MIX2 – RAS & 6VS
Wednesday, 7 June 2023, 16:00-17:30, Room H1
Session Chair: Özlem Tuğfe Demir (TOBB University of Economics and Technology, Turkey)
A Comparative Performance Analysis of LoRaWAN in Two Frequency Spectra: EU868 MHz and 2.4 GHz
Riccardo Marini and Giampaolo Cuozzo (CNIT, National Laboratory of Wireless Communications (WiLab), Italy) During the latest years, LoRaWAN has emerged as a valid alternative to cellular technologies for Internet of Things applications, thanks to its long-range capabilities and high energy efficiency at the expense of a very limited bit rate. To support highly demanding applications, in 2017, Semtech released the first LoRa module working at 2.4 GHz, but the LoRaWAN protocol has not been updated to support such a version, despite the wider frequency channels available in this frequency portion. In this paper, we perform a quantitative analysis on the exploitation of this frequency spectrum in the currently released LoRaWAN version, comparing it with the standard EU868 MHz version. By means of extensive ad-hoc network simulations obtained with the open-source LoRaWANSim, we point out the main advantages of the 2.4 GHz version (e.g., higher throughput) and disadvantages (e.g., coverage issues), for both transmission directions, and as a function of different parameters, such as the number of devices, gateways, area size, and payload. Numerical results show that LoRaWAN at 2.4 GHz is a viable solution for scenarios characterized by a high density of nodes in a limited area, whereas LoRaWAN in the EU868 MHz spectrum remains the preferred choice when considering wide environments.
Towards Enabling Performance-Guaranteed Slice Management and Orchestration in 6G
Serae Kim, Sunghyun Jin, Junseon Kim and Kyunghan Lee (Seoul National University, Korea (South))
Next-generation network services (e.g., XR, mobile hologram, digital twin) often expect both latency and bandwidth guarantees. In the 5G network, network slicing techniques that enable the isolated management of multiple virtual networks are devised for ensuring quality-of-service (QoS). However, existing network slicing frameworks are inherently insufficient to provide guaranteed performance to those new network services. Even with ultra-reliable low-latency communications (URLLC), lowlevel performance pertaining to the delivery of radio frames or packets rather than service-level performance has only been dealt with, although there exists a large gap between them. The discrepancy comes from the fact that none of those services runs based on the packets. They run based on their own application data units (ADUs) whose size is dynamic and mostly much larger than just a packet. In this regard, in order to directly guarantee the service-level performance, we propose a new slice management and orchestration framework that can make the time duration to complete the transmission (i.e., completion time) of variable ADUs over fluctuating wireless channels constant through two techniques leveraging the knowledge of ADUs: a time budget orchestration and a radio resource management for ADU completion. We provide detailed specifications of our framework.
Energy Savings Under Performance Constraints via Carrier Shutdown with Bayesian Learning
Lorenzo Maggi (Nokia Bell Labs, France); Claudiu Mihailescu (Nokia, France); Qike Cao (Nokia, Finland); Alan Tetich (Sii Poland, Poland); Saad Khan (Nokia, Poland); Simo Aaltonen (Nokia, Finland); Arndt Ryo Koblitz (Nokia Bell Labs, United Kingdom (Great Britain)); Maunu Holma (Nokia, Finland); Samuele Macchi and MariaElena Ruggieri (Nokia, Italy); Igor Korenev and Bjarne Klausen (Nokia, Denmark)
By shutting down frequency carriers, the power consumed by a base station can be considerably reduced. However, this typically comes with traffic performance degradation, as the congestion on the remaining active carriers is increased. We leverage a hysteresis carrier shutdown policy that attempts to keep the average traffic load on each sector within a certain min/max threshold pair. We propose a closed-loop Bayesian method optimizing such thresholds on a sector basis and aiming at minimizing the power consumed by the power amplifiers while maintaining the probability that KPI’s are acceptable above a certain value. We tested our approach in a live customer 4G network. The power consumption at the base station was reduced by 11% and the selected KPI’s met the predefined targets.
Dueling-DQN Based Spectrum Sharing Between MIMO Radar and Cellular Networks
Atiquzzaman Mondal and Aparajita Dutta (Indian Institute of Information Technology Guwahati, India); Sudip Biswas (Indian Institute of Information Technology, Guwahati, India)
We investigate a two-tier distributed spectrum sharing framework between a multi-cell multi-user mobile broadband network (MBN) and a multiple-input multiple-output (MIMO) radar. While multi-agent reinforcement learning (RL) is used for transmit power allocation at MBN’s base-stations to improve the quality of service of its users subject to the constraint of the probability of detection of the radar, the interference from the radar towards the MBN is mitigated via null-space based waveform projection. In the RL framework, the multiple cells in the MBN operate as agents, and the average signalto-noise ratio value is the reward. Accordingly, we propose a deep RL network called dueling deep Q-network (DDQN) to enable co-existence by taking into account the physical layer parameters of the MBN and radar communication. The DDQN is compared to two other baseline RL algorithms, namely Qlearning and deep Q-network (DQN). Numerical results show that DDQN learns to obtain the best power allocation policies for distributed spectrum access without needing a centralized controller to control interference towards the radar. In particular, over time, the advantage network of DDQN allows the agent to take actions having higher advantage value, thus leading to faster convergence and a more stable spectrum sharing framework.
Inter O-DUs Coordination for Scheduling of Massive Users in O-RAN
Aamir Latif and Muhammad Mahtab Alam (Tallinn University of Technology, Estonia); Yannick Le Moullec (Tallinn University of Technology (TalTech), Estonia)
Open Radio Access Network (O-RAN) is a decentralized, intelligent, and open network architecture. RAN Network Functions (RNF) are split into Open Central Unit (OCU), Open Distributed Unit (O-DU), and Open Radio Unit (ORU), which are deployed on commercial hardware and cloud nodes as containerized Network Functions (CNFs). This paper presents a coordination architecture between O-RAN Distributed Units (O-DU) for scheduling massive number of users. We propose a Non-Orthogonal Multiple Access (NOMA) based access technique for interference avoidance and scheduling of such massive number of users through a cooperative strategy. This involves sharing individual scheduling tables to compute interference for subsequent User Equipment (UEs) transmissions in the network. The results of the interference calculations are then fed into ODU schedulers. The scheduler of each O-DU shares the radio resources among the UEs to have minimum interference. The proposed scheme’s effectiveness in terms of throughput, energy efficiency, and connectivity is examined through simulations. The results show that the proposed method reduces the effects of interference, making it better suited to the needs of dense deployments. Users operating under the proposed approach experience up to +30% and -10% improvements in achieved user data rates and energy consumption, respectively, when compared to the absence of a coordination architecture.
