Sahoo recognized as distinguished lecturer by IEEE
Bibhudatta Sahoo, professor, Department of Electrical Engineering discussing research with one of his students.
By Peter Murphy
Published May 5, 2025
Bibhudatta Sahoo, professor in the Department of Electrical Engineering, will serve as a distinguished lecturer in the Institute of Electrical and Electronics Engineers (IEEE) Circuits and Systems Society’s (CASS) Distinguished Lecturer Program (DLP) for two years.
As a lecturer, Sahoo will discuss future impacts on the semiconductor industry, analog computing techniques and their impact on AI and machine learning algorithms, and computing blocks needed to realize various neural network architectures.
According to IEEE CASS, the program serves the needs of CASS members and enhances “their professional knowledge and vitality by keeping them informed of the latest research results and their practical applications.”
Twenty researchers were selected as distinguished lecturers from around the world, this year.
Sahoo’s research focuses on analog and mixed signal circuit design; hardware accelerators for machine learning; signal processing inspired analog circuits; and other areas critical to circuits and systems research. As a CASS distinguished lecturer, Sahoo is expected to give a minimum of three talks per year to any CASS society worldwide on any of his three selected topics to promote research in the area of circuits and systems.
Data Converter Design—A Signal Processing Perspective
Sahoo’s first topic addresses Moore’s Law and the critical points of the integrated circuit industry. Moore’s Law observes that the number of transistors on a computer chip doubles every two years, increasing the performance and decreasing the cost. The transistors decrease in size, allowing engineers to shrink the size of the chips. However, analog and mixed-signal circuits—found in smartphones, medical devices and cars—have not kept pace with advancements in transistors, and this has created a bottleneck in many of these critical systems.
Sahoo suggests that we have hit an inflection point in designing analog-to-digital systems, and that designers need to be as innovative in both the analog and digital signal processing domains.
Analog and Mixed Signal Circuits and Systems for Emerging Applications
Sahoo will also share why some classic electronics are finding a resurgence in advanced technologies like quantum computing and artificial intelligence and machine learning. While quantum computing has significantly enhanced raw computing power, parts of it can be built with traditional electronics like operational amplifiers.
In his second talk, Sahoo will discuss analog computing techniques that could enable AI and machine learning algorithms, and dive into mixed-signal computing. He will also provide details of novel memory devices, viz and memristors to enable energy efficient computing. This topic will also cover how classical analog hardware can emulate quantum algorithms.
Mixed Signal Approaches to Machine Learning Hardware Accelerator for Inference Engines
Sahoo’s third lecture also discusses the concept of using conventional analog technologies to continue to develop smart connected devices—like smartwatches, sensors and appliances—that produce huge amounts of data. According to Sahoo, we have reached a turning point in designing smart devices, and some engineers are using analog computing technologies to continue to develop these devices. Sahoo will cover neural network architectures and the computing needed to realize them.
Sahoo joined UB in 2023. He leads the Smart-Signal Processing and Integrated Circuit Engineering Laboratory at Buffalo and is a faculty member in UB’s Center for Advanced Semiconductor Technologies.