Coding Theory, Foundation and Trendsin Communications and Information Theoiy, vol. 2, nos. 4 and 5, pp. 241-381, 2005.Term II


Modeling and Simulation of Photonic Devices and Circuits II(Active and Functional Devices)

Dr. Xun LiE-mail: [email protected]

Instructor:

Electromagnetic Fields (3FI4), Solid State Devices I (4E03),Solid State Devices II (4F03), Modeling and Simulation ofPhotonic Devices and Circuits I (ECE-754)

Prerequisite:

Recommended Texts: Course notes

Course Description: Photonic devices and circuits are key components used forlightwave generation, amplification, transmission and detectionin communication systems and networks. Photonic devices andcircuits that utilize primarily photons, in conjunction withelectrons can offer the tremendous bandwidth which is the key toa variety of applications, especially broadband communicationsystems and networks. This course will focus on the modeling ofactive and functional device physics through numericalapproaches, the simulation of device terminal performancesthrough mixed analytical and numerical methods and theextraction of device behavior models.

1. Introduction to optoelectronic device modeling2. Optical wave propagation3. Optical property of semiconductors4. Carrier transport in semiconductors5. Thermal property of semiconductors6. Numerical solution techniques7. Selected device modeling and simulation examples

Course Outline:

Midterm minor project 35%, Final major project 65%Grading:

Term: II


Design of Lightwave Communication Systems and Networks

Instructor: Dr. S. KumarCRL-204, ext: 26008Email: [email protected]

Communication Systems (3TI4), Discrete Time Systems and RandomProcesses (3TJ4), Computer Communication Networks (4DK4)

Prerequisite:

"Optical Fiber Telecommunications IIIA and IIIB", edited by I.P.Kaminow and T.L. Koch, Academic Press, ISBN 0123951704 (IHA) andISBN 0123951712 (DIB)

"Fiber-Optic Communication Systems", Govind P. Agrawal, John Wileyand Sons, Inc., 1997, ISBN 0-471-17540-4

Recommended Texts:

Course Description: Lightwave communication has emerged as the undisputed transmissionmethod of choice in almost all areas of telecommunication, mainlybecause it offers unrivaled transmission capacity at low cost. Startingwith the design of photonic devices for lightwave generation, modulation,amplification and detection and optical fibers for lightwave transmission,this course will mainly focus on the design of lightwave communicationsystems and networks based on these photonic devices and optical fibers.1. Lightwave generation and modulation2. Fiber Transmission3. Lightwave amplification4. Lightwave detection5. Advanced components for multiplexing and networking6. Transmitter design7. Amplifier design8. Receiver design9. Transmission protocols and line coding10. Design of point to point WDM system11. Transport networks, access networks and packet switched networks.

Course Outline:

Grading: Project - 50%Final Exam - 50%

IITerm:


Numerical Techniques in Electromagnetics

Instructor: Prof. Mohamed BakrITB-A219, ext. 24079, e-mail: mbakr@,mail,ece.mcmaster.ca

Prof. Natalia NikolovaITB-A220, ext. 27141

Support Lecturer:

EE3FI4 Theory and Applications in Electromagnetics or equivalentPrerequisite:

Course Description: This course provides a solid understanding of the computationalelectromagnetic techniques used to model electromagnetic phenomenarelated to microwave and millimetre-wave engineering, antennaengineering and wireless technology. We adopt a systematic approachin which the complexity and dimension of the explained techniques areincreased starting with simple ID problems. Lectures will cover thefollowing topics:1. Fundamentals of electromagnetic theory-revision.2. Green's functions in electromagnetic equations.3. Method of Moments (MoM) and applications.4. Finite Difference techniques and the Finite Difference Time

Domain (FDTD) method.5. Huygen's principle and the time domain Transmission Line

Modeling (TLM) method.6. Variational approaches in electromagnetics and the Finite Element

Method (FEM).7. The Mode Matching Method.8. Recent advances in numerical electrodynamics-open discussion.

Recommended Texts: 1. C.A. Balanis, Advanced Engineering Electromagnetics, JohnWiley and Sons, 1989

2. R.F. Harrington, Time-Harmonic Electromagnetic Fields,McGraw-Hill, Inc., 1961.

3. M.N.O. Sadiku, Numerical Techniques in Electromagnetics, CRCPress, 1992.

4. R.C. Booton, Jr., Computational Methods for Electromagneticsand Microwaves, John Wiley & Sons, 1992

Additional Resources: A selection of literature papers.

Four projects for 25% eachStudents are expected to give presentations explaining theirapproaches. Part of the project grade is assigned to the presentation.All implementations are expected in Matlab.

Evaluation:

Term: I


Algorithms for Parameter and State Estimation

Dr. T. KirubarajanOffice: ITB-A313, ext. 24819e-mail: kiruba@,mcmaster.ca

Instructor:

This course presents parameter and state estimation algorithms fornoisy dynamic systems. The objective is to present a comprehensivecoverage of advanced estimation techniques with applications tocommunications, signal processing and control. In addition to theory,the course also covers practical issues like filter initialization, softwareimplementation, and filter model mismatch. Advanced topics onnonlinear estimation and adaptive estimation will be discussed as well.The concepts will be put into practice by the students on realisticestimation projects.

Outline:

Engineering mathematics, linear systems, probability and stochasticprocesses

Prerequisites:

References: 1. Y. Bar-Shalom, X. Rong Li and T. Kirubarajan, Estimation withApplications to Tracking and Navigation, John Wiley & Sons,2001 .

2. R. G. Brown and P. Y. C. Hwang, Introduction to Random Signalsand Applied Kalman Filtering, John Wiley & Sons, 1992.

3. F. L. Lewis, Optimal Estimation, John Wiley & Sons, 1986.4. D. Manolakis, Statistical and Adaptive Signal Processing: Spectral

Estimation, Signal Modeling, Adaptive Filtering and ArrayProcessing, McGraw-Hill, 2000.

Course Outline: 1. Basic concepts:a. Maximum likelihood (ML) estimationb. Maximum a posteriori (MAP) estimationc. Least squares (LS) estimationd. Minimum mean square error (MMSE) estimatione. Linear MMSE (LMMSE) estimation

2. LS estimation for linear and nonlinear systems3. Modeling stochastic dynamic systems4. The Kalman filter for discrete time linear dynamic systems with

Gaussian noise

5. Steady static filters for noisy dynamic systems6. Adaptive multiple model estimation techniques7. Nonlinear estimation techniques8. Computational aspects of discrete time estimation9. Extensions to autocorrelated noise and smoothing10. Continuous time state estimation

Grading: Exams 50%; Projects 40%; Homework assignments 10%

Term: I

Electrical and Computer Engineering 111NEURAL NETWORKS AND LEARNING MACHINES

Dr, S. HaykinEmail: [email protected]

Instructor:

1. Introduction: Models of Neurons:. Related items

2. Support-vector Machines:. Classification of input data. Example 1

3. Multilayer Perceptrons:. Back-propogation algorithm. Advantages and limitations. Example 2

4. Deep-learning Algorithms:. Long short-term memory (LSTM). Back-prpogation algorithm revisited. Example 3. Adaptive critic

5. References6. Projects:

. Mid-term exam 25%

. Final exam 75%

Term 1

Electrical and Computer Engineering 775COGNITIVE DYNAMIC SYSTEM THEORY

Dr. S. HaykinEmail: [email protected]

Instructor;

Abstract of the Course:

1. Introduction

2. Cognitive Dynamic System Simulating Certain Features of the Brain3. Principles of Cognition4. Perception-action Cycle5. Perceptor: Bayesian Dynamics

. Bayesian generative modeling

. Bayesian filtering reciprocally couple to the generative model

. Entropic state of the preceptor

6. Feedback Channel Linking the Perceptor and the Executive. Internal rewards

7. Executive: Cognitive Control. Reinforcement learning. Shunt cycle. Action space. Planner for set of prospective actions. Policy for decision-making. Cognitive Action on The Environment

8. Conclusions9. References for the Course10. Course Project

. Mid-term project: 25%

. Final project: 75%

Term 1


Sensory and Neuromuscular Engineering

Dr. Hubert de BruinITB-A211, X24171. [email protected]

Instructor:

To give the student a more detailed knowledge of engineeringapplications to sensory and neuromuscular physiology and medicine.The student will be introduced to sensory and neuromuscularphysiology from an engineering perspective including equivalentcircuits and models. The student will also gain experience incollecting electrophysiological or other physiological signals and inanalyzing electroneurographic and electromyographic signals.

Objective:

S. Deutsch and A. Deutsch, Understanding the Nervous System: AnEngineering Perspective, IEEE Press, 1993. ISBN 0-87942-296-3.Copies will be ordered as required

Text:

J.G. Webster, Medical Instrumentation, Application and Design,Third Edition, Houghton Mifflin, 1998.A.C. Guyton, Basic Human Physiology, Normal Function andMechanisms of Disease, (a number of different editions andvariations )B. Katz, Nerve, Muscle and Synapse,R.B. Stein, Nerve and Muscle Membranes, cells and Systems,Plenum Press, 1980

References:

• Sensory and neuromuscular anatomy and physiology.• Acquisition and analysis of Electromyographic and Electro-

neurographic signals to determine normal and pathologicalneuromuscular function

• Models of the myelinated and unmyelinated nerves includingapplied stimulating electrical fields

• Electrical fields in tissue resulting from surface and subcutaneousapplied stimuli

• Surface and subcutaneous electrical fields in tissue resulting fromsingle or populations of active nerve or muscle fibres

• Models of neuromuscular control• Magnetic and electrical stimulation of neural structures• Functional Electrical Stimulation (FES) and Magnetic Stimulation

(FMS) in rehabilitation• Neuroprostheses and sensory system interfaces.

Course Outline:(Subject to Change)

Term Project (60%), Midtenn (20%), Final (20%)Grading:

IITerm:


Models of the Neuron

Instructor: Dr. Ian Bruce,ITB/A213, Ext. 26984.e-mail: [email protected]

To provide a solid conceptual and quantitative background in themodelling of biological neurons and how they function as computationaldevices. Practical experience will be gained in modelling neurons from anumber of perspectives, including equivalent electrical circuits, nonlineardynamical systems, and random point-processes, and an introduction tothe mathematics required to understand and implement these differentengineering methodologies will be given.

Objective:

C. Koch, Biophysics of computation: information processing in singleneurons, Oxford University Press, 1998. (ISBN: 0195104919)

Text:

References: P. Dayan and L. F. Abbott, Theoretical neuroscience, MIT Press, 2001.(ISBN: 0262041995)D. Johnston and S. M.-S. Wu, Foundations of cellular neurophysiology,MIT Press, 1994. (ISBN: 0262100533)C. Koch and L Segev, Methods in neuronal modeling - 2nd edition, MITPress, 1998. (ISBN: 0262112310)H. Wilson, Spikes decisions and actions: Dynamical foundations ofneuroscience, Oxford University Press, 1999. (Hdbk: ISBN0-19-852431-5; Pbk: ISBN 0-19-852430-7)W. Gerstner and W. Kistler, Spiking neuron models: single neurons,populations, plasticity, Cambridge University Press , 2002. (Hdbk: ISBN0-521-81384-0; Pbk: ISBN 0-521-89079-9) Link to online version.F. Rielce, D. Warland, R. de Ruyter van Steveninck, and W. Bialek,Spikes: exploring the neural code, MIT Press, 1996. (ISBN: 0262181746)D. L. Snyder and M. I. Miller, Random point processes in time and space,Springer-Verlag, 1991. (ISBN: 0387975772)S. H. Strogatz, Nonlinear dynamics and chaos: with applications inphysics, biology, chemistry, and engineering, Perseus Books, 2001.(ISBN: 0738204536)

There will be eleven 3-hour lectures, with the possibility of one extra, ifrequired.

Lectures:

A basic undergraduate understanding of electrical circuits, linear systems,ordinary and partial differential equations, probability and randomprocesses.

Prerequisite:

Introduction to Biological Neurons and Neural Computation (1 Lecture)Basic anatomy and physiology of neurons, membrane potential, spiking,spike propagation, synapses, excitation and inhibition, basics of neuralcomputation;Simple Detenninistic Models of Neural Excitation (2 Lectures)Integrate-and-fire models, discharge-rate models, simple neural networks;Stochastic Models of Neural Activity (2 Lectures)Poisson- and renewal-process models, random-walk models;Nonlinear Dynamical Models of Neural Excitation (3 Lectures)The Hodglcin-Huxley model, ionic channels, activation and inactivationstates, action potential generation, phase-plane analysis of neuralexcitability, nonlinear dynamics;Axons and Dendritic Trees (3 Lectures)Linear cable theory, modeling dendritic trees, action potentialpropagation, compartmental models.

Course Outline:

Assignments (45%); Midterm (25%); Final (30%).Grading:

IITerm:

2

Policy Reminders:Senate and the Faculty of Engineering require all course outlines to include the fol-lowing reminders:“The Faculty of Engineering is concerned with ensuring an environment that is freeof all adverse discrimination. If there is a problem, that cannot be resolved by discussionamong the persons concerned, individuals are reminded that they should contact theDepartment Chair, the Sexual Harassment Officer or the Human Rights Consultant, assoon as possible."“Students are reminded that they should read and comply with the Statement onAcademic Ethics and the Senate Resolutions on Academic Dishonesty as found in theSenate Policy Statements distributed at registration and available in the Senate Office.”“The instructor and university reserve the right to modify elements of the courseduring the term. The university may change the dates and deadlines for any or allcourse in extreme circumstances. If either type of modification becomes necessary,reasonable notice and communication with the students will be given with explanationand the opportunity to comment on changes. It is the responsibility of the student tocheck their McMaster email and course websites weekly during the term and to noteany changes.”“Academic dishonesty consists of misrepresentation by deception or by other fraudu-lent means and can result in serious consequences, e.g. the grade of zero on anassignment, loss of credit with a notation on the transcript (notation reads: “Grade of Fassigned for academic dishonesty"), and/or suspension or expulsion from the university.It is your re-sponsibility to understand what constitutes academic dishonesty. Forinformation on the various kinds of academic dishonesty please refer to the AcademicIntegrity Policy, specifically Appendix 3, located athttp://www.mcmaster.ca/senate/academic/ac integrity.htmThe following illustrates only three forms of academic dishonesty:

1. Plagiarism, e.g. the submission of work that is not one’s own or for which othercredit has been obtained. (Insert specific course information, e.g. style guide)2. Improper collaboration in group work. (Insert specific course information)3. Copying or using unauthorized aids in tests and examinations.(If applicable) In this course we will be using a software package designed to revealplagiarism. Students will be required to submit their work electronically and in hardcopy so that it can be checked for academic dishonesty.”

Documents

DEPARTMENT OF ELECTRICAL AND COMPUTER ENGINEERING · Electrical and Computer Engineering 704 Advanced Engineering Mathematics (MEng or MASc only)Dr. T. Field Email: [email protected]