Joint RWW/ARFTG Plenary Session
Tuesday, 18 January 2022 10:10 - 12:00
Artificial Intelligence and Machine Learning for the era of Hyperconnected IoT Devices
Speaker: Dr. Frank Schirrmeister, Cadence
We are at the cusp of an Era of Hyperconnectivity and Hyperscale Computing, fueled by billions of devices in the Internet of Things (IoT) and its industrial sibling, the Industrial IoT (IIOT). Ericsson predicts that data traffic through commercial networks grows to 164 exabytes per month in 2025. Video already accounted for 63% of the traffic of 2019’s 33 exabytes per month and will become 76% of the estimated 164 exabytes per month in 2025. At that time, 5G adoption could reach 2.8 billion subscriptions, and 5G population coverage is forecast at 55%. Seagate and IDC predict storage in the “global datasphere” to grow to 175 zettabytes by 2025, up from 45 zettabytes in 2019.
This presentation will discuss requirements for designing systems on chips (SoCs) and systems enabling the era of Hyperconnectivity. Specifically, we will and introduce solutions that the Electronic Design Automation (EDA) industry provides today for Artificial intelligence (AI) and machine learning (ML), as well as trends to address future challenges. In addition, we will also discuss how AI/ML technologies increase development productivity and optimize EDA design processes.
Specifically, we will discuss enabling processor and design IP and high-level synthesis to enable optimized circuitry for AI/ML algorithms. Furthermore, we will introduce the requirements for optimized AI/ML designs and specific verification tools for this design category. Advanced node and low power implementation are vital to linking verification to SoC implementation, and we will discuss particular optimizations and 3DIC and Chiplet based integration and analysis. To utilize AI/ML for EDA, we will introduce trends and experiences using AI/ML for formal verification, simulation, and implementation.
Frank Schirrmeister is senior group director, solutions & ecosystem at Cadence. He leads a team translating customer challenges in the hyperscale, communications, consumer, automotive, aerospace/defense, industrial, and healthcare vertical domains into specific requirements and solutions. In addition, his team focuses on cross-product technical solutions such as 5G, artificial intelligence, machine learning, safety, security, digital twins, and partner collaborations. Frank holds a Dipl.-Ing. in electrical engineering from the Technical University of Berlin, Germany. Before joining Cadence, Frank held senior engineering and product management positions in embedded software, semiconductor, and system development, both in Europe and the United States.
6G: Is it really DC to Daylight?
Speaker: Roger Nichols, Keysight Technologies, Inc.
By the time of this ARFTG event, the primary group driving 5G standardization will have finished the scoping of the 2nd update (Rel-18) to the implementable 5G standard. 5G brought new capabilities to commercial radio systems placing new demands on RF measurement. These include things like “massive MIMO”: active transceiver-chain and antenna systems that adapt to user need, physical location and movement, and user-density; taking new radio bands mainstream—everything from 3-7GHz to 28-52. These drove the need for complex measurements of active systems, wide bandwidths with complex modulation, and so much integration that over-the-air is the only feasible (and indeed the only standardized) technique to make some measurements. But even though most Rel-16 capabilities are not yet in production networks, the industry is in active dialogue and research on the sixth generation. 6G presentations are full of glorious descriptions of how wireless will improve society. While this all may be true, we technologists have plenty of work to do to ensure the system works. Like 5G, realizing 6G will take far more than simply adding new frequency bands and increasing the related bandwidths. This talk will cover an overview of the 6G vision and then dive into a few examples of how this vision will impact RF, microwave, and millimeter-wave systems and the associated challenges of measurements. Some of these are obvious (but difficult) next steps, like radio information bandwidths of at least 15 GHz, and some are not-so-obvious like AI-driven real-time determination of waveform and modulation schemes. I will attempt to cover examples that span DC-to-Daylight issues and show that analogy as closer to reality than what we have seen before.
Roger Nichols’ 37 years of engineering and management experience in wireless test and measurement at Hewlett-Packard, Agilent Technologies, and Keysight spans roles in marketing, manufacturing, and R&D. He has managed projects, programs, and departments beginning with analog cellular radio, evolving to 6G, and on every standard in between. He has been directing Keysight’s 5G program since 2014, and Keysight’s 6G program since 2019. He is also the director of Keysight’s work in wireless standards.
Roger holds a BSEE from the University of Colorado, Boulder.