skip to main content


Title: Deep Q-Network for 5G NR Downlink Scheduling
The Third Generation Partnership Project (3GPP) introduced the fifth generation new radio (5G NR) specifications which offer much higher flexibility than legacy cellular communications standards to better handle the heterogeneous service and performance requirements of the emerging use cases. This flexibility, however, makes the resources management more complex. This paper therefore designs a data driven resource allocation method based on the deep Q-network (DQN). The objective of the proposed model is to maximize the 5G NR cell throughput while providing a fair resource allocation across all users. Numerical results using a 3GPP compliant 5G NR simulator demonstrate that the DQN scheduler better balances the cell throughput and user fairness than existing schedulers.  more » « less
Award ID(s):
2030291 2120442
NSF-PAR ID:
10356294
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
Deep Q-Network for 5G NR Downlink Scheduling
Page Range / eLocation ID:
312 to 317
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. In December 2017, the Third Generation Partnership Project (3GPP) released the first set of specifications for 5G New Radio (NR), which is currently the most widely accepted 5G cellular standard. 5G NR is expected to replace LTE and previous generations of cellular technology over the next several years, providing higher throughput, lower latency, and a host of new features. Similar to LTE, the 5G NR physical layer consists of several physical channels and signals, most of which are vital to the operation of the network. Unfortunately, like for any wireless technology, disruption through radio jamming is possible. This paper investigates the extent to which 5G NR is vulnerable to jamming and spoofing, by analyzing the physical downlink and uplink control channels and signals. We identify the weakest links in the 5G NR frame, and propose mitigation strategies that should be taken into account during implementation of 5G NR chipsets and base stations. 
    more » « less
  2. IoT systems require a wireless infrastructure that supports 5G devices, including handovers between heterogeneous and/or small cell radio access networks. These networks are subject to increased radio link failures and loss of IoT network function. 3GPP new radio (NR) applications include multihoming, i.e., simultaneously connecting devices, and handover, i.e., changing the point of access to the network. This work leverages the open radio access network (O-RAN) alliance, which specifies a new open architecture with intelligent controllers, to improve handover management. A new feedback-based time-to-trigger (TTT) handover mechanism is introduced. Improved throughput and reduced radio link failures over other techniques were achieved. 
    more » « less
  3. Fifth-generation (5G) new radio (NR) deployments are being rolled out in both the C–band (3.3 - 5.0 GHz) and millimeter-wave (mmWave) band (24.5 - 29.5 GHz). For outdoor scenarios, the C–band is expected to provide wide area coverage and throughput uniformity, whereas the mmWave band is expected to provide ultra-high throughput to dedicated areas within the C-band coverage. Due to the differences in the frequency bands, both systems are expected to be designed with different transmit and receive parameters, naturally resulting in performance variations proportional to the chosen parameters. Unlike many previous works, this paper presents measurement evaluations in central Auckland, New Zealand, from a precommercial deployment of a single-user, single-cell 5G-NR system operating in both bands. The net throughput, coverage reliability, and channel rank are analyzed across the two bands with baseband and analog beamforming. Our results show that the C-band coverage is considerably better than mmWave, with a consistently higher channel rank. Furthermore, the spatial stationarity region (SSR) for the azimuth angles-of-departure (AODs) is characterized, and a model derived from the measured beam identities is presented. The SSR of azimuth AODs is seen to closely follow a gamma distribution. 
    more » « less
  4. With the growing demand for locating services in a variety of commercial applications, positioning techniques have been considered as a vital part in cellular networks. With the evolution from 2G to 5G, the positioning techniques have been enhanced in various aspects. In this paper, we summarize the evolution of positioning standards in the Third Generation Partnership Project (3GPP) and briefly introduce the new positioning standards in 5G NR (New Radio), which include new positioning requirements, the general positioning structure, new positioning reference signals, and general positioning methods. 
    more » « less
  5. Henderson, Thomas ; Imputato, Pasquale ; Liu, Yuchen ; Gamess, Eric (Ed.)
    Physical (PHY) layer abstraction is an effective method to reduce the runtimes compared with link simulations but still accurately characterize the link performance. As a result, PHY layer abstraction for IEEE 802.11 WLAN and 3GPP LTE/5G has been widely configured in the network simulators such as ns-3, which achieve faster system-level simulations quantifying the network performance. Since the first publicly accessible 5G NR Sidelink (SL) link simulator has been recently developed, it provides a possibility of implementing the first PHY layer abstraction on 5G NR SL. This work deploys an efficient PHY layer abstraction method (i.e., EESM-log-SGN) for 5G NR SL based on the offline NR SL link simulation. The obtained layer abstraction which is further stored in ns-3 for use aims at the common 5G NR SL scenario of OFDM unicast single layer mapping in the context of Independent and Identically Distributed (i.i.d.) frequency-selective channels. We provide details about implementation, performance, and validation. 
    more » « less