This Summer School will be organized in line with international summer schools conducted by the IEEE Communications Society. As the event title suggests, the theme this year will revolve around the emerging use of machine learning and artificial intelligence in the design of next-generation communication networks. Targeting graduate students and researchers, the objective of the event is to equip the audience with the technical tools necessary to carry out cutting-edge research in this important direction. Staying true to the event title of “school”, the event will attempt to achieve this objective through concerted teaching sessions coupled with tutorial-type talks in advanced research topics from international and local speakers.
Final Year Students
IEEE and Luminite: 2000 PKR
Other Students: 2500 PKR
Faculty/Professionals: 5000 PKR
Final Year Students
Dates: Aug 07 – Aug 10, 2018
Timings: 9:30am – 5:30pm
Dr. Shahid Masud
Professor - Dean SBASSE, LUMS
Dr. Momin Uppal
Focal Person / Associate Professor, LUMS
Dr. Wasif Tanveer
Patron, IEEE / Assistant Professor, LUMS
Mr. Kashif Basheer
Chair IEEE COSMOC Lahore / Manager HUTIC UET Lahore.
Organizing Committee (click to view)
|Names||Affiliation||Designation in Event|
|Dr. Momin Uppal||Associate Professor,
EE Dept, LUMS.
|Dr. Wasif Tanveer||Patron IEEE LUMS.
EE Dept, LUMS.
|Member Advisory Committee|
|Dr. Shahid Masud||
Dean SBASSE LUMS.
|Chair Advisory Committee|
|Dr. Tariq Jadoon||
Head of EE,
|Co-Chair Advisory Committee|
|Dr. Ijaz Naqvi||
EE Dept, LUMS.
|Member Advisory Committee|
|Mr. Kashif Bashir||Chair IEEE COSMOC Lahore Section.
Manager HUTIC UET Lahore.
Mr. Affan Ahmad
|Senion Officer EE Dept. LUMS||Chair Program Committee|
Mr. Naeem ur Rehman
|Administrative Assistant||Administrative Head|
Mr. Usman Munawar
|Secretary IEEE Lahore Section||Organizing Committee|
Chair IEEE LUMS
|Chair Core Team|
|Azan Bin Zahid||Vice Chair IEEE LUMS||Co-Chair Program Committee|
|M. Usama Rizwan||Convener IEEE LUMS||Chair Publication Committee|
|Subbul Zaid||Treasurer IEEE LUMS||Chair Finance Committee|
|M. Mufeez Malik||General Secretary IEEE LUMS||Co-Chair Core Team|
|Usman Muhammad Ali||SAC Chair IEEE Lahore Section||Chair Student Participation Committee|
Dr. Ali Imran
Assistant Professor, University of Oklahoma-Tulsa, USA
Dr. Faqir Zarrar Yousaf
NEC Laboratories Europe | NEC · 5G Networks Group
Dr. Usman Javed
Global Technology Director for Internet of Things (IoT) at Vodafone Group, UK
Dr. Ali Raza Zaidi
University Academic Fellow, University of Leeds, UK
TALK ABSTRACT OF FOREIGN INVITED SPEAKERS
|Name:||Dr. Ali Imran|
|Title:||Leveraging AI for Designing Lean, Elastic, Agile and Proactive (LEAP) Wireless Networks of the Future: Case Studies, Challenges and Opportunities|
|Abstract:||This tutorial is aimed to interactively share with the summer school audience very timely answers to following questions: 1) How and why the plethora of diverse requirements emerging from a motley of emerging human to human and IoT applications, can be essentially boiled down to four fundamental design features i.e., lean, elastic, agile and proactive (LEAP) Network, and how these four requirements can be characterized quantitatively and qualitatively?;2) What constraints these four fundamental requirements impose on the design space of future wireless networks?; 3) What are AI enabled CDSA and BSON, how a cellular system design based on AI enabled CDSA and BSON can meet LEAP requirements?; 4) What is dark data and how it can be leveraged to achieve AI for the LEAP future cellular networks?. 5) What machine learning methods have immediate applications for enabling LEAP Networks. 6) How does an AI-based LEAP network design fare in terms of different KPIs against a conventional design such as legacy SON based HetNets?.The tutorial will also include an in-depth coverage of carefully selected machine learning tools that have been shown to be highly useful towards LEAP network design. These tools include anomaly detection, transfer learning, multi-arm in bandit learning and Markov chains, deep learning, entropy field decomposition and more. The tutorial will leverage insights from very recent research conducted in several ongoing NSF-USA and Industry funded projects led by the presenter.|
|Name:||Dr. Ali Zaidi|
|Title:||Emerging Technologies for Intelligent Communication Network Design and Deployment|
|Abstract:||In recent past, we have envisioned tremendous growth in research and development activities to realize ultra-reliable low latency wireless connectivity. Such wireless connectivity is key enabler for what is known as ‘Internet-of-Things’. This talk will focus on recent technological developments and highlight open issues which must be tackled to enable ubiquitous communication. In particular, we will explore role of robotics and autonomous systems in facilitating physical infrastructure mobility. We will also discuss key aspects of edge and fog computing which can be exploited for the overall network optimization.|
|Name:||Dr. Usman Javaid|
|Title:||Internet of Things (IoT) – A network which makes things come alive and creates a digitally connected society|
|Abstract:||With the Internet of Things (IoT), it’s the first time as an industry we are not building technology for humans but for machines to ‘digitally’ transform our lives and businesses. In this tutorial, we will study fundamental concepts behind Internet of Things (IoT) technology, business models, industry ecosystem, value chain and market dynamics. With real examples of IoT applications in different vertical industries, we will ‘bring IoT to life’ and learn how IoT is helping to solve complex business problems. In the end, we will have an open forum to draw out conclusions, innovation topics and potential research areas. If you are excited about the potential offered by Digital technologies like IoT, this is an opportunity not to missed!|
|Name:||Dr. Zarrar Yousuf|
|Title:||NFV and SDN – Key Technology Enablers for 5G Networks|
|Abstract:||Communication networks are undergoing their next evolutionary step towards 5G. The 5G networks are envisioned to provide a flexible, scalable, agile and programmable network platform over which different services with varying requirements can be deployed and managed within strict performance bounds. In order to address these challenges a paradigm shift is taking place in the technologies that drive the networks, and thus their architecture. Innovative concepts and techniques are being developed to power the next generation mobile networks. At the heart of this development lie Network Function Virtualization and Software Defined Networking technologies, which are now recognized as being two of the key technology enablers for realizing 5G networks, and which have introduced a major change in the way network services are deployed and operated. For interested readers that are new to the field of SDN and NFV this paper provides an overview of both these technologies with reference to the 5G networks. Most importantly it describes how the two technologies complement each other and how they are expected to drive the networks of near future.|
List of National Invited Speakers
|1||Dr. Momin Uppal||LUMS|
|2||Dr. Muhammad Tahir||LUMS|
|3||Dr. Ubaid Ullah Fayaz||UET, Lahore|
|4||Dr. Junaid Qadir||ITU|
|5||Dr. Ali Ahmad||ITU|
|6||Dr. Mehboob Ur Rehman||ITU|
Dr. Muhammad Tahir
Title: Cross roads between signal processing and machine learning
Abstract: In this talk, we will discuss how we approach modern machine learning trends from signal processing point of view. We will discuss signal processing techniques – mostly on time series data – which are in common practice for machine learning applications. Mostly, the effectiveness of any machine learning algorithm depends on choosing the best representation of the available data either through transformation, feature extraction, de-noising or dimensionality reduction. Signal processing can be viewed as a set of tools to achieve this purpose. Using real signal and data, we will demonstrate a number of concepts for obtaining insightful data analysis using signal processing tools.
Dr. Momin Ayub Uppal
Title: OFDM: The theory of Practice
Abstract: Orthogonal frequency division multiplexing (OFDM) has become the de-facto waveform for modern-day wideband wireless communication. Indeed, OFDM is the underlying waveform in many communication standards including, but not limited to, LTE, WiFi, WiMax, DSL, and power line networks. The overarching goal of this short yet rigorous course will be to enable the participants to design, build, and customize their own implementation of an OFDM-based physical layer – an implementation that could readily be deployed for real-world over-the-air transmission of information using software defined radios. Going beyond a typical academic instruction of OFDM, the course contents will be geared towards (a) modeling real-world channel artifacts (e.g. Doppler, carrier offsets, timing mismatch, multipaths, attenuation) using a mathematical framework, and (b) using the mathematical models to design and implement efficient algorithms to compensate for these artifacts. While the concepts covered are applicable to any OFDM-based physical layer, the instruction will primarily use the IEEE 802.11 (WiFi) physical layer as a case study. The goal described above will be achieved by a careful mix of in-class chalk-and-board instruction of theory, algorithms, and computer-based demonstrations.
Dr. Mahboob ur Rehman
Title: Towards intelligent receiver design for nano-scale communication systems operating in Terahertz band.
Abstract: We initiate the discussions on the design an intelligent/cognitive nano receiver operating in Terahertz (THz) band. Specifically, we investigate two essential ingredients of an intelligent nano receiver—modulation mode detection (to differentiate between pulse-based modulation and carrier based modulation), and modulation classification (to identify the exact modulation scheme in use). To implement modulation mode detection, we construct a binary hypothesis test in nano-receiver’s passband, and provide closed-form expressions for the two error probabilities. As for modulation classification, we aim to represent the received signal of interest by a Gaussian mixture model (GMM). This necessitates the explicit estimation of the THz channel impulse response, and its subsequent ompensation (via deconvolution). We then learn the GMM parameters via Expectation-Maximization algorithm. We then do Gaussian approximation of each mixture density to compute symmetric Kullback-Leibler divergence in order to differentiate between various modulation schemes (i.e., M-ary phase shift keying, M-ary quadrature amplitude modulation). The simulation results on mode detection indicate that there exists a unique Pareto-optimal point (for both SNR and the decision threshold) where both error probabilities are minimized. The main takeaway message by the simulation results on modulation classification is that for a pre-specified probability of correct classification, higher SNR is required to correctly identify a higher order modulation scheme.
On a broader note, this work aims to trigger the interest of the community in the design of intelligent/cognitive nano receivers (capable of performing various intelligent tasks, e.g., modulation prediction etc.)
Dr. Ali Ahmad
Title: Blind Deconvolution using Deep generative priors
Abstract: We consider the problem of recovering two unknown vectors from their circular convolution. We make the structural assumption that the two vectors are members of know subspaces. Although the observed convolution is nonlinear in the unknowns, it is linear in the rank-1 matrix formed by their outer product. This observation allows us to recast the deconvolution problem as low-rank matrix recovery problem from linear measurements, whose natural convex relaxation is a nuclear norm minimization program. We discuss this result in the context of blind channel estimation in communications. We will then focus on a data driven approach to handle the blind deconvolution problem, where we employ a deep neural network to act as a regularizer in the inverse problem. We showcase state of the art image deblurring results obtained using this approach.
Dr. Junaid Qadir
Title: Analyzing Self-Driving Networks: A Systems Thinking Approach
Abstract: Along with recent networking advances (such as software-defined networks, network functions virtualization, and programmable data planes), the networking field in a bid to construct highly optimized self-driving and self-organizing networks is increasingly embracing artificial intelligence and machine learning. It is worth remembering that the modern Internet that interconnects millions of networks is a `complex adaptive social system’, in which interventions not only cause effects but the effects have further knock-on effects (not all of which are desirable or anticipated). We believe that self-driving networks will likely raise new unanticipated challenges (particularly in the human-facing domains of ethics, privacy, and security). In this talk, we propose the use of insights and tools from the field of “systems thinking”—a rich discipline developing for more than half a century, which encompasses nonlinear models of complex social systems—and highlight their relevance for studying the long-term effects of network architectural interventions, particularly for self-driving networks.
Dr. Ubaid Ullah Fayyaz
Title: Factor Graphs in Channel Coding
Abstract: In communication receivers, we are interested in finding marginal distributions for transmitted symbols, given channel observations. The marginal distributions usually come from a multivariate distribution which can further be factored into multiple functions. Factor graphs represent factorization of a function and help in efficiently calculating such marginal functions. Almost all modern channel decoders such as belief propagation decoders for low-density parity-check (LDPC) and polar codes heavily depend on factor graph representations of their decoders. In this tutorial talk, we will explore the basics of factor graphs and their application in modern channel coding theory.
The event will provide a platform for graduate students, faculty members, and industry professionals from across the country to share ideas in the emerging research area of the use of artificial intelligence in designing future generation communication networks. Most importantly, it will provide an opportunity to graduate students from across the country to not only hone their fundamental concepts, but also to explore the important research directions being pursued across the globe.