A Survey on Low Latency Towards 5G: RAN, Core Network and Caching Solutions (1708.02562v2)

Abstract: 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: RAN, core network, and caching. We also present a general overview of 5G cellular networks composed of software defined network (SDN), network function virtualization (NFV), caching, and mobile edge computing (MEC) capable of meeting latency and other 5G requirements.

• The paper demonstrates that RAN enhancements, including flexible TTI structures and advanced waveforms, can achieve sub-millisecond latency.
• The paper shows that core network innovations like SDN, NFV, and MEC streamline routing to significantly reduce data transmission delays.
• The paper identifies effective caching strategies that place content closer to users, thereby alleviating backhaul congestion and lowering latency.

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

The evolution of mobile communication technologies has progressively enhanced network capabilities. As we move towards the fifth generation (5G) of wireless networks, key objectives include unprecedented improvements in capacity, reliability, energy efficiency, and notably, drastic reductions in latency. The comprehensive paper "A Survey on Low Latency Towards 5G: RAN, Core Network and Caching Solutions" explores the architectural and technological advancements required to achieve end-to-end (E2E) latency on the order of 1 ms in 5G networks.

Overview of 5G Low Latency Requirements

The paper underscores the importance of low latency in 5G, which is vital for applications such as tactile internet, high-resolution video streaming, autonomous driving, smart grids, telemedicine, and industrial automation. These applications necessitate stringent requirements characterized by throughput, reliability, and remarkably low latency. For instance, mission-critical communications demand E2E latency as low as 1 ms and reliability as high as 99.99%.

Radio Access Network (RAN) Solutions

The paper delineates various enhancements at the RAN level to meet low latency requirements:

• Frame/Packet Structure: The introduction of flexible TTI sizes, numerology, and subframe structures is essential. The scalable TDD frame structure ensures latency reduction by adapting TTI sizes to traffic requirements.
• Advanced Multiple Access Techniques/Waveforms: Techniques like Filtered OFDM (f-OFDM), Universal Filtered Multi-Carrier (UFMC), and Sparse Code Multiple Access (SCMA) promise to enhance latency and reliability by mitigating issues related to orthogonality and synchronization intrinsic to traditional OFDM.
• Modulation and Coding: The paper highlights Polar codes and improved Turbo decoding algorithms for small packet transmissions, enhancing reliability and reducing re-transmission delays.
• Control Signaling: Enhancements in control channel sparse encoding and allocating dedicated uplink control channels for sporadic traffic significantly reduce latency.
• Symbol Detection: Low-complexity receivers, compressed sensing techniques, and advanced MIMO detection algorithms are vital for reducing processing time and enhancing throughput.

A key emphasis is on the exploitation of mmWave spectrum for its potential to provide large bandwidths, thereby significantly reducing latency.

Core Network Innovations

Transformation in the core network is equally critical:

• Software Defined Networking (SDN): SDN decouples the control plane from the data plane, optimizing route selection and enabling adaptive and programmable network management, which is instrumental in reducing latency.
• Network Function Virtualization (NFV): NFV orchestrates network functions through software virtualization, alleviating dependencies on hardware and allowing dynamic scaling of network resources.
• Mobile Edge Computing (MEC)/Fog Networks: MEC enhances network performance by deploying computation and storage resources closer to the user, thus minimizing the E2E latency.

Backhaul Solutions

Efficient and scalable backhaul solutions are integral to achieving low latency:

• Unified Packet-Based Transport Networks: Approaches like the MAC-in-MAC Ethernet and the use of SDN controllers ensure seamless integration of fronthaul and backhaul traffic, reducing procedural latency.
• mmWave Backhaul: The use of mmWave technologies in fronthaul and backhaul operations is explored to provide high capacity and low latency, addressing the needs of ultra-dense 5G networks.

Caching Solutions

Edge caching significantly contributes to low latency communications:

• Distributed and Centralized Caching: Placing popular content closer to the network edge mitigates backhaul congestion and reduces content delivery time. The paper investigates cache placement strategies and content delivery algorithms tailored to latency-critical applications.
• Latency-Storage Trade-off: The paper probes into the trade-offs between storage capacity and delivery latency, offering insights into optimal caching strategies that balance network resource utilization and latencies.

Field Tests, Trials, and Experiments

The paper compiles results from various field tests and system trials, underscoring practical implications and feasibility of the proposed solutions. For instance, experiments validate that tailored frame structures, efficient control channel designs, and advanced modulation schemes can achieve sub-millisecond latencies under real-world conditions.

Implications and Future Research

The insights provided in this paper have substantial practical and theoretical implications:

• Practical Applications: Implementing these solutions in 5G networks will facilitate latency-sensitive applications, improving user experience in IoT, VR/AR, and critical infrastructure communications.
• Theoretical Advances: The paper prompts further research in optimizing cross-layer design, enhancing mmWave communication, and developing robust network function virtualization frameworks.
• Sustainability and Scalability: Ensuring the scalability of latency-reducing solutions while maintaining energy efficiency and minimizing control overhead remains a critical focus for future research.

Conclusion

In conclusion, the quest for achieving ultra low latency in 5G involves comprehensive advancements in various network domains including the RAN, core network, and caching strategies. While significant strides have been made, ongoing research addressing open issues and challenges will pave the way for fully realizing the 5G vision. The synthesized solutions and their empirical validations in this paper provide a solid foundation for both academic researchers and industry practitioners striving to meet the stringent latency requirements of next-generation wireless networks.