Much has been written about the technologies that are driving 5G, especially how those technologies will improve the user experience in terms of connectivity. Similarly, much has been said about how ongoing technological developments will usher in a new generation of network-enabled applications. In this article, we will discuss a key aspect of 5G technology and how it will affect the evolution of wireless network capacity.
This is one of the more important, but often neglected, aspects of wireless communications development. This is another big reason why the convergence of cloud computing and wireless communication makes so much sense. In short, many of the complex problems associated with 5G wireless networks can be solved with the help of software, eliminating the need for costly, time consuming, and often slow developing hardware that has been used in the past.
Cloud and Telecommunications: Fits perfectly with next generation networks
It is well known that some of the most advanced technologies that make up 5G can be implemented in software that runs on standard servers. This is exciting because we can slowly but surely move away from the specialized hardware that has been used in all of the previous four generations of telecommunications networks. The move to software will help lower the total cost of capital and operating costs for telecommunications providers. Equally important, this hardware-to-software shift will future-proof such networks as it enables the telecommunications industry to become agile and aggressive with the regular introduction of desirable features, rather than having to wait a decade or so for the next generation of standards to emerge. Innovation will thrive if we create a world where moving from one generation to another is a software upgrade, just as the cloud industry has been doing for over a decade.
We’ll say more about this in future blogs, but today we want to discuss wireless capacity – or technically spectrum efficiency. Hopefully we will convince you that computing power can be used to increase the capacity of cellular networks, and advances in software-based machine learning and data analysis techniques can be used to improve the efficiency of 5G and future networks. When you add this to the other elements of the ecosystem, the connection between cloud computing and telecommunications networks fits perfectly.
5G core technologies: Massive multi-user MIMO
Multiple-Input and Multiple-Output (MIMO) is a method of multiplying the capacity of a radio link using multiple transmit and receive antennas in order to take advantage of multipath propagation. MIMO is an essential element of the wireless communication standards in Wi-Fi, 3G and 4G. However, 5G takes it to the next level with massive multi-user (MU) MIMO by massively scaling the number of antennas and supporting many users at the same time. This technology is key to the promise of 5G, which promises 1,000 times the capacity gain over 4G.
The science behind massive MU-MIMO lies in the complex math involved in manipulating signals sent to and received by each antenna so that the channels of communication are preserved with each user and to survive the environmental distortion. This has been the subject of many reference books and scientific studies, but a simplified version can be found in the figure below.
Massive MU-MIMO contains many matrix multiplications and transpositions, all of which require a considerable amount of computation. It is a direct function of the number of users served by the cell tower and the number of antennas the cell tower has. In addition, this calculation takes place every few milliseconds for thousands of sub-carriers. The implication here is that significant computing power and energy are required. As network operators increase the number of antennas, the computational effort increases significantly, along with other related problems.
User patterns also affect the computational effort required. The precoding method described in the above figure works best when users are stationary or moving slowly. Otherwise the precoding matrix has to be recalculated frequently, which requires even more calculations. An alternative technique known as conjugate beamforming may work better in this case, but the number of antennas must far exceed the number of users and wireless capacity is typically reduced.
The total capacity that the network delivers depends directly on how much computing power the operator is willing to buy and use on each of its thousands of cell towers. Edge computingthat offers the ability to easily scale computing is perfect for this. While some operators may not need a lot of capacity right away, it is still important to understand whether the network should be built so that it can easily be scaled as network capacity needs increase.
Microsoft has invested heavily in computing technology that can deliver massive MU-MIMO for 5G networks. Back in 2012, Microsoft Research invested in a hands-on solution to implement MU-MIMO by using distributed pipelines with a rack of standard servers (an edge data center) to meet timing specifications and scale to hundreds of antennas (the Technology was state-of-the-art and a report was posted at SIGCOMM 2013).
Deep learning for wireless capacity
5G is moving towards an open architecture with many options for optimizing a network. While this approach adds complexity, deep learning techniques can be used to address these complexities, which are usually beyond human capabilities. In the above case of precoding for massive MIMO, we can apply deep learning techniques to select an algorithm that will reduce power consumption while minimizing the reduction in capacity. With predictive analytics and modern software that adapts to dynamic network loads, 5G networks can become more intelligent.
Microsoft has invested heavily in machine learning and AI and has supported the work of the world’s leading experts in the field. And we’re working to expand telecommunications networks by developing deep learning algorithms that incorporate domain knowledge. In addition to the example above, we are actively investigating how deep learning techniques can be used to control transmit power in order to reduce interference and thus increase capacity.
Continuous machine learning (powered by flexible edge computing to model the dynamic radio frequency environment and user mobility patterns), along with managing the signal processing pipeline, creates a tremendous value proposition for the telecommunications industry. This massive step forward enables research breakthroughs to be rapidly incorporated into the system – not only to increase wireless capacity, but also to improve the overall operational efficiency of 5G networks.
Azure: Where edge computing, cloud and telecommunications operators come together
To the More than 10 years, Microsoft has invested a lot in Edge computing and continues to do so. In particular, Azure is working to provide calculations near cell towers, which is where carriers will benefit most if they want to cost-effectively scale their network. Additionally through his Azure for Operators initiative, Microsoft is continuously working to enable new first- and third-party solutions that further improve and simplify edge computing, from network connectivity to on-demand computing to full orchestration.
Given Microsoft’s ability to scale computations as far and as often as operators demand, the power of technology on the fringes – including massive MU-MIMO – is the answer telecom operators have been looking for. Azure is here to help telecom operators meet their capacity increase goals as the network grows and evolves. As telecommunications providers increase the number of antennas and cell towers, Microsoft’s ability to set up servers at scale and manage them from anywhere in the world makes Azure the perfect solution for 5G and beyond for telecommunications networks.
read this Azure for Operators eBook to learn more.