DeepSig is proud to sponsor the upcoming IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN)—the very first focused on the intersection of AI/ML and communication. The event, held at the KTH Royal Institute of Technology in Stockholm, Sweden, focuses on the intersection of AI/ML and communication technologies, a rapidly advancing area of research that promises to redefine the future of communication systems and networks.
Insightful Sessions and Engagements
DeepSig will actively engage in multiple sessions, sharing their insights and innovations at the conference. Some of the highlights of their participation include:
Learning to Transmit and Receive Session
This interactive session, led by scholars from the University of Maryland and DeepSig, will present a novel deep learning-based beamforming design for a single-input multiple-output (SIMO) radio system. The approach is based on a multi-objective optimization model that uses neural networks to predict multi-task performance measures of the radio system. The method is generalizable for various wireless communication scenarios, with performance improvements over traditional beamforming designs.
Panel on AI-Native 6G Radio Interfaces – What Can Hardware Acceleration Really Bring Out?
In recent years, advanced technologies have been integrated into the radio interface, bringing performance advantages and increasing the need for complex management solutions. This complexity can be addressed through increased network automation with the adoption of AI/ML. While 5G-Advanced has introduced AI/ML-based improvements, 6G wireless networks are expected to fully integrate AI/ML into the air interface design, facing challenges such as computational intensity and infrastructure complexities. This panel session brings together senior researchers to discuss the challenges and future developments in AI-native 6G systems, focusing on algorithmic and hardware acceleration. The implications of this development are significant economic and political in terms of technological sovereignty in the US and the EU.
The panel, moderated by Dr. Carles Antón-Haro, director EU programmes & industry contracts and research director for Centre Tecnologic de Telecomunicacions de Catalunya, will explore the intersection of AI/ML algorithmic and hardware acceleration developments in 6G, featuring insights from leading researchers in the field.
DeepSig CTO Tim O’ShO’Sheal will be joined by Dr. Bo Goransson, senior expert multi-antenna systems at Ericsson, Deniz Gündüz, professor at Imperial College London and Zoran Zivkovic, principal engineer and system architect at Intel.
Demo on OmniPHY Real-Time Pre-6G AI-Native Air-Interface Over the Air
One of today’s most exciting design considerations for the 6G Air Interfaces is moving to an AI-Native approach for the physical layer. Such an approach offers many significant advantages, such as removing pilots, adapting to local channel conditions or hardware impairments, and even cross-layer optimization for semantic communications. To this end, DeepSig will demonstrate OmniPHY, its AI-Native Air Interface, which can be used today for real-time high-rate over-air tests and pre-6G physical layer techniques validation.
Tim O’Shea will be joined by Director of ML Tamoghna Roy, Principal Engineer Dan DePoy, Principal Engineer Raj Bhattacharjea, Senior ML Engineer Nitin Nair and Principal Applications Engineer Jacob Gilbert will illustrate the real-time high-rate over-air tests and validation of pre-6G physical layer techniques.
DeepSig’s participation in ICMLCN 2024 underscores its commitment to shaping the future of AI-driven communication networks, fostering a dialogue among academia and industry leaders on pioneering technological advancements.
