The Role and Future of Machine Learning in Massive MIMO Wireless Systems

, ETH Zurich
Future wireless systems will combine large antenna arrays with millimeter-wave communication in order to increase network throughput and improve quality-of-service. Such massive MIMO (multiple-input and multiple-output) wireless systems must acquire and process massive amounts of data at very high rates, which requires innovation at the physical layer. We'll showcase two examples to demonstrate that this data deluge is not just a curse, but also a blessing. Specifically, we'll show that the large amount of measured channel state information enables machine learning (ML) for user equipment localization without needing global navigation satellite systems. We'll also show that ML tools can reduce the complexity of some of the most complex physical layer tasks, including data detection and precoding. Finally, we'll discuss the role and efficacy of graphics processing units to assist such ML tasks in future massive MIMO wireless systems.
活动: GTC Digital November
日期: November 2021
级别: 中级技术
话题: IoT / 5G / Edge
行业: 电信
语言: 英语
话题: Signal & Sensor Processing
所在地: