@article{kilinc_dreifuerst_kim_heath_2024, title={Beam Training in mmWave Vehicular Systems: Machine Learning for Decoupling Beam Selection}, ISBN={["979-8-3503-5186-6"]}, ISSN={["2375-8236"]}, DOI={10.1109/BLACKSEACOM61746.2024.10646235}, journal={2024 IEEE INTERNATIONAL BLACK SEA CONFERENCE ON COMMUNICATIONS AND NETWORKING, BLACKSEACOM 2024}, author={Kilinc, Ibrahim and Dreifuerst, Ryan M. and Kim, Junghoon and Heath, Robert W., Jr.}, year={2024}, pages={54–59} }
 @article{nahum_lopes_dreifuerst_batista_correa_cardoso_klautau_heath jr_2024, title={Intent-Aware Radio Resource Scheduling in a RAN Slicing Scenario Using Reinforcement Learning}, volume={23}, ISSN={["1558-2248"]}, DOI={10.1109/TWC.2023.3297014}, abstractNote={Network slicing at the radio access network (RAN) domain, called RAN slicing, requires elasticity, efficient resource sharing, and customization. In this scenario, radio resource scheduling (RRS) is responsible for dealing with scarce and limited frequency spectrum resources available at the RAN domain while fulfilling the slice intents. The wide variety of scenarios supported in 5G and beyond 5G networks makes the RRS problem in RAN slicing scenario a significant challenge. This paper proposes an intent-aware reinforcement learning method to perform the RRS function in a RAN slicing scenario. The slice's quality of service intents is described in a common intent model in a service-level agreement. The proposed method tries to prevent intent faults by making the management of radio resources available among slices. This method uses slices' and user equipment network metrics in the observation space. The proposed method is evaluated under different network conditions and outperforms different baselines considering the slices' intents fulfillment.}, number={3}, journal={IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS}, author={Nahum, Cleverson Veloso and Lopes, Victor Hugo L. and Dreifuerst, Ryan M. and Batista, Pedro and Correa, Ilan and Cardoso, Kleber Vieira and Klautau, Aldebaro and Heath Jr, Robert W.}, year={2024}, month={Mar}, pages={2253–2267} }
 @article{dreifuerst_heath jr_2024, title={Machine Learning Codebook Design for Initial Access and CSI Type-II Feedback in Sub-6-GHz 5G NR}, volume={23}, ISSN={["1558-2248"]}, DOI={10.1109/TWC.2023.3331313}, abstractNote={Beam codebooks are a recent feature to enable high dimension multiple-input multiple-output in 5G. Codebooks comprised of customizable beamforming weights can be used to transmit reference signals and aid the channel state information (CSI) acquisition process. Codebooks are also used for quantizing feedback following CSI measurement. In this paper, we unify the beam management stages–codebook design, beam sweeping, feedback, and data transmission–to characterize the impact of codebooks throughout the process. We then design a neural network to find codebooks that improve the overall system performance. The proposed neural network is built on translating codebook and feedback knowledge into a consistent beamspace basis similar to a virtual channel model to generate initial access codebooks. This beamspace codebook algorithm is designed to directly integrate with current 5G beam management standards without changing the feedback format or requiring additional side information. Our simulations show that the neural network codebooks improve over traditional codebooks, even in dispersive sub-6GHz environments. We further use our framework to evaluate CSI feedback formats with regard to multi-user spectral efficiency. Our results suggest that optimizing codebook performance can provide valuable performance improvements, but optimizing the feedback configuration is also important in sub-6GHz bands.}, number={6}, journal={IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS}, author={Dreifuerst, Ryan M. and Heath Jr, Robert W.}, year={2024}, month={Jun}, pages={6411–6424} }
 @misc{dreifuerst_heath_2023, title={Massive MIMO in 5G: How Beamforming, Codebooks, and Feedback Enable Larger Arrays}, volume={61}, ISSN={["1558-1896"]}, DOI={10.1109/MCOM.001.2300064}, abstractNote={Massive multiple-input multiple-output (MIMO) is an important technology in fifth generation (5G) cellular networks and beyond. To help design the beamforming at the base station, 5G has introduced new support in the form of flexible feedback and configurable antenna array geometries that allow for arbitrarily massive physical arrays. In this article, we present an overview of MIMO throughout the mobile standards, highlight the new beam-based feedback system in 5G NR, and describe how this feedback system enables massive MIMO through beam management. Finally, we conclude with challenges related to massive MIMO in 5G.}, number={12}, journal={IEEE COMMUNICATIONS MAGAZINE}, author={Dreifuerst, Ryan M. and Heath, Robert W., Jr.}, year={2023}, month={Dec}, pages={18–23} }
 @article{lopes_nahum_dreifuerst_batista_klautau_cardoso_heath jr_2022, title={Deep Reinforcement Learning-Based Scheduling for Multiband Massive MIMO}, volume={10}, ISSN={["2169-3536"]}, DOI={10.1109/ACCESS.2022.3224808}, abstractNote={Fifth-generation (5G) cellular communication systems have embraced massive multiple-input-multiple-output (MIMO) in the low- and mid-band frequencies. In a multiband system, the base station can serve different users in each band, while the user equipment can operate only in a single band simultaneously. This paper considers a massive MIMO system where channels are dynamically allocated in different frequency bands. We treat multiband massive MIMO as a scheduling and resource allocation problem and propose deep reinforcement learning (DRL) agents to perform user scheduling. The DRL agents use buffer and channel information to compose their observation space, and the agent’s reward function maximizes the transmitted throughput and minimizes the packet loss rate. We compare the proposed DRL algorithms with traditional baselines, such as maximum throughput and proportional fairness. The results show that the DRL models outperformed baselines obtaining a 20% higher network sum rate and an 84% smaller packet loss rate. Moreover, we compare different DRL algorithms focusing on training time to assess the online implementation of the DRL agents, showing that the best agent needs about 50K training steps to converge.}, journal={IEEE ACCESS}, author={Lopes, Victor Hugo L. and Nahum, Cleverson Veloso and Dreifuerst, Ryan M. and Batista, Pedro and Klautau, Aldebaro and Cardoso, Kleber Vieira and Heath Jr, Robert W.}, year={2022}, pages={125509–125525} }
 @article{dreifuerst_heath_yazdan_2022, title={Massive MIMO Beam Management in Sub-6 GHz 5G NR}, DOI={10.1109/VTC2022-Spring54318.2022.9860458}, abstractNote={Beam codebooks are a new feature of massive multiple-input multiple-output (M-MIMO) in 5G new radio (NR). Codebooks comprised of beamforming vectors are used to transmit reference signals and obtain limited channel state information (CSI) from receivers via the codeword index. This enables large arrays that cannot otherwise obtain sufficient CSI. The performance, however, is limited by the codebook design. In this paper, we show that machine learning can be used to train site-specific codebooks for initial access. We design a neural network based on an autoencoder architecture that uses a beamspace observation in combination with RF environment characteristics to improve the synchronization signal (SS) burst codebook. We test our algorithm using a flexible dataset of channels generated from QuaDRiGa. The results show that our model outperforms the industry standard (DFT beams) and approaches the optimal performance (perfect CSI and singular value decomposition (SVD)-based beamforming), using only a few bits of feedback.}, journal={2022 IEEE 95TH VEHICULAR TECHNOLOGY CONFERENCE (VTC2022-SPRING)}, author={Dreifuerst, Ryan M. and Heath, Robert W., Jr. and Yazdan, Ali}, year={2022} }
 @article{dreifuerst_heath_2022, title={SignalNet: A Low Resolution Sinusoid Decomposition and Estimation Network}, volume={70}, ISSN={["1941-0476"]}, DOI={10.1109/TSP.2022.3201336}, abstractNote={The detection and estimation of sinusoids is a fundamental signal processing task for many applications related to sensing and communications. While algorithms have been proposed for this setting, quantization is a critical, but often ignored modeling effect. In wireless communications, estimation with low resolution data converters is relevant for reduced power consumption in wideband receivers. Similarly, low resolution sampling in imaging and spectrum sensing allows for efficient data collection. In this work, we propose SignalNet, a neural network architecture that detects the number of sinusoids and estimates their parameters from quantized in-phase and quadrature samples. We incorporate signal reconstruction internally as domain knowledge within the network to enhance learning and surpass traditional algorithms in mean squared error and Chamfer error. We introduce a worst-case learning threshold for comparing the results of our network relative to the underlying data distributions. This threshold provides insight into why neural networks tend to outperform traditional methods and into the learned relationships between the input and output distributions. In simulation, we find that our algorithm is always able to surpass the threshold for three-bit data but often cannot exceed the threshold for one-bit data. We use the learning threshold to explain, in the one-bit case, how our estimators learn to minimize the distributional loss, rather than learn features from the data.}, journal={IEEE TRANSACTIONS ON SIGNAL PROCESSING}, author={Dreifuerst, Ryan M. and Heath, Robert W., Jr.}, year={2022}, pages={4454–4467} }