Optimization of Radio Access Networks Architecture for 5G

This post is prepared for non-technical readers and the general public.

Introduction

Mobile networks are growing fast and pushed by various applications, demanding lower latency, higher traffic volumes, and ultra-reliable connectivity. This led industry to renew their networks while keeping the costs down in order to meet diversified demands with competitive prizes.

The International Telecommunication Union (ITU) categorized 5G mobile network services into three use cases [1]:

• Enhanced Mobile Broadband (eMBB) which tries to meet the services that need higher- bandwidth, such as high definition (HD) videos, virtual reality (VR), and augmented reality (AR) (Figure 1).

Figure 1. The real-time mixed reality with AR and VR services [2]

• Ultra-reliable and Low-latency Communications (uRLLC) focuses on the demanding digital industry to meet low latency services, such as assisted and automated driving (Figure 2).
• Massive Machine Type Communications (mMTC) meet demands for the services that include high requirements for connection density, such as smart city and smart agriculture.

Figure 2. The assisted and automated driving [3]

To fully meet these application demands, RAN architecture entails to be flexible enough in order to support improved resource pooling and to have higher spectral efficiency over various transport network configurations.

The new architecture of RAN, centralized RAN (C-RAN) that has recently gained momentum, is capable of splitting Base Station (BS) functions from Remote Radio Heads (RRHs) to be pooled at Central Units (CUs). All RRHs are connected to CUs through the transport network. Figure 3, shows the structure of C-RAN overview with Core Network and Internet. In C-RAN there exists three RRHs with Virtualized Network Functions (VNF). RRHs are connected to a CU via the transport network, and the CU is connected to the Core network via the backhaul link. Finally, the Core is connected to the Internet.

Figure 3. The architecture overview of C-RAN with Core network and Internet

This architecture of C-RAN helps mobile network operators to support the various demands in 5G networks. It also brings advantages in both network systems and user experience part.

The Open Challenges and Countermeasures

To be clearer, our approach within the framework of SPOTLIGHT will elaborate the current/upcoming challenges with the three most common use cases:

First, we study a stadium in which the audience tries to send the status of the match or upload some nice photos to Instagram. This case is a clear example of the massive connectivity required by mMTC.

The second use case under study is the eMBB. A good example of this use case is when a user tries to connect to the internet and watches this match from home with Full HD video quality.

The third case is the applications of Tactile internet, where the expertise of medical doctors connected through the Internet during surgery, remote diagnosis and/or treatment. This use case falls into the uRLLC.

In such cases, where we have a different kind of application requirements, splitting network functions of C-RAN into CU or RRH allocation is vital to optimize the network performance. The main objective of this technique is to fairly split NFs in either CU or RRHs. Indeed this technology allows us to manage the network resources based on the application requirements. For example, for uRLLC users which need low-latency, most NFs must be allocated in RRH side. Conversely, for eMBB users which needs higher traffic load, most of NFs should be migrated to CU to be managed centrally. Other useful techniques can be the utilization of Software Defined Network (SDN) and Network Function Virtualization (NFV). These technologies help virtualizing the hardware of network (here RAN architecture) and using resources of the network more efficiently. The resources of the network can be CPU, memory, computation capacities of BSs and spectrum.

Our Solution (ongoing)

Our ongoing work is focused on the utilization of the concepts of both the aforementioned technologies and optimize the deployment cost of BSs while satisfying maximum possible services. Hence, we are virtualizing NFs of RAN in order to meet different requirements of users and better use the network resources. Furthermore, we are splitting NFs based on the received demands, and meantime, minimizing the deployment cost which finally will end with the lower price for the customers.

In fact, our research work integrates the recent technologies to jointly optimize the system gains (i.e., network resources) and the user’s experiences.

References:

[1] Recommendation ITU-R M.2083-0. IMT Vision—Framework and Overall Objectives of the Future Development of IMT for 2020 and Beyond.

[2] Yu, H.; Lee, H.; Jeon, H. What is 5G? Emerging 5G Mobile Services and Network Requirements. Sustainability2017, 9, 1848.

[3] Available online: https://www.avl.com/-/system-engineering-for-assisted-autonomous-driving