Hosted by the College of Engineering and the Elmore Family School of Electrical and Computer Engineering

Abstract

Professor Nambirajan Seshadri, a renowned expert in wireless communication systems and theory, signal processing, systems on a chip, and networking, will present a stimulating discussion about his work applying machine learning to the design of end-to-end communications systems. Machine learning has the potential to quickly crunch, analyze, and learn from big data in order to continuously optimize these complex communications networks by reducing noise and interference, power consumption, and computational requirements. These will be vital improvements, as we connect billions of heterogeneous devices for applications across sectors like healthcare, autonomous driving, smart factories and buildings, and so forth. Professor Seshadris talk will cover the application of machine learning to a variety of communication channels, including the additive white Gaussian noise (AWGN) channel, the multi-user interference channel, and the MIMO fading channel. In the second part of his presentation, he will consider the problem of synthesis and generation of facial images, with the goal of achieving low-bit-rate image/video compression for transmission over noisy channels. This can reduce the amount of data required to show an image or video while minimizing loss of quality " also a crucial optimization challenge, as communications networks wrestle with the enormous and endlessly growing data streams that will flow through them in the future.

Biography

Professor Seshadri has a track record of contributing to successful research programs at AT&T Bell Labs and AT&T Labs, as well as creating and building a multi-billion-dollar wireless business at Broadcom Corporation. He received his B.E. degree in Electronics and Communications from Regional Engineering College (now called NIT), Tiruchirappalli, India in 1982, and M.S. and Ph.D. from Rensselaer Polytechnic Institute, Troy, NY in 1984 and 1986, respectively. He joined Broadcom in 1999 and was the first employee dedicated to developing the companys wireless strategy. As VP and CTO of the Mobile Platforms and Wireless Connectivity business groups, he helped drive Broadcoms entry into 2G, 3G and 4G cellular, mobile multimedia, low power Wi-Fi for handsets, combo chips that integrate multiple wireless connectivity technologies, GPS, as well as development of a strong intellectual property rights (IPR) portfolio. From 2011 to 2014 he also served as Senior VP and General Manager of the Mobile Platforms Solutions business unit. This group was responsible for developing and marketing cellular baseband, RF and power management products, as well as complete smart phone handset reference platforms. From 2014 until his departure from Broadcom he served as Senior VP and CTO of the Broadband and Connectivity group. Prior to joining Broadcom, he worked for more than 13 years at AT&T Bell Laboratories as a member of the technical staff and as head of communications research at AT&T Shannon Labs, where he focused on developing techniques for reliable transmission of data, speech, and audio for mobile communications. He is a Fellow of the IEEE, member of the US National Academy of Engineering, Foreign Member of the Indian National Academy of Engineering, and Distinguished Alumnus of the National Institute of Technology in Tiruchirappalli, India. He holds some 200 patents, and is a co-recipient of the 1999 IEEE Information Theory Society Award for best paper, Space-time Codes for High Data Rate Wireless Communications: Performance Criterion and Code Construction. His published body of work has almost 30,000 Google Scholar citations.