Teaching

Fall 2014

EE 562a: Random Processes in Engineering

Course information:

Venue: Monday-Wednesday 11:00-12:20 pm, OHE 100B

Discussion Section: Friday 5:00-5:50 pm, OHE 100C

Office Hours: Monday 12:30-2:30 pm, EEB526, or by appointment

Webpage: blackboard at https://www.uscden.net

Grading: In-class participation (3%), Homework (17%), Midterm 1 (25%), Midterm 2 (25%), Final Exam (30%)

Textbook: Probability, Statistics, and Random Processes For Electrical Engineering, Third Edition, by Alberto Leon-Garcia. Prentice Hall, 2008.

Course description:

Electrical and computer engineers have played a central role in the design of modern information and communication systems. These highly successful systems work reliably and predictably in highly variable and chaotic environments. Probability models are one of the tools that enable the designer to make sense out of the chaos and to successfully build systems that are efficient, reliable, and cost effective. This course describes the theory underlying probability models as well as the basic techniques used in the development of such models.

In particular we will cover the following topics:

Topics:

  1. Review probability (chapters 1-6)

  2. Jointly Gaussian random variables (section 6.4)

  3. Estimation (section 6.5)

  4. LLN and CLT (chapter 7)

  5. Random processes (chapter 9)

  6. Stochastic signal processing (chapter 10)

  7. Markov chains (chapter 11)

  8. Introduction to queuing theory  (sections 12.1-12.3)


Spring 2013

ECE 5630: Fundamentals of Information Transmission

Course information:

Venue: Tuesday-Thursday 1:25-2:40 pm, PHL 407

Office Hours: Tuesday 2:45-4:45 pm, RH 325, or by appointment

Webpage: http://blackboard.cornell.edu

Grading: In-class participation (3%), Homework (17%), Prelim (30%), Project (20%), Final Exam (30%)

Textbook: Elements of Information Theory, Second Edition, by T. M. Cover and J. A. Thomas, 2006.

Pre-requisites: Probability and Random Process (ECE 4100).

Course description:

This course covers fundamental theories and algorithms for reliable information transfer over communication channels and networks. In particular, the course covers the following topics: entropy and information measures, lossless data compression, typicality and joint typicality, channel coding theorem, low density parity check codes and fountain codes, differential entropy, additive Gaussian noise channels, multiple access channels, broadcast channels, relay channels, information flow over wireline relay networks, deterministic modeling of wireless networks, and information flow over deterministic wireless relay networks, linking systems and generalization of max-flow min-cut theorem to deterministic relay networks, and several open problems beyond unicast and multicast information flow.

In particular we will cover the following topics:

Topics:

  1. Entropy and Information Measures

  2. Entropy Inequalities

  3. Optimal Data Compression

  4. Asymptotic Equipartition Property

  5. Channel Coding Theorem

  6. Differential Entropy and Entropy Maximization

  7. Additive Gaussian Noise Channel

  8. Bandlimited Channel, Parallel Channel, and Channels with Feedback

  9. Multiple Access Channels

  10. Broadcast Channels

  11. Relay Channels

  12. Information Flow over Wireline Relay Networks

  13. Deterministic Modeling of Wireless Channels

  14. Information Flow over Wireless Deterministic Relay Networks

  15. Linking Systems and its Connection of Wireless Network Information Flow

  16. Beyond Unicast and Multicast in Wireless Networks


Fall 2012

ECE 4110: Random Signals in Communications and Signal Processing

Course information:

Venue: Tuesday-Thursday 1:25-2:40 pm, PHL 407

Discussion Section: Wednesday 7:30-8:20 pm, PHL 219 (by prior notice only)

Office Hours: Tuesday 2:45-4:45 pm, RH 325, or by appointment

Webpage: http://blackboard.cornell.edu

Grading: In-class participation (3%), Homework (17%), Prelim 1 (25%), Prelim 2 (25%), Final Exam (30%)

Textbook: Probability, Statistics, and Random Processes For Electrical Engineering, Third Edition, by Alberto Leon-Garcia. Prentice Hall, 2008.

Pre-requisites: Signals and Information (ECE 2200) and Probability (ECE 3100).

Course description:

Electrical and computer engineers have played a central role in the design of modern information and communication systems. These highly successful systems work reliably and predictably in highly variable and chaotic environments. Probability models are one of the tools that enable the designer to make sense out of the chaos and to successfully build systems that are efficient, reliable, and cost effective. This course describes the theory underlying probability models as well as the basic techniques used in the development of such models.

In particular we will cover the following topics:

Topics:

  1. Review probability (chapters 1-6)

  2. Jointly Gaussian random variables (section 6.4)

  3. Estimation (section 6.5)

  4. LLN and CLT (chapter 7)

  5. Random processes (chapter 9)

  6. Stochastic signal processing (chapter 10)

  7. Markov chains (chapter 11)

  8. Introduction to queuing theory  (sections 12.1-12.3)


Spring 2012

ECE 4670/5670: Digital Communications

Course information:

Venue: Tuesday-Thursday 8:40-9:55, PHL 403

Office Hours: Tuesday 10:00-12:00, RH 325, or by appointment

Webpage: http://blackboard.cornell.edu

Course Material: Lecture notes will be provided

Pre-requisites: Signals and Information (ECE 2200), 

                        Probability and random processes (ECE 3100, ECE 4110 )

Instructor: Prof. Salman Avestimehr. 325 Rhodes Hall. x5-9915, avestimehr [at] ece                   

Course description:

Digital communication systems are the backbone of today’s Information High-way. Examples include the Internet, advanced wireless and wireline communication systems such as DSL, Cellular and WiFi, compact discs, etc. This course illustrates the basic principles underlying the design and analysis of digital communication systems. . In particular we will cover the following topics:

Statistical channel model

MAP and ML detection - Calculating the detection error probability

Modulation schemes

Sequential and block communication

Energy-efficient and rate-efficient communication 

Reliable communication with erasures -- Linear block codes -- Hard and soft decoding

Capacity of the Gaussian channel

Signal space and orthonormal basis

Optimal receiver for waveform channels -- Matched filtering

Passband transmission

Channels with memory -- Pulse shaping and sampling

Inter Symbol Interference (ISI) and Maximum Likelihood Sequence Detection (MLSD)

The Viterbi Algorithm

Symbol by symbol detection

Equalization

Orthogonal Frequency Division Multiplexing (OFDM) 

Complex baseband representation of passband channels

Modeling of multipath wireless channels

Delay spread -- coherence bandwidth -- coherence time -- doppler spread

Channel fading -- Non-coherent and coherent detection in flat fading channels

Time and antenna diversity techniques -- Alamouti scheme

Single-carrier systems with ISI equalization

Direct sequence spread spectrum

Fall 2011

ECE 4110: Random Signals in Communications and Signal Processing

Course information:

Venue: Tuesday-Thursday 1:25-2:40 pm, PHL 307

Discussion Section: Wednesday 7:30-8:20 pm, PHL 403

Office Hours: Tuesday 2:45-4:45 pm, RH 325, or by appointment

Webpage: http://blackboard.cornell.edu

Grading: In-class participation (3%), Homework (17%), Prelim 1 (25%), Prelim 2 (25%), Final Exam (30%)

Textbook: Probability, Statistics, and Random Processes For Electrical Engineering, Third Edition, by Alberto Leon-Garcia. Prentice Hall, 2008.

Pre-requisites: Signals and Information (ECE 2200) and Probability (ECE 3100).

Teaching Assistant: Sina Lashgari (sl2232 [at] ece)

TA Office Hours: Wednesday 4:30-6:30 pm, PHL 429

Course description:

Electrical and computer engineers have played a central role in the design of modern information and communication systems. These highly successful systems work reliably and predictably in highly variable and chaotic environments. Probability models are one of the tools that enable the designer to make sense out of the chaos and to successfully build systems that are efficient, reliable, and cost effective. This course describes the theory underlying probability models as well as the basic techniques used in the development of such models.

In particular we will cover the following topics:

Topics:

  1. Review probability (chapters 1-6)

  2. Jointly Gaussian random variables (section 6.4)

  3. Estimation (section 6.5)

  4. LLN and CLT (chapter 7)

  5. Random processes (chapter 9)

  6. Stochastic signal processing (chapter 10)

  7. Markov chains (chapter 11)

  8. Introduction to queuing theory  (sections 12.1-12.3)


Spring 2011

ECE 5670: Digital Communications

Course information:

Venue: Tuesday-Thursday 8:40-9:55, PHL 403

Office Hours: Tuesday 10:00-12:00, RH 325, or by appointment

Webpage: http://blackboard.cornell.edu

Grading: Homework (30%), Midterm (30%), Final Exam (40%)

Course Material: Lecture notes will be provided

Pre-requisites: Signals and Information (ECE 2200), 

                        Probability and random processes (ECE 3100, ECE 4110 )

Instructor: Prof. Salman Avestimehr. 325 Rhodes Hall. x5-9915, avestimehr [at] ece                   

Course description:

Digital communication systems are the backbone of today’s Information High-way. Examples include the Internet, advanced wireless and wireline communication systems such as DSL, Cellular and WiFi, compact discs, etc. This course illustrates the basic principles underlying the design and analysis of digital communication systems. . In particular we will cover the following topics:

Statistical channel model

MAP and ML detection - Calculating the detection error probability

Modulation schemes

Sequential and block communication

Energy-efficient and rate-efficient communication 

Reliable communication with erasures -- Linear block codes -- Hard and soft decoding

Capacity of the Gaussian channel

Signal space and orthonormal basis

Optimal receiver for waveform channels -- Matched filtering

Passband transmission

Channels with memory -- Pulse shaping and sampling

Inter Symbol Interference (ISI) and Maximum Likelihood Sequence Detection (MLSD)

The Viterbi Algorithm

Symbol by symbol detection

Equalization

Orthogonal Frequency Division Multiplexing (OFDM) 

Complex baseband representation of passband channels

Modeling of multipath wireless channels

Delay spread -- coherence bandwidth -- coherence time -- doppler spread

Channel fading -- Non-coherent and coherent detection in flat fading channels

Time and antenna diversity techniques -- Alamouti scheme

Single-carrier systems with ISI equalization

Direct sequence spread spectrum


Fall 2010

ECE 4110: Random Signals in Communications and Signal Processing

Course information:

Venue: Monday-Wednesday 10:10-11:25, PHL 403

Office Hours: Wednesday 12:30-2:30, RH 325, or by appointment

Webpage: http://blackboard.cornell.edu

Grading: In-class participation (3%), Homework (17%), Prelim 1 (25%), Prelim 2 (25%), Final Exam (30%)

Textbook: Probability, Statistics, and Random Processes For Electrical Engineering, Third Edition, by Alberto Leon-Garcia. Prentice Hall, 2008.

Pre-requisites: Signals and Information (ECE 2200) and Probability (ECE 3100).

Teaching Assistant: Alireza Vahid (av292 [at] ece)

TA Office Hours: Thursday 4:00-6:00, PHL 201

Course description:

Electrical and computer engineers have played a central role in the design of modern information and communication systems. These highly successful systems work reliably and predictably in highly variable and chaotic environments. Probability models are one of the tools that enable the designer to make sense out of the chaos and to successfully build systems that are efficient, reliable, and cost effective. This course describes the theory underlying probability models as well as the basic techniques used in the development of such models.

In particular we will cover the following topics:

Topics:

  1. Review probability (chapters 1-6)

  2. Jointly Gaussian random variables (section 6.4)

  3. Estimation (section 6.5)

  4. LLN and CLT (chapter 7)

  5. Random processes (chapter 9)

  6. Stochastic signal processing (chapter 10)

  7. Markov chains (chapter 11)

  8. Introduction to queuing theory  (sections 12.1-12.3)



Spring 2010

ECE 5980: Advanced Topics in Network Information Flow

Course information:

Venue: Monday-Wednesday 2:55--4:10, Room PHL 407

Webpage: http://blackboard.cornell.edu

Format: Seminar. no exam. projects and presentations

Course Material: Research papers and book chapters

Pre-requisites: Probability and random processes (ECE 4110 or equivalent), 

                        Information Theory (those interested are encouraged to take ECE 5620 simultaneously)

Instructor: Prof. Salman Avestimehr. 325 Rhodes Hall. x5-9915, avestimehr [at] ece

Course description:

Claude Shannon characterized the fundamental limit of reliable communication over a point-to-point communication channel. Until recently, there has been only limited success in extending the theory to a network of interacting nodes. Progress has been made in the past decade, driven both by engineering interest in wireless networks as well as conceptual advances such as network coding. This seminar indents to cover these advances and provide a fresh perspective on the state-of-the-art of the field. Specific topics include recent generalizations of the Ford-Fulkerson Theorem, applications of “Matroid Theory” and “Linking Systems” in network information flow,  cooperation techniques in wireless relay networks, and interference management strategies in wireless networks.

Tentative Topics:

Single-source single-destination wireline networks. Ford-Fulkerson Theorem.

Multicast in wireline networks. Advantage of network coding over routing.

Linear network coding. Multicast capacity of wireline networks.

Algorithms for directed acyclic graphs. Algorithms for cyclic graphs.

Beyond multicast. Insufficiency of linear codes.

Two-unicast problem.

Wireless networks. Deterministic modeling of wireless channels.

Single-source single-destination deterministic wireless networks (i.e. wireless relay networks).

Capacity of deterministic relay networks. Generalization of Ford-Fulkerson Theorem.

Connections to Matroid Theory and Linking Systems.

Connections between the deterministic model for wireless channels and the Gaussian model.

Approximate capacity of Gaussian relay networks.

Beyond single--source single--destination wireless networks. Interference channel.

Capacity of the interference channel.



Spring 2010

ECE 5950: Probabilistic Methods in Communication Networks

Course information:

Venue: Friday 12-3:00, Room UPS 211

Webpage: http://blackboard.cornell.edu

Format: Seminar. no exam. projects and presentations

Course Material: Research papers and book chapters

Pre-requisites: Probability and random processes (ECE 4110 or equivalent), 

                        Information Theory (those interested are encouraged to take ECE 5620 simultaneously)

Instructors: Prof. Salman Avestimehr. 325 Rhodes Hall. x5-9915, avestimehr [at] ece

                   Prof. Lang Tong. 384 Rhodes Hall. x5-3900, ltong [at] ece

Course description:

The probabilistic method is a powerful tool in tackling many problems in discrete mathematics. Roughly speaking, the method works as follows: Trying to prove that a structure with desired property exists, one defines an appropriate probability space of structures and then shows that the desired properties hold in this space with positive probability. In this seminar series we aim to learn this powerful tool and investigate its application in communication theory problems.

Tentative Topics:

Illustration of the probabilistic method:

    Basic examples

    Distributing keys in multi-user crypto-systems

    Broadcasting in radio networks

    Leader-election game

    Local Lemma and simple applications of it

Linearity of Expectation and its Applications

Codes, Games and Entropy

Lovasz Local Lemma and its applications to routing and tracking

Correlation Inequalities

Martingales


Fall 2009

ECE 5670: Digital Communications

Course information:

Venue: Monday-Wednesday 10:10-11:25, THR 202

Office Hours: Wednesday 11:30-1:30, RH 325, or by appointment

Webpage: http://blackboard.cornell.edu

Grading: Homework (15%), Midterm (30%), Final Exam (40%), Project (15%)

Course Material: Lecture notes will be provided

Pre-requisites: Signals and Information (ECE 2200), 

                        Probability and random processes (ECE 3100, ECE 4110 (preferred not required))

Instructor: Prof. Salman Avestimehr. 325 Rhodes Hall. x5-9915, avestimehr [at] ece                   

Course description:

Digital communication systems are the backbone of today’s Information High-way. Examples include the Internet, advanced wireless and wireline communication systems such as DSL, Cellular and WiFi, compact discs, etc. This course illustrates the basic principles underlying the design and analysis of digital communication systems. . In particular we will cover the following topics:

Statistical channel model

MAP and ML detection - Calculating the detection error probability

Modulation schemes

Sequential and block communication

Energy-efficient and rate-efficient communication 

Reliable communication with erasures -- Linear block codes -- Hard and soft decoding

Capacity of the Gaussian channel

Signal space and orthonormal basis

Optimal receiver for waveform channels -- Matched filtering

Passband transmission

Channels with memory -- Pulse shaping and sampling

Inter Symbol Interference (ISI) and Maximum Likelihood Sequence Detection (MLSD)

The Viterbi Algorithm

Symbol by symbol detection

Equalization

Orthogonal Frequency Division Multiplexing (OFDM) 

Complex baseband representation of passband channels

Modeling of multipath wireless channels

Delay spread -- coherence bandwidth -- coherence time -- doppler spread

Channel fading -- Non-coherent and coherent detection in flat fading channels

Time and antenna diversity techniques -- Alamouti scheme

Single-carrier systems with ISI equalization

Direct sequence spread spectrum