Free Registration: https://sjsu.zoom.us/meeting/register/tZMlc-mspjItGNTBJ2LBC0aFKKw4bJIj1oC1 Synopsis: In this talk, the speaker will present potent solutions towards offloading reasoning subtasks in the case of mathematical problem solving: how to teach an auxiliary (and potentially frugal) language model to coordinate with black-box solvers, symbolic or language model-based, to successfully answer mathematical problems. The talk will focus on successfully teaching language models to perform reasoning from non-human feedback and how rewards beyond just the correctness of the final answer are essential for better learning Speaker(s): Dr. Chakraborty, Dr. Vishnu S. Pendyala Virtual: https://events.vtools.ieee.org/m/405310

The design of closed-loop controllers for high precision servo positioning systems in high-volume production is severely challenged by the tradeoff between the controller's robustness and performance. This presentation will review an industrial control design process for a population of dual-stage hard disk drive servo actuators. A multi-rate digital controller is developed to suppress high-frequency resonant modes and track low-frequency desired trajectories. The controller includes a proportional-integral controller cascaded with a set of notch and peak filters. The controller parameters are tuned via an optimization process based on a number of closed-loop system objectives such as stability margins and bandwidth, directly applied to the plants' frequency response data. To further enhance the performance of the closed-loop controller, the plant population is clustered into groups of similar plants, and separate controllers are designed for each cluster. Results indicate considerable improvements in the robustness and performance of the system using the clustering-based control design method. For long-range motions, the controllers can be supplemented with polynomial reference and feedforward input trajectories for fast and smooth transition from the initial to the target position. A new polynomial design process will be presented, which reduces the transient vibrations in long-range motion by incorporating the flexibility of the system in the reference trajectory. Simulations and experimental tests indicate significant improvements in the settling time of flexible servo systems using the proposed polynomial design framework. Speaker(s): Dr. Saeid Bashsah Agenda: 6:00 - 6:30 PM : Networking - Light dinner 6:30 - 7:30 PM : Talk and Q&A 7:30 - 8:00 PM : Networking Room: SCDI - 2301, Bldg: Sobrato Campus for Discovery and Innovation - SCDI, Santa Clara University, 500 El Camino Real , Santa Clara, California, United States, 95053

Abstract The talk will focus on Optical MEMS technology where the MEMS device is utilized for an Optical function. Optical MEMS is playing a critical role in many important areas such as Optical Networks, LIDAR, Virtual Reality, 3D Imaging, and various Biomedical applications. Devices that operate on the phase of an Optical beam require less movement and can be much faster. Examples of such devices include GLV (grating light valve), Diffractive MEMS, and Tunable Fabry Perot filters. Many applications are enabled by Micromirror devices that rotate in one or two axes. Different approaches are used depending on mirror size, speed, rotation angle, and whether used in a controlled or resonant mode. One or two dimensional arrays of mirrors are needed for some applications such as Optical Cross-connect Switches. The talk will also review design methodologies and process fabrication methods. Speaker(s): Dr. Asif A. Godil, Agenda: 6:30 – 7:00 PM Registration & Networking 7:00 – 7:45 PM Invited Talk 7:45 – 8:00 PM Questions & Answers Room: ECC1 (Conference Center), Bldg: Building E, 1st Floor, 2900 Semiconductor Drive, Santa Clara, California, United States, 95051, Virtual: https://events.vtools.ieee.org/m/408613

Abstract: Advancements in technology scaling have ushered in larger systems boasting enhanced functionality, increased operational speed, and expanded data bandwidth. However, these benefits come with more demanding clocking requirements, including extended distribution distances and heightened timing precision. Furthermore, technology scaling has rendered traditional analog design challenging. Wider PVT variations necessitate intensive calibration efforts, and increased integration levels call for resilience against external noise sources. Moreover, the fact that reference frequency and loop bandwidth do not scale at the same rate as technology leads to prohibitive costs for oversized loop filters. While pure analog implementations offer intuitive operation and elegant analysis, clocking circuits incorporating digital elements offer effective solutions to these challenges. This presentation will cover how digital circuits can enhance clock generation and distribution through techniques like calibration and signal processing. Beginning with well-established methods that harness the mixed-signal nature of PLLs, such as delta-sigma modulation for the MDD in fractional-N PLLs, the presentation will shift toward digital-intensive architectures. It will focus on techniques that leverage digital implementations for error detection and enhance timing accuracy through either analog or digital correction. State-of-the-art designs featuring runtime calibration and power noise cancellation for clock generation and distribution will also be introduced. This talk will conclude with insights into future challenges and trends. Speaker Bio: Ping-Hsuan Hsieh received the B.S. degree in electrical engineering from National Taiwan University, Taipei, Taiwan, in 2001, and the M.S. and Ph.D. degrees in electrical engineering from the University of California, Los Angeles, Los Angeles, CA, in 2004 and 2009, respectively. From 2009 to 2011, she was with the IBM T.J. Watson Research Center, Yorktown Heights, NY. In 2011 she joined the Electrical Engineering Department of National Tsing Hua University, Hsinchu, Taiwan, where she is currently an Associate Professor. Her research interests focus on mixed-signal integrated circuit designs for high-speed electrical data communications, clocking and synchronization systems, and energy-harvesting systems. Prof. Hsieh served in the Technical Program Committee of the IEEE International Solid-State Circuits Conference, and is currently a member of the Technical Program Committees of the IEEE Asian Solid-State Circuits Conference and the IEEE Custom Integrated Circuits Conference. She served as an Associate Editor for the IEEE Internet of Things Journal from 2014 to 2018, a Guest Editor for the IEEE Journal of Solid-State Circuits Special Issue in 2021, and is currently an Associate Editor for the IEEE Open Journal of Circuits and Systems and IEEE Solid-State Circuits Letters. Virtual: https://events.vtools.ieee.org/m/408965

Will AI replace 50% of jobs by 2030? It seems likely that humans must change what we do and adapt. With its widespread use, AI promises to allow humans to be released from simple repetitive tasks, and will surely start something new and creative. As a background, Hollywood Sci-Fi movies are biased toward “machines will destroy us,” but Japanese Sci-Fi cartoons have a different paradigm of friendly robots who help us. Friend or foe, which will it be? We will watch this interesting video and discuss how we think and feel about its message. Join us to discuss the implications of AI for management and leadership. Speaker(s): Tomo Noda, Carl Angotti Virtual: https://events.vtools.ieee.org/m/408814

You're cordially invited to an in-person professional networking event hosted by the IEEE Signal Processing Society (SPS) East Bay Chapter! Whether you're just starting your career, an experienced professional in industry or academia, or an entrepreneur, this event is the perfect opportunity to connect with fellow SPS members, expand your network, discover new opportunities, and stay updated on industry trends. Complimentary pizza and drinks will be provided for all attendees. To help us secure our reservation of the venue's private room, please RSVP at your earliest convenience. We look forward to seeing you there! Rocco's Ristorante Pizzeria, 2909 Ygnacio Valley Rd, Walnut Creek, California, United States, 94598

Agenda 3:30 PM Registration 4:00 PM Hands on Tutorials 5:00 PM LLM Talk 6:00 PM Panel: Future of AI 6:30 PM Dinner & Networking Speaker(s): Ipsita Mohanty, Mrinal Karvir, Nasibeh Nasiri, Andreea Munteanu, Nathalie Baracaldo Room: 180, Bldg: SC-12 Auditorium, 3600 Juliette Lane, Santa Clara, California, United States, 95054

This event will be available live at SEMI, as well as over Zoom. Get to SEMI by 6:30pm to network – and enjoy great pizza and refreshments! All attendees MUST register using the Eventbrite form (see link). This talk will explore the power of Google Cloud’s AlloyDB Omni, a groundbreaking database technology poised to shape 2024’s technological landscape using analytics acceleration and built-in machine learning for optimizing performance and managing complex workloads. This technology enables data platforms for dynamic business analytics and cutting-edge Generative AI (Gen AI) applications, and is compatible with the widely popular PostgreSQL open source database. This presentation will discuss how these platforms are deployed seamlessly across the Edge and Data Center Core on the Lenovo platform, and with Google Cloud for optimizing online transaction processing (OLTP) and real-time data analytics. Extensive vendor-neutral information will be shared as well. There will also be a deep dive into the fascinating realm of state-of-the-art Retrieval Augmented Generation (RAG) using Vector Databases. The talk will conclude with a live demonstration showcasing both Basic and Advanced RAG functionalities. Speaker(s): Prasad Venkatachar, 567 Yosemite Dr, Milpitas, California, United States, 95035, Virtual: https://events.vtools.ieee.org/m/403835

Swedish Neutral has pioneered what is arguably one of the most effective technologies for mitigating the risk of wildfires caused by powerlines: the Ground Fault Neutralizer (GFN). Some of the most devastating wildfires of the 21st century can be attributed to distribution powerlines. A practical approach to mitigate the ignition risk would involve transforming exposed conductors into insulated cables and placing powerlines underground. However, the associated expense of approximately $3M per mile of conductor — an economic burden too substantial for electric utility providers to bear. An alternative strategy involves substituting exposed overhead wires with covered conductors. Given the considerable cost of approximately $1-1.5M per mile, this solution does not offer a consistently reliable outcome. Costing just a fraction of the expense incurred by approaches above, the Ground Fault Neutralizer (GFN) system delivers robust results. As affirmed in the Functional Performance Report and Cost-Benefit Analysis published in 2020 by Victoria's Energy Safety Regulator (ESV), the comprehensive evaluation confirms that the operational performance of the deployed GFN systems aligns with expectations concerning bushfire risk mitigation. In certain cases, these installed GFNs have even surpassed expectations by effectively addressing more intricate faults and thereby diminishing the risk of bushfires. In addition to wildfire risk mitigation advantages, the GFN significantly increases reliability - obviates the necessity for Public Safety Power Shutoffs; and public safety (risk of electrocution) of energy distribution assets. Speaker(s): Viacheslav Levashov, , Virtual: https://events.vtools.ieee.org/m/409260

Joana Mendes will present a detailed background on current status of diamond film development. Artificial diamond films are deposited from a mixture of methane and hydrogen — and the deposition of the material is not an expensive process. However, the integration of diamond films and electronic devices requires the development and optimization of new processing lines, which is a costly procedure. LEDs, for instance, have become low-cost components. Are manufacturers willing to trade the current low cost for higher efficiency/performance and longer lifetime? Specific upcoming possibilities are for HEMT packaging, 5G/6G RF modules in cell phones, and in heterogeneous integration such as for high-power SiP modules. The second talk: Diamond Foundry, Inc. produces CVD single-crystal diamond (SCD) at scale, and recently achieved the historic milestone of the creation of the world’s first 100mm SCD wafer. SCD as heat spreader achieves performance enhancements of 3-6x and temperature reductions of 65C due to its unmatched high isotropic thermal conductivity (2,200W/m-K). Jeroen van Duren will describe the manufacturing of SCD dies and wafers and use for AI and power electronics packaging Speaker(s): Joana Catarina, Jeroen Virtual: https://events.vtools.ieee.org/m/388160

Quantum Semiconductor has invented atomistically-ordered superlattices made of Group IV elements (C, Si, Ge, Sn, Pb) strained to silicon surfaces, that have direct bandgaps across the infrared spectrum, and that can be monolithically integrated with CMOS. This approach overcomes the limitations of pure silicon devices - sensing light outside the visible range of wavelengths, and light emission. Also, the photon collection process is decoupled from CMOS junction engineering, thereby allowing these devices to track Moore’s Law with each new design generation, as well as use the most advanced substrates for state-of-the-art CMOS such as fully depleted thin-film SOI. This CMOS-compatible technology platform enables innovative new products for IR image sensing, photonics, and AI. Carlos Augusto, Ph.D., is a co-founder and the Chief Technology Officer of Quantum Semiconductor. A prolific inventor, he is responsible for Quantum Semiconductor’s core technology. Dr. Augusto has been in the semiconductor industry for over 25 years. Previously, he was at IMEC in Leuven, Belgium where he was a member of the research staff in the Advanced Silicon Devices Group and worked on the device, process, modeling and fabrication of SiGe Vertical MOSFETs and DRAMs. Agenda: In-Person Meeting. Register: (https://ieee.us6.list-manage.com/track/click?u=4f23d76911d28dcc51bf71bce&id=467218c0a6&e=d63c66b7c5) 11:30am: Networking & Pizza Noon-1PM: Seminar Cost: $4 to $6 Location: EAG Laboratories - (https://www.google.com/maps/search/810+Kifer+Road,+Sunnyvale?entry=gmail&source=g) ==> Use corner entrance: Kifer Road / San Lucar Court ==> Do not enter at main entrance on Kifer Road Bldg: ==> Use corner entrance: Kifer Road / San Lucar Court ==> Do not enter at main entrance on Kifer Road, 810 Kifer Road, Sunnyvale, California, United States, 95051

Discover the future of edge AI in our upcoming talk by Dr. Jun Zhang, an IEEE Fellow and Associate Professor at the Hong Kong University of Science and Technology. Delve into the shift from traditional data-oriented communications to task-oriented approaches, optimizing data transmission for specific inference tasks. Learn about the development of effective feature encoders and the introduction of EdgeGPT, an autonomous edge AI system. This presentation will highlight innovations in edge video analytics and mobile robotics, offering insights into achieving high accuracy and low latency in resource-constrained devices. Join us to explore cutting-edge strategies for enhancing edge computing solutions. Abstract Deep learning has achieved remarkable successes in many application domains, such as computer vision, image processing, and natural language processing. However, deploying powerful deep learning models on resource-constrained mobile devices (e.g., wearable or IoT devices) faces great challenges. Recently, edge AI techniques that rely on the emerging mobile edge computing platforms have been proposed, which forward intermediate features to be processed by a powerful edge server. To achieve high-accuracy and low-latency inference, effective feature encoders with low complexity and high compression capability will be needed. This calls for a paradigm shift in wireless communications, from “data-oriented communications”, which maximize data rates, to “task-oriented communications”, where the data transmission is an intermediate step to be optimized for the downstream inference task. This talk will introduce recent progresses on task-oriented communication for edge inference. An effective design principle based on information bottleneck will be firstly introduced, which will then be extended to multi-device cooperative perception based on a distributed information bottleneck framework. Use cases on edge video analytics and edge-assisted localization for mobile robots will be presented, followed by introduction of EdgeGPT, an autonomous edge AI system empowered by large language models. Bio: Jun Zhang received his Ph.D. degree in Electrical and Computer Engineering from the University of Texas at Austin. He is an IEEE Fellow and an IEEE ComSoc Distinguished Lecturer. He is an Associate Professor in the Department of Electronic and Computer Engineering at the Hong Kong University of Science and Technology. His research interests include wireless communications and networking, mobile edge computing and edge AI, and cooperative AI. Dr. Zhang co-authored the book Fundamentals of LTE (Prentice-Hall, 2010). He is a co-recipient of several best paper awards, including the 2021 Best Survey Paper Award of IEEE Communications Society, the 2019 IEEE Communications Society & Information Theory Society Joint Paper Award, and the 2016 Marconi Prize Paper Award in Wireless Communications. He also received the 2016 IEEE ComSoc Asia-Pacific Best Young Researcher Award. He is an Editor of IEEE Transactions on Communications and IEEE Transactions on Machine Learning in Communications and Networking, and was an editor of IEEE Transactions on Wireless Communications (2015-2020). Co-sponsored by: Pradeep Kumar Virtual: https://events.vtools.ieee.org/m/406837

3D integration technology enables heterogeneous system scaling by offering higher I/O densities, shorter interconnect lengths, higher bandwidth and smaller form factors compared to 2D packaging solutions. In recent years, many products have been released taking advantage of this technology to combine logic, memory and imaging and/or optical components into 3D stacked die, or 2.5D interposer configurations. Due to the vertical integration of thinned silicon chips, the strong thermal coupling between the tiers in the 3D stack, and the difficulty to remove heat from within the 3D die stack, thermal management is one of the major challenges of 3D integration technology. The incorporation of 3DIC and co-packaged optics on 2.5D interposer architectures to boost the compute throughput, has further complicated the cooling challenge. In this talk, the thermal impact of the recent scaling trends in 3D die and wafer stacking and CPO, including hybrid bonding, backside power delivery, and BEOL scaling will be discussed. Furthermore, the thermal opportunities of 3D functional partitioning and advanced liquid cooling solutions, to enable 2.5D HPC applications of multiple kW, will be addressed. Speaker(s): Herman Oprins, Virtual: https://events.vtools.ieee.org/m/411225

Administrative Committee meeting for planning for future Technical, Professional and other events. Virtual: https://events.vtools.ieee.org/m/413321

[]Blockchain is a distributed ledger technology that enables secure, transparent, and immutable recording of transactions across a network of computers. At its core, a blockchain consists of a chain of blocks, where each block contains a list of transactions. It requires network consensus to ensure validity. Once consensus is met, a block can be added to the ledger as an immutable record of transaction. This rather simple concept has already changed our transactional landscape a great deal since it emerged in popularity. Conceptually, blockchain technology has the potential to disrupt traditional systems, increase efficiency, transparency, and trust, and empower individuals and organizations across various domains. But, how did we get here? How is the history of Blockchain tied to Engineering and Technology? Is Blockchain only significant in the Data sciences and Computer Sciences Industries? Speaker(s): Amber Orr, Virtual: https://events.vtools.ieee.org/m/411663

Free but required Registration: https://sjsu.zoom.us/meeting/register/tZUlfumqpjotGtTrgOueGSpOTxC2L08dm2vW After registering, you will receive a confirmation email containing information about joining the meeting. Synopsis: The panel discussion will explore the ground-breaking potential of generative AI in revolutionizing medical care. Generative AI holds immense promise for transforming healthcare, but ethical considerations, thorough validation, and responsible implementation are essential. Careful attention to these challenges will unlock the full potential of AI to improve diagnosis, treatment, and overall patient care. However, navigating the ethical and practical challenges will be crucial to ensure equitable, responsible, and impactful implementation for the benefit of all. This expansion includes specific examples, highlights potential benefits beyond patient care, personalized diagnosis, prevention of infectious diseases and Medical image processing. The emphasis should be collaboration and stakeholder involvement in addressing the challenges. The panellists will delve into a range of crucial questions on leveraging the predictive power of GenAI for personalized medicine, enhanced diagnostics, ethical dilemmas, mental health support, privacy-preserving practices, trust and transparency, and more. Speaker(s): Ruchi, Dr. Vishnu S. Pendyala, Chinmay, Lakshmanan Virtual: https://events.vtools.ieee.org/m/407733

Existing telecom infrastructure is inefficient and not energy scalable to future demand. The villain here is physics. Using today’s technology, universal 5G coverage for the United States alone would require a 16x increase in base stations and a 48x increase in power usage. Totaling over $1.2T in cost, the amount of power that would need to be consumed is not feasible or sustainable. Eridan has developed an entirely new way to generate radio frequency signals at power, reducing the cost to deploy universal, high-speed wireless coverage by 10-100x. Eridan's Miracle is the world’s first switch-mode direct polar transmitter in gallium nitride. No other transmitter provides power efficiency, signal cleanliness, and spectrum agility all at the same time. Ultimately, this solution is applicable to every wireless market. In a small cell, the Miracle transmitter is twice as efficient as systems using linear power amplifiers and covers the same area at half the output power, due to the ultra-clean signals of its switched circuits. At scale, the Miracle reduces the cost of high-speed wireless coverage by up to 10x, and by up to 100x in sparse rural areas. Miracle’s frequency agility also opens the door for new spectrum allocation models, solving the artificial spectrum scarcity created by today’s rigid, frequency-fixed approach. In the near future, Miracle will also be built into a massive MIMO proof of concept system. Linear amplifiers are inefficient, and the future must be efficient. The future is fundamentally new transmitter architecture in gallium nitride, and the future is now. Co-sponsored by: IEEE Power Electronics Society, San Francisco Bay Area Chapter Speaker(s): Douglas, Mashrur Room: Rm 4021, 500 El Camnino Real, Sobrato Campus for Discovery and Innovation, Santa Clara, California, United States, 95053

MTT SCV/SF Chapter officer meeting March 22, 2024 (covering for March'24) Attendees: Venkata Gadde, Darin Phelps, Yashika Sharma, Vinay Iyer Virtual: https://events.vtools.ieee.org/m/413840

5:30pm social and food 6:15pm presentation starts Getting the money and space for implementing more anechoic chambers serving various test purposes can be a difficult path. The question “Dear Test & Measurement Solution Supplier, I have a large full-vehicle EMC chamber. Could you modify it, so it can do both EMC and OTA measurements?” is increasingly asked these days. Even though the evaluation of unintended electromagnetic radiation phenomena and intended radiation performance have a lot in common, they also differ in many respects. This talk discusses the trade-off that has to be made when choosing to combine EMC and OTA in one test environment and discusses the scope of specific combinations that work better than others. Speaker(s): Benoit Derat, 1120 Fulton Pl, Fremont, California, United States, 94539

The past decades have witnessed rapid growth in imaging as a major form of communication between individuals. Due to recent advances in capture, storage, delivery, and display technologies, consumers demand improved perceptual quality while requiring reduced storage. In this context, research and innovation in lossy image compression have steered towards methods capable of achieving high compression ratios without compromising the perceived visual quality of images. Subjective and objective visual quality assessment of images play a fundamental role in defining quality as perceived by human observers. Although the field of image compression is constantly evolving towards efficient solutions for higher visual qualities, standardized subjective visual quality assessment protocols are still limited to those proposed in ITU-R Recommendation BT.500 and JPEG AIC standards. Similarly, objective image quality metrics often exhibit limited correlation with subjective quality scores across various distortion types and intensities, and little to no work has been devoted to assessing the performance of these metrics on images with compression artifacts in the high to nearly visually lossless quality range. This talk discusses the effectiveness of both subjective and objective image quality assessment methods on coding artifacts with higher visual quality. Moreover, ongoing solutions investigated in the context of the JPEG AIC activity are presented. Speaker(s): Michela Testolina, Virtual: https://events.vtools.ieee.org/m/411480

Abstract Over the last few decades, we have witnessed various microsystems revolutionized fundamental and applied science. Due to their small size and low damping, these devices often exhibit significant nonlinearity, which narrows the operational range of these impressive applications. Therefore, understanding the mechanisms leading to nonlinearity in such systems is crucial for eliminating obstacles to their further development. Motivated by the need to advance current capabilities of Micro-Electro-Mechanical Systems (MEMS), our research has been focused on the implementation of intentional intrinsic nonlinearity in the design of micro resonators. We have demonstrated that harnessing nonlinearity enables the exploitation of various nonlinear phenomena, not attainable in linear settings, such as broadband resonances, dynamic instabilities, nonlinear hysteresis, and passive targeted energy transfers. We have developed a comprehensive analytical, numerical, and experimental methodology to consider structural nonlinearity as a main design factor, enabling to tailor mechanical resonances and achieve targeted performance. We investigated the mechanism of geometric nonlinearity in a non-prismatic microresonator and suggested strategies to tailor the various types of nonlinear resonance. Our more recent works focus on exploiting nonlinearity and multimodality simultaneously by internally coupling two or more modes through the mechanism of internal resonance. This talk will introduce various types of nonlinear phenomena realized in micro systems and discuss their unique behavioral features that can be exploited in the field of MEMS. Speaker(s): Dr. Hanna Cho, Agenda: 6:30 – 7:00 PM Registration & Networking 7:00 – 7:45 PM Invited Talk 7:45 – 8:00 PM Questions & Answers Virtual: https://events.vtools.ieee.org/m/410528

IEEE President Tom Coughlin will share his goals and objectives and Joseph Wei will present how Region 6 will support Tom in achieving his goals. This will be an interactive session where Tom and Joseph will take questions and feedback from the audience on how IEEE can better serve its members in the Bay area and around the world. [] Reference: https://techblog.comsoc.org/2024/02/05/ieee-presidents-priorities-and-strategic-direction-for-2024/ Speaker(s): Tom Coughlin, Joseph Wei, Alan Weissberger Agenda: 5:00 - 5:30pm Registration and networking 5:30 - 7:00pm Opening statements by Tom Coughlin and Joseph Wei, followed by a discussion about IEEE key issues and initiatives, moderated by Alan Weissberger 7:00 - 7:25pm Audience Q & A 7:25 - 7:30pm Closing remarks and thanks from the participants Room: Room SCDI 1302 & 1308, 500 El Camino Real, Santa Clara, California, United States, 95053