US Patent - System & Method for Throughput Prediction for Cellular Networks

2 Jul 2020

US Patent - System & Method for Throughput Prediction for Cellular Networks 

Professor Cormac Sreenan, Dr Ahmed Zahran and Dr. Darijo Raca (recent CSIT PhD graduate) are inventors on a new US patent (No. 10,693,575) entitled "System and method for throughput prediction for cellular networks". This is a joint invention with AT&T Labs - Research (USA) and arose from the iVID research project that was funded by Science Foundation Ireland. The invention uses machine and deep learning techniques to predict the data throughput of wireless links in mobile phone networks. Having accurate predictions can be used by applications to improve their performance, one example being video streaming apps, which can use this informationto decide on what video quality can be sustained without causing playback stalls.