Fast Initial Access with Deep Learning for Beam Prediction in 5G mmWave Networks
Fast Initial Access with Deep Learning for Beam Prediction in 5G mmWave Networks
复制标题
通过深度学习快速初始访问 5G 毫米波网络中的波束预测
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Y. Sagduyu
中科院分区：
文献类型：
--
作者：
Tarun S. Cousik;V. Shah;Jeffrey H. Reed;T. Erpek;Y. Sagduyu
DeepIA是一种深度学习解决方案，可在定向网络（如5G系统中的毫米波网络）中实现快速、可靠和安全的初始接入（IA）。通过在IA过程中仅利用波束的子集，DeepIA消除了对穷举波束搜索的需要，从而减少了IA中的波束扫描时间。训练深度神经网络（DNN）以学习从利用减少数量的波束收集的接收信号强度（RSS）到接收器的最佳空间波束（在较大的波束集合中）的复杂映射。在测试时，DeepIA仅从少量波束中测量RSS，并运行DNN来预测IA的最佳波束。我们表明，DeepIA通过扫描更少的波束减少了IA时间，并且在视距（LoS）和非视距（NLoS）毫米波信道条件下的波束预测精度都明显优于传统IA。
We present DeepIA, a deep learning solution for a fast, reliable and secure initial access (IA) in directional networks such as the mmWave networks in 5G systems. By utilizing only a subset of beams during the IA process, DeepIA removes the need for an exhaustive beam search thereby reducing the beam sweep time in IA. A deep neural network (DNN) is trained to learn the complex mapping from the received signal strengths (RSSs) collected with a reduced number of beams to the optimal spatial beam of the receiver (among a larger set of beams). In test time, DeepIA measures the RSSs only from a small number of beams and runs the DNN to predict the best beam for IA. We show that DeepIA reduces the IA time by sweeping fewer beams and significantly outperforms the conventional IA's beam prediction accuracy in both line of sight (LoS) and non-line of sight (NLoS) mmWave channel conditions.