Radio Access Network Research Articles

Recent Advances in Machine Learning for Network Automation in the O-RAN.

The evolution of network technologies has witnessed a paradigm shift toward open and intelligent networks, with the Open Radio Access Network (O-RAN) architecture emerging as a promising solution. O-RAN introduces disaggregation and virtualization, enabling network operators to deploy multi-vendor and interoperable solutions. However, managing and automating the complex O-RAN ecosystem presents numerous challenges. To address this, machine learning (ML) techniques have gained considerable attention in recent years, offering promising avenues for network automation in O-RAN. This paper presents a comprehensive survey of the current research efforts on network automation usingML in O-RAN.We begin by providing an overview of the O-RAN architecture and its key components, highlighting the need for automation. Subsequently, we delve into O-RAN support forML techniques. The survey then explores challenges in network automation usingML within the O-RAN environment, followed by the existing research studies discussing application of ML algorithms and frameworks for network automation in O-RAN. The survey further discusses the research opportunities by identifying important aspects whereML techniques can benefit.

High cost-effective photonic assisted in-band interference cancellation scheme for centralized radio access network towards future 5G communication

A photonic-assisted scheme capable of self-interference cancellation (SIC) and image rejection mixing (IRM) is demonstrated to solve the intractable problem of in-band interference for in-band full duplex (IBFD) systems with high frequency and large bandwidth. In the proposed scheme, a single optical path configuration is constructed based on an integrated modulator applied as the main device. The SIC operation is implemented in the optical domain making it immune to the influence caused by fiber dispersion. Owing to the intermediate frequency (IF) signal amplitude regulation mechanism under the combined action of fiber dispersion induced phase and double sideband (DSB) modulation, the scheme can realize IRM without increasing the system complexity. Experimental results show that the 50 MHz 16-quadrature amplitude modulation (16-QAM) signal of interest (SOI) centered at different frequencies are down-converted to 2.5 GHz. The depth of SIC and image-rejection (IR) is over 35 dB and 25 dB after transmitted through a span of 10.1 km optical fiber. The received 16-QAM down-converted signal can be recovered with an error vector magnitude (EVM) which satisfies the requirement of 3GPP-specified EVM limit. The signal recovery performance, SIC depth and image rejection ratio (IRR) under different signal to interference ratio (SIR) is also investigated. In addition, a comparison of relevant literature with this scheme from the perspective of technical discussion is also presented. The proposed scheme can realize effective functional integration with compact structure, simple parameter tuning mechanism, improved stability and high cost-effectiveness to solve the in-band interference problem which is of great value to be applied in IBFD centralized radio access networks (CRAN) towards future 5G communication.

A three-level slicing algorithm in a multi-slice multi-numerology context

Fifth generation (5G) mobile networks feature a new concept called Bandwidth Part (BWP) that permits applying various multiple Orthogonal Frequency-Division Multiplexing (OFDM) subcarrier spacing, also coined numerologies, in the same band. With flexible numerology, the sought-after purpose is achieving lower latency for the Ultra Reliable Low Latency Communications (URLLC) service. However, the lucrative multiple numerology gains might not stand because of the ensuing Inter-Numerology Interference (INI) problems. In this paper, we assess the feasibility, as well as the profitability, of multi-numerology in a multi-slice setting. In such a context, it is paramount to study the radio resource allocation problem at the Radio Access Network (RAN) level. To that aim, we propose a three-level slicing algorithm that efficiently selects the BWP serving the URLLC users from a set of BWPs using different numerologies, designs a guard band among these BWPs to reduce INI without wasting radio resources, and attributes a bandwidth to each slice depending on multiple parameters such as the users’ Key Performance Indicators (KPIs), channel conditions, INI, and the incurred monetary cost. This work considers the three 5G services, namely the enhanced Mobile BroadBand (eMBB) service, the URLLC service and the massive Machine-Type Communications (mMTC) service. The proposed algorithm combines various tools from Game Theory, Deep-Q learning Networks as well as savvy heuristics. The performance evaluation demonstrates the high efficiency of our solution in terms of latency and throughput compared to well-known approaches in the State-of-the-Art.

User Preference-Based Heterogeneous Network Management System for Vertical Handover

Vertical handover management plays an essential role in wireless network technologies, mainly due to the rapid development of various radio access technologies (RATs) that require users to connect seamlessly from one RAT to another. However, in multiple RAT environments, vertical handover management encounters different challenges, including unnecessary handovers, handover failures, ping-pong handovers, and unsuitable access network selection. Essential in vertical handover management is maintaining the desired quality of service (QoS) by the mobile device user. The seamless movement of mobile device users as they run various applications depends on a well-performing vertical handover decision-making algorithm. This bears special significance in a heterogeneous network environment. This paper proposes a vertical handover algorithm that considers user preferences (i.e., a vertical handover algorithm that evaluates the application currently running on a user device). The main objective of the algorithm is to determine when it is necessary to perform the handover, depending on the applications running on the mobile device. The proposed algorithm utilizes a fuzzy logic system to assess whether the handover is necessary and a multiattribute decision-making (MADM) method to select the best available radio access network. A simulation scenario involving different applications at various mobile device velocities was developed. The results proved the algorithm’s effectiveness compared to some of the earlier proposed vertical handover algorithms. At velocities below 10 m/s and 30 m/s, the proposed algorithm had 0% and 15.02% unnecessary handovers, respectively, while the technique for order preference by similarity to ideal solution (TOPSIS) utility’s function-based algorithm obtained 12.38% and 23.24% at the same velocities, respectively. In addition, compared to TOPSIS, the obtained results of the proposed algorithm demonstrated a lower handover failure rate and ping-pong rate for a velocity span of 1–30 m/s for the considered user applications.

Minimizing energy consumption of IoT devices for O-RAN based IoT systems

The implementation of Open Radio Access Network (O-RAN) architecture in Internet of Things (IoT) systems has garnered significant attention as a means to fulfill the stringent requirements of ultra-low latency and ultra-low energy consumption in future IoT systems. Although the traditional edge network architecture has been extensively employed, it continues to pose challenges in terms of synchronizing the integration of global and local information for the design of an optimal offloading strategy in edge servers, thus hindering the reduction of latency and energy consumption in IoT devices. In an effort to decrease the latency and energy consumption of IoT devices (IoTDs), we propose a computation offloading problem and employs the Successive Convex Approximation (SCA) algorithm to convert the non-convex problem into a convex problem. The proposed strategy aims to minimize the energy consumption of IoTDs while ensuring Quality of Service (QoS) requirements. The results of the experiment demonstrate that the average energy consumption of terminal devices can be reduced by 20%. Additionally, this strategy is found to be more effective in reducing IoTDs' latency and energy consumption as compared to the traditional edge network architecture.

On the design and performance of scheduling policies exploiting spatial diversity for URLLC

In this paper, we study the performance of packet scheduling schemes for Ultra-Reliable Low-Latency Communications (URLLC) services. We exploit spatial diversity (i.e., redundant coverage of users) guaranteed in numerous 5G radio access network scenarios to examine the impact of multi-connectivity. Thus, we consider a set of URLLC users connected to two frequency layers or Radio Access Technologies (RATs) to ensure minimal queuing time. We review four packet scheduling and redundancy schemes, namely Join-the-Shortest-Queue (JSQ), the shortest expected delay (SED), systematic Redundancy (RED), and redundancy with Cancellation upon completion (CAN). We choose the outage probability as a metric, defined as the packet’s probability of arriving after some given target delay. We show that RED performs well at low load, whereas JSQ and SED are better when the load rises. Besides, CAN outperforms all other schemes. We then discuss the trade-off between performance and implementation complexity.

Oceania’s 5G Multi-Tier Fixed Wireless Access Link’s Long-Term Resilience and Feasibility Analysis

Information and communications technologies play a vital role in achieving the Sustainable Development Goals (SDGs) and bridging the gap between developed and developing countries. However, various socioeconomic factors adversely impact the deployment of digital infrastructure, such as 5G networks, in the countries of Oceania. The high-speed broadband fifth-generation cellular network (5G) will improve the quality of service for growing mobile users and the massive Internet of Things (IoT). It will also provide ultra-low-latency services required by smart city applications. This study investigates the planning process for a 5G radio access network incorporating sub-6 GHz macro-remote radio units (MRRUs) and mmWave micro-remote radio units (mRRUs). We carefully define an optimization problem for 5G network planning, considering the characteristics of urban macro-cells (UMa) and urban micro-cells (UMi) with appropriate channel models and link budgets. We determine the minimum number of MRRUs and mRRUs that can be installed in each area while meeting coverage and user traffic requirements. This will ensure adequate broadband low-latency network coverage with micro-cells instead of macro-cells. This study evaluates the technical feasibility analysis of combining terrestrial and airborne networks to provide 5G coverage in Oceania, with a special emphasis on Fiji.

Enabling containerized Central Unit live migration in 5G radio access network: An experimental study

5G mobile networks have gained considerable attention and traction by offering numerous benefits such as high data rates, low latency, reliability, and application specific connectivity. To effectively achieve these milestones mobile network functions must be virtualized using Virtual Machine (VM) and/or Containers. The latter has gained momentum in recent years due to its reduced resource usage and virtualization overhead, improving efficiency and flexibility while sharing compute and network resources. While containerization offers numerous benefits, it still faces specific challenges, one of them being the ability to provide robustness and resiliency of the containerized network functions under challenging network conditions, e.g., varying load conditions and application types. The currently available open-source container platform(s) do not achieve a full-fledged containerized mobile network protocol stack as they fail to support successful live migration of containers that execute specific 5G software functionalities.The contribution of this study is to present a proof-of-concept implementation that completes the live migration of a containerized 5G Central Unit (CU) module between two servers without causing permanent service disruption to the mobile user. The authors’ implementation makes use of an open-source migration software (CRIU) that has been modified to support the protocol (SCTP) used by the CU to exchange signaling messages with other 5G modules. The authors’ implementation is experimentally tested in a federated testbed environment by applying four container migration techniques, namely Cold, Pre-Copy, Post-Copy, and Hybrid-Copy migration. A performance comparison reveals that the end-user service recovery time due to the CU live migration can be reduced by up to 36% when applying the hybrid migration technique compared to the conventional cold migration technique. Additionally, experimental results reveal that compared to state-of-the-art VM migration the container-based migration approach reduces both migration time and service downtime by 96.7% and 75.2%, respectively.

Resource allocation and energy trading in the short blocklength regime for radio access networks

AbstractWith the large‐scale deployment of communication and computation units, the conventional radio access network has been gradually transformed into a sophisticated Cyber‐Physical System. Since the access network is put forward higher expectations for ultra‐reliable and low‐latency services, Short‐Packet Communications (SPC) based on Finite Blocklength Codes can be adopted to further decrease network latency and enhance communication reliability. Considering the interaction among the service provider, the network operator, and users, a three‐stage Stackelberg game is constructed to dynamically arrange communication and computation resources. In this paper, the tripartite interaction problem is modelled as a generalized Nash equilibrium problem, and the reverse induction method is employed to achieve the sub‐game equilibrium. To solve the problem of maximizing tripartite utilities, the authors propose an iterative algorithm for jointly managing resource allocation and tripartite pricing. Simulation results show that compared with the benchmark, the proposed scheme can effectively improve the utilities of users, the service provider, and the network operator and achieve a win–win situation.

ColO-RAN: Developing Machine Learning-Based xApps for Open RAN Closed-Loop Control on Programmable Experimental Platforms

Cellular networks are undergoing a radical transformation toward disaggregated, fully virtualized, and programmable architectures with increasingly heterogeneous devices and applications. In this context, the open architecture standardized by the O-RAN Alliance enables algorithmic and hardware-independent Radio Access Network (RAN) adaptation through closed-loop control. O-RAN introduces Machine Learning (ML)-based network control and automation algorithms as socalled xApps running on RAN Intelligent Controllers. However, in spite of the new opportunities brought about by the Open RAN, advances in ML-based network automation have been slow, mainly because of the unavailability of large-scale datasets and experimental testing infrastructure. This slows down the development and widespread adoption of Deep Reinforcement Learning (DRL) agents on real networks, delaying progress in intelligent and autonomous RAN control. In this paper, we address these challenges by discussing insights and practical solutions for the design, training, testing, and experimental evaluation of DRLbased closed-loop control in the Open RAN. To this end, we introduce ColO-RAN, the first publicly-available large-scale O-RAN testing framework with software-defined radios-in-the-loop. Building on the scale and computational capabilities of the Colosseum wireless network emulator, ColO-RAN enables ML research at scale using O-RAN components, programmable base stations, and a ”wireless data factory”. Specifically, we design and develop three exemplary xApps for DRL-based control of RAN slicing, scheduling and online model training, and evaluate their performance on a cellular network with 7 softwarized base stations and 42 users. Finally, we showcase the portability of ColO-RAN to different platforms by deploying it on Arena, an indoor programmable testbed. The lessons learned from the ColO-RAN implementation and the extensive results from our first-of-its-kind large-scale evaluation highlight the importance of experimental frameworks for the development of end-toend intelligent RAN control pipelines, from data analysis to the design and testing of DRL agents. They also provide insights on the challenges and benefits of DRL-based adaptive control, and on the trade-offs associated to training on a live RAN. ColO-RAN and the collected large-scale dataset are publicly available to the research community.

Consortium Blockchain-Based Spectrum Trading for Network Slicing in 5G RAN: A Multi-Agent Deep Reinforcement Learning Approach

Network slicing (NS) is envisioned as an emerging paradigm for accommodating different virtual networks on a common physical infrastructure. Considering the integration of blockchain and NS, a secure decentralized spectrum trading platform can be established for autonomous radio access network (RAN) slicing. Moreover, the realization of proper incentive mechanisms for fair spectrum trading is crucial for effective RAN slicing. This paper proposes a novel hierarchical framework for blockchain-empowered spectrum trading for NS in RAN. Specifically, we deploy a consortium blockchain platform for spectrum trading among spectrum providers and buyers for slice creation, and autonomous slice adjustment. For slice creation, the spectrum providers are infrastructure providers (InPs) and buyers are mobile virtual network operators (MVNOs). Then, underloaded MVNOs with extra spectrum to spare, trade with overloaded MVNOs, for slice spectrum adjustment. For proper incentive maximization, we propose a three-stage Stackelberg game framework among InPs, seller MVNOs, and buyer MVNOs, for joint optimal pricing and demand prediction strategies. Then, a multi-agent deep reinforcement learning (MADRL) method is designed to achieve a Stackelberg equilibrium (SE). Security assessment and extensive simulation results confirm the security and efficacy of our proposed method in terms of players’ utility maximization and fairness, compared with other baselines.

Next Generation Mobile Networks' Enablers: Machine Learning-Assisted Mobility, Traffic, and Radio Channel Prediction

Machine learning (ML) is an important component for enabling automation in radio access networks (RANs). The work on applying ML for RAN has been under development for several years and is now also drawing attention in 3GPP standardization fora. A key component of multiple features, highlighted in the recent 3GPP specification work, is the use of mobility, traffic and radio channel prediction. These types of predictions form intelligence enablers to leverage the potentials of ML for RAN enhancements, in both current and future wireless networks. Our contributions are twofold, first we provide an overview with representative evaluation results of current applications that utilize these intelligence enablers. Next, we discuss how those enablers likely will be a cornerstone for emerging 6G use cases such as wireless energy harvesting. As the journey to 6G remains an open research area, we highlight how the development of these enablers can unlock new features in future mobile networks.

Boosting 5G on Smart Grid Communication: A Smart RAN Slicing Approach

Fifth-generation (5G) and beyond systems are expected to accelerate the ongoing transformation of power systems towards the smart grid. However, the inherent heterogeneity in smart grid services and requirements pose significant challenges towards the definition of a unified network architecture. In this context, radio access network (RAN) slicing emerges as a key 5G enabler to ensure interoperable connectivity and service management in the smart grid. This article introduces a novel RAN slicing framework which leverages the potential of artificial intelligence (AI) to support IEC 61850 smart grid services. With the aid of deep reinforcement learning, efficient radio resource management for RAN slices is attained, while conforming to the stringent performance requirements of a smart grid self-healing use case. Our research outcomes advocate the adoption of emerging AI-native approaches for RAN slicing in beyond-5G systems, and lay the foundations for differentiated service provisioning in the smart grid.

Knowledge Transfer and Reuse: A Case Study of Ai-Enabled Resource Management in RAN Slicing

An efficient resource management scheme is critical to enable network slicing in 5G networks and in envisioned 6G networks, and artificial intelligence (AI) techniques offer promising solutions. Considering the rapidly emerging new machine learning techniques, such as graph learning, federated learning, and transfer learning, a timely survey is needed to provide an overview of resource management and network slicing techniques of AI-enabled wireless networks. This article provides such a survey along with an application of knowledge transfer in radio access network (RAN) slicing. In particular, we first provide some background on resource management and network slicing, and review relevant state-of-the-art AI and machine learning (ML) techniques and their applications. Then, we introduce our AI-enabled knowledge transfer and reuse-based resource management (AKRM) scheme, where we apply transfer learning to improve system performance. Compared with most existing works, which focus on the training of standalone agents from scratch, the main difference of AKRM lies in its knowledge transfer and reuse capability between different tasks. Our paper aims to be a roadmap for researchers to use knowledge transfer schemes in AI-enabled wireless networks, and we provide a case study over the resource allocation problem in RAN slicing.

Optimization of Channel Segregation-Based Fractional Frequency Reuse for Inter-Cell Interference Coordination in Cellular Ultra-Dense RAN

To cope with ever growing mobile data traffic, we recently proposed a concept of cellular ultra-dense radio access network (RAN). In the cellular ultra-dense RAN, a number of distributed antennas are deployed in the base station (BS) coverage area (cell) and user-clusters are formed to perform small-scale distributed multiuser multi-input multi-output (MU-MIMO) transmission and reception in each user-cluster in parallel using the same frequency resource. We also proposed a decentralized interference coordination (IC) framework to effectively mitigate both intra-cell and inter-cell interferences caused in the cellular ultra-dense RAN. The inter-cell IC adopted in this framework is the fractional frequency reuse (FFR), realized by applying the channel segregation (CS) algorithm, and is called CS-FFR in this paper. CS-FFR divides the available bandwidth into several sub-bands and allocates multiple sub-bands to different cells. In this paper, focusing on the optimization of the CS-FFR, we find by computer simulation the optimum bandwidth division number and the sub-band allocation ratio to maximize the link capacity. We also discuss the convergence speed of CS-FFR in a cellular ultra-dense RAN.

Impact of information update generation on the radio access network of cellular Internet of Things

Impact of information update generation on the radio access network of cellular Internet of Things

Location Aware Beamforming in Millimeter-Wave Band Ultra-Dense Radio Access Networks. Part 1. Model of Two Links

The evolution of 1G to 4G radio access networks (RANs) over the past 40 years has shown that beamforming (BF) capabilities add an additional spatial dimension to traditional device multiplexing methods. When base stations (gNodeB - gNB) and user equipment (UE) form narrow antenna radiation patterns (APPs), in addition to frequency, time and code division of channels, an additional spatial dimension appears that implements spatial multiplexing. This concept has been known for quite a long time, but the full implementation of its capabilities in practice is expected with the widespread adoption of millimeter wave (mmWave) ultra-dense networks (UDN) of the fifth (5G) and subsequent (B5G) generations. To control APP, the approach of preliminary analysis of training sequences about the current situation in the radio channel - CSI (Channel State Information) - can be used, but its overhead costs become unacceptably high in conditions of ultra-dense distribution of devices. An alternative approach is positioning-based BF. The validity, relevance and prospects of this approach are determined by the fact that for 5G networks, unlike previous generations, for the first time the requirements for UE positioning accuracy up to one meter are formalized. Initial research in the field of location-aware BF has already been carried out over the past years, however, mainly for particular scenarios of one or more radio links between gNBs and fixed UEs. In this work, for the first time, a scientifically based methodology for controlling the beam pattern of a stationary gNB based on the positioning of a mobile UE for a two-radio link scenario is formalized and implemented in software. The problem of practical implementation of BF is the difficulties to predict level of interference due to the mutual influence of radio links with mobile UEs. When estimating the instantaneous signal-to-interference ratio in a two-radio link scenario between two fixed gNBs that perform BF based on the current location of mobile UEs as they move, it is necessary to take into account the mutual influence of each other's radio links on each other. In such a scenario, a transmitter on one radio link acts both as a source of a wanted signal for one UE and as a source of an interfering signal for another UE. The task of assessing interference for such a scenario is complicated by the nonlinearity of the transmitter and/or receiver ARPs. The model developed and implemented in software in this work uses the functions of the Phased Array System Toolbox Matlab extension package. The simulation results show a significant scatter (tens of dB) of the instantaneous signal-to-interference ratio depending on the territorial separation of devices and can be used to justify scenarios for the construction and operation of 5G/B5G UDN.

Joint optimization of IoT devices access and bandwidth resource allocation for network slicing in edge-enabled radio access networks

Joint optimization of IoT devices access and bandwidth resource allocation for network slicing in edge-enabled radio access networks

An Enhanced Deep Reinforcement Learning-based Slice Acceptance Control System (EDRL-SACS) for Cloud–Radio Access Network

In 5G networks, Network Slicing (NS) allows service providers (SPs) to create customized standalone networks on shared platforms provided by infrastructure providers (InPs). However, the challenge arises when prioritizing slice requests sent by SPs to InPs due to limited and shared resources. The Cloud Radio Access Network (C-RAN) architecture is a novel approach that supports various services in 5G networks, but efficient resource utilization and increased revenue across the three layers of C-RAN are necessary. To address this issue, we propose an Enhanced Deep Reinforcement Learning-based Slice Acceptance Control System (EDRL-SACS) technique to maximize network utility by accepting suitable requests. EDRL-SACS considers time-varying resource situations, various service requirements, and the frequency of requests from SPs to endorse model-free solutions for Slice Acceptance Control (SAC). By adding effective parameters to the deep reinforcement learning algorithm, the proposed methodology improves its efficiency and performance resulting in better resource utilization, coverage, revenue, and rejection of users’ request terms. The system’s evaluation shows that EDRL-SACS performs well and effectively maximizes the utility of the network by providing slices to SPs in response to their slice requests.

Simple and low-latency multipoint-to-point aggregation architecture using commercial optical transceivers for uplink fronthaul.

The increasing demand of real-time applications poses a huge challenge to building next-generation radio access network (NG-RAN) with higher stability and lower system complexity. Parallel signal detection (PSD), which aggregates signals of different intermediate frequencies (IFs) on different wavelengths with a single photodiode (PD), becomes a promising candidate for uplink mobile fronthaul with the advantage of low-latency. However, high requirements on the transmitters inhibit the large-scale deployment of radio units (RU). In this paper, we propose an economical, low-latency, multipoint-to-point (MP2P) uplink fronthaul architecture capable of aggregating four end-users with commercial 25G-class optical modules and a single PD. With delta-sigma modulation (DSM), commercial off-the-shelf optical modules can replace analog transmitters in traditional systems. As a demonstration, we aggregated 4 × 380.16-MHz 5 G new radio (NR) orthogonal frequency division multiplexing (OFDM) signals in an IF band with a fixed interval of 400 MHz over 20 km fiber with 4 users.