Day-1: 13th September 2021, 8am-12pm PST

Title of the Lecture: LDPC based Advanced Error Correction Coding and 5G

Low-Density Parity-Check (LDPC) codes are now being used in Hard disk drive read channels, Wireless (IEEE 802.11n/ IEEE 802.11ac, IEEE 802.16e WiMax), 10-GB, DVB-S2, Flash SSD, and more recently in 5G-NR cellular radio. This lecture covers Low-Density Parity-Check (LDPC) code-based Advanced Error Correction Coding Algorithms and Architectures. LDPC codes now have been firmly established as coding techniques for communication and storage channels. This talk gives an overview of the development of low complexity iterative LDPC solutions for communication channels. Complexity is reduced by developing new or modified algorithms and new hardware architectures.

Day-2: 14th September 2021, 8am-12pm PST

Title of the Lecture: LDPC based Advanced Error Correction Coding and 5G

This lecture gives an overview of HLS tools and evaluation of HLS flow vs. manually optimized traditional RTL design flow for LDPC decoders. Then this talk gives an overview of LDPC implementations for 5G-NR on various platforms such as GPUs, ASICs, and FPGAs.

Day-3: 15th September 2021, 8am-12pm PST

Title of the Lecture: GPU Acceleration for 5G Signal Processing and Machine Learning

As the rollout of 5G progresses and research for 6G begins, the key themes of softwarization, virtualization, open systems and artificial intelligence form foundational principles for communication systems of the future.

The application of AI/MLto wireless communications an extremely active research area with many 10’s to 100’s of papers published weekly reporting new results on the application of AI/ML to the physical layer (L1), MAC layer (L2) and at the network optimization level.

To realize the Industry’s vision of an AI/ML powered wireless future, a full stack solution supporting a software defined radio (SDR) approach for the vRAN, together with optimized silicon for AI, coupled with application development frameworks for AI/ML development is essential. NVIDIA GPU technology and associated CUDA programming model, together with arich suite of AI/ML SDKs (Software Development Kits) provides these capabilities.

In this talk we present The Aerialsoftware-defined GPU-based cloud native 5G NR RAN platform. Aerial implements not only 5G NR the baseband signal processing, but using GPU virtualization supports additional concurrently operating workloads, such as AI/ML inference, training and data analytics on this one hyper converged system. We provide an overview of the L1 signal processing pipeline and describe efficient mechanisms for data movement between the GPU and NIC-based fronthaul interface using a GPU-enabled Data Plane Development Kit (DPDK). A brief survey of some of the promising deep learning approachesfor L1 and L2 enhancements is presented.


To realize the Industry’s vision of an AI/ML powered wireless future, a full stack solution, spanning silicon to application development frameworks.

This platform is the nexus for future generation wireless that will be built on the foundation of artificial intelligence and machine learning for the vRAN itself and also for the application layer.

Day-4: 16th September 2021, 8am-12pm PST

Title of the Lecture: Building low latency and low power Smart City Applications

Computer vision and machine learning technology have advanced considerably in the past few years and are now being fused together to create a Smarter World of solutions that will soon improve our lives. For example, collectively retail stores lose billions of dollars annually, the National Retail Federation estimates that it was up to $60B in 2019. From Smart Cities to Factories, Hospitals, and Buildings, our world is experiencing an explosion of computational power being applied to real-time video analytics. Imagine a missing child is spotted by a SmartCity application and returned to their parents. In a hospital setting, the application monitoring the cameras in a patient’s room alerts the staff to a critical situation minutes or even seconds earlier. However, the entire end-to-end pipeline needs to be built for supporting low latency and low power for the edge where Smart city applications reside. Wireless 5G networks provide the fabric on which the video and AI/ML-driven computations and results travel. With latency targets for these smart technologies in the 10s of milliseconds, no other broadband network is up to the task. ML and video, combined with 5G, are making the Smart City solution deployable and more cost-effective. In this presentation, we will discuss what technologies are available to build the optimal Smart City solutions and what the differences are.

Day-5: 17th September 2021, 8am-12pm PST

Title of the Lecture: Machine Learning for Next Generation Wireless Communication Systems 5G/6G

With the emergence of fifth-generation (5G) networks, there has been a shift in the research focus towards exploring new technologies for the next-generation communication systems, sixth-generation (6G). The potential target expectations from 6G are to achieve even higher data rates, further reduction in latency and ultra massive machine type connection density compared to 5G. In this search for new technologies, there has been a significant interest in applying machine learning and artificial intelligence to communication systems.

In this lecture, we motivate the AI/ML for wireless communications by starting with simple machine learning applications and similarities to communication use cases. During the course, you will learn about various wireless communication system blocks and the application of ML to them. We cover the applications at the physical and higher layers at both the base station and user equipment of the wireless communication system. Later we briefly discuss the possibility of an end-to-end conventional communication system replaced by an ML-trained communication system. We will observe that ML/AI does not give benefits all the time.

In the later part of this lecture, we cover model-based and model-free systems and how they evolve with continuous improvements. We will further discuss and exercise a step-by-step procedure on designing an ML application for a wireless communication system with the constraints of complexity, timing, performance, and training requirements.

By the end of this lecture, you will have learned about various blocks of communication systems that ML-trained systems can replace and where to apply and where not to apply ML/AI in wireless communication systems.