The fifth generation (5G) wireless network technology is to be standardized by 2020, where main goals are to improve capacity, reliability, and energy efficiency, while reducing latency and massively increasing connection density. An integral part of 5G is the capability to transmit touch perception type real-time communication empowered by applicable robotics and haptics equipment at the network edge. In this regard, we need drastic changes in network architecture including core and radio access network (RAN) for achieving end-to-end latency on the order of 1 ms. In this paper, we present a detailed survey on the emerging technologies to achieve low latency communications considering three different solution domains: 1) RAN; 2) core network; and 3) caching. We also present a general overview of major 5G cellular network elements such as software defined network, network function virtualization, caching, and mobile edge computing capable of meeting latency and other 5G requirements.

A Survey on Low Latency Towards 5G: RAN, Core Network and Caching Solutions

Ubiquitous sensors and smart devices from factories and communities are generating massive amounts of data, and ever-increasing computing power is driving the core of computation and services from the cloud to the edge of the network. As an important enabler broadly changing people’s lives, from face recognition to ambitious smart factories and cities, developments of artificial intelligence (especially deep learning, DL) based applications and services are thriving. However, due to efficiency and latency issues, the current cloud computing service architecture hinders the vision of “providing artificial intelligence for every person and every organization at everywhere”. Thus, unleashing DL services using resources at the network edge near the data sources has emerged as a desirable solution. Therefore, edge intelligence , aiming to facilitate the deployment of DL services by edge computing, has received significant attention. In addition, DL, as the representative technique of artificial intelligence, can be integrated into edge computing frameworks to build intelligent edge for dynamic, adaptive edge maintenance and management. With regard to mutually beneficial edge intelligence and intelligent edge , this paper introduces and discusses: 1) the application scenarios of both; 2) the practical implementation methods and enabling technologies, namely DL training and inference in the customized edge computing framework; 3) challenges and future trends of more pervasive and fine-grained intelligence. We believe that by consolidating information scattered across the communication, networking, and DL areas, this survey can help readers to understand the connections between enabling technologies while promoting further discussions on the fusion of edge intelligence and intelligent edge , i.e., Edge DL.

Convergence of Edge Computing and Deep Learning: A Comprehensive Survey

Ubiquitous sensors and smart devices from factories and communities are generating massive amounts of data, and ever-increasing computing power is driving the core of computation and services from the cloud to the edge of the network. As an important enabler broadly changing people's lives, from face recognition to ambitious smart factories and cities, developments of artificial intelligence (especially deep learning, DL) based applications and services are thriving. However, due to efficiency and latency issues, the current cloud computing service architecture hinders the vision of "providing artificial intelligence for every person and every organization at everywhere". Thus, unleashing DL services using resources at the network edge near the data sources has emerged as a desirable solution. Therefore, edge intelligence, aiming to facilitate the deployment of DL services by edge computing, has received significant attention. In addition, DL, as the representative technique of artificial intelligence, can be integrated into edge computing frameworks to build intelligent edge for dynamic, adaptive edge maintenance and management. With regard to mutually beneficial edge intelligence and intelligent edge, this paper introduces and discusses: 1) the application scenarios of both; 2) the practical implementation methods and enabling technologies, namely DL training and inference in the customized edge computing framework; 3) challenges and future trends of more pervasive and fine-grained intelligence. We believe that by consolidating information scattered across the communication, networking, and DL areas, this survey can help readers to understand the connections between enabling technologies while promoting further discussions on the fusion of edge intelligence and intelligent edge, i.e., Edge DL.

The sixth generation (6G) wireless communication networks are envisioned to revolutionize customer services and applications via the Internet of Things (IoT) towards a future of fully intelligent and autonomous systems. In this article, we explore the emerging opportunities brought by 6G technologies in IoT networks and applications, by conducting a holistic survey on the convergence of 6G and IoT. We first shed light on some of the most fundamental 6G technologies that are expected to empower future IoT networks, including edge intelligence, reconfigurable intelligent surfaces, space-air-ground-underwater communications, Terahertz communications, massive ultra-reliable and low-latency communications, and blockchain. Particularly, compared to the other related survey papers, we provide an in-depth discussion of the roles of 6G in a wide range of prospective IoT applications via five key domains, namely Healthcare Internet of Things, Vehicular Internet of Things and Autonomous Driving, Unmanned Aerial Vehicles, Satellite Internet of Things, and Industrial Internet of Things. Finally, we highlight interesting research challenges and point out potential directions to spur further research in this promising area.

6G Internet of Things: A Comprehensive Survey

Massive machine-type communications (mMTC) is one of the main three focus areas in the 5th generation (5G) of wireless communications technologies to enable connectivity of a massive number of Internet of things (IoT) devices with little or no human intervention. In conventional human-type communications (HTC), due to the limited number of available channel resources and orthogonal resource allocation techniques, users get a transmission slot by making scheduling/connection requests. The involved control channel signaling, negligible with respect to the huge transmit data, is not a major issue. However, this may turn into a potential performance bottleneck in mMTC, where huge number of devices transmit short packet data in a sporadic way. To tackle the limited radio resources and massive connectivity challenges, non-orthogonal multiple access (NOMA) has emerged as a promising technology that allows multiple users to simultaneously transmit their data over the same channel resource. This is achieved by employing user-specific signature sequences at the transmitting devices, which are exploited by the receiver for multi-user data detection. Due to its massive connectivity potential, NOMA has also been considered to enable grant-free transmissions especially in mMTC, where devices can transmit their data whenever they need without the scheduling requests. The existing surveys majorly discuss different NOMA schemes, and exploit their potential, in typical grant-based HTC scenarios, where users are connected with the base station, and various system parameters are pre-defined in the scheduling phase. Different from these works, this survey provides a comprehensive review of the recent advances in NOMA from a grant-free connectivity perspective. Various grant-free NOMA schemes are presented, their potential and related practical challenges are highlighted, and possible future directions are thoroughly discussed at the end.

Grant-Free Non-Orthogonal Multiple Access for IoT: A Survey

Fifth-generation cellular mobile networks are expected to support mission critical URLLC services in addition to enhanced mobile broadband applications. This article first introduces three emerging mission critical applications of URLLC and identifies their requirements on end-to-end latency and reliability. We then investigate the various sources of end-to-end delay of current wireless networks by taking 4G LTE as an example. Then we propose and evaluate several techniques to reduce the end-to-end latency from the perspectives of error control coding, signal processing, and radio resource management. We also briefly discuss other network design approaches with the potential for further latency reduction.

Ultra-Reliable Low Latency Cellular Networks: Use Cases, Challenges and Approaches

In this paper, we consider a mobile edge computing system with both ultra-reliable and low-latency communications services and delay tolerant services. We aim to minimize the normalized energy consumption, defined as the energy consumption per bit, by optimizing user association, resource allocation, and offloading probabilities subject to the quality-of-service requirements. The user association is managed by the mobility management entity (MME), while resource allocation and offloading probabilities are determined by each access point (AP). We propose a deep learning (DL) architecture, where a digital twin of the real network environment is used to train the DL algorithm off-line at a central server. From the pre-trained deep neural network (DNN), the MME can obtain user association scheme in a real-time manner. Considering that the real networks are not static, the digital twin monitors the variation of real networks and updates the DNN accordingly. For a given user association scheme, we propose an optimization algorithm to find the optimal resource allocation and offloading probabilities at each AP. The simulation results show that our method can achieve lower normalized energy consumption with less computation complexity compared with an existing method and approach to the performance of the global optimal solution.

Deep Learning for Hybrid 5G Services in Mobile Edge Computing Systems: Learn From a Digital Twin

In cloud radio access networks (C-RANs), the baseband processing units (BBUs) from traditional base stations (BSs) are virtualized and moved into a single centralized location. By virtualization, the computing resources of BBUs can be dynamically shared among all cells, enabling a significant improvement in computing resource utilization and power efficiency. In this letter, we propose a BBUs virtualization scheme that minimizes the power consumption with a linear computational complexity order. The scheme is based on a heuristic simulated annealing (HSA) algorithm, which combines a bin packing algorithm with SA. Simulation results show that the HSA effectively decreases system power consumption when compared to standard approaches.

Baseband Processing Units Virtualization for Cloud Radio Access Networks

In future 6th generation networks, URLLC will lay the foundation for emerging mission-critical applications that have stringent requirements on end-to-end delay and reliability. Existing works on URLLC are mainly based on theoretical models and assumptions. The model-based solutions provide useful insights, but cannot be directly implemented in practice. In this article, we first summarize how to apply data-driven supervised deep learning and deep reinforcement learning in URLLC, and discuss some open problems of these methods. To address these open problems, we develop a multi-level architecture that enables device intelligence, edge intelligence, and cloud intelligence for URLLC. The basic idea is to merge theoretical models and realworld data in analyzing the latency and reliability and training deep neural networks (DNNs). Deep transfer learning is adopted in the architecture to fine-tune the pre-trained DNNs in non-stationary networks. Further considering that the computing capacity at each user and each mobile edge computing server is limited, federated learning is applied to improve the learning efficiency. Finally, we provide some experimental and simulation results and discuss some future directions.

Deep Learning for Ultra-Reliable and Low-Latency Communications in 6G Networks

In 3GPP Long Term Evolution (LTE) networks, the frequency reuse schemes such as fractional frequency reuse (FFR) and soft frequency reuse (SFR) are used to improve system capacity. The allocation of transmit power and subcarriers to each cell in these schemes are fixed prior to network deployment. This limits the potential performance of these frequency reuse schemes. In this paper, we propose to improve the capacity of SFR scheme by jointly optimizing subcarrier and power allocation in multi-cell LTE networks. An iterative algorithm that can adaptively vary the number of major subcarriers and adjust the transmit power for each cell according to wireless traffic loads is proposed. Simulation results show that the proposed algorithm outperforms the existing Reuse 1, FFR and static SFR schemes in both system throughput and cell edge user performance.

Wibowo Hardjawana

Papers

Ultra-Reliable Low Latency Cellular Networks: Use Cases, Challenges and Approaches

Deep Learning for Hybrid 5G Services in Mobile Edge Computing Systems: Learn From a Digital Twin

Baseband Processing Units Virtualization for Cloud Radio Access Networks

Deep Learning for Ultra-Reliable and Low-Latency Communications in 6G Networks

Inter-cell interference coordination through adaptive soft frequency reuse in LTE networks