Bringing 5G networks indoors
With 5G networks now on the air from Tier One operators around the world, there is significant excitement about this new technology and what it means for the future. This optimism surrounding 5G is justified and expected given the range of new and improved capabilties including massive increases in broadband speeds, ultra low latency, support of massive numbers of IoT devices and mission-critical use cases requiring the highest levels of reliability and security. Some of these gains will be achievable with a software upgrade of existing LTE systems, while others are tied directly to new spectrum, radio technologies and deployment architectures.
When it comes to indoor networks, 5G means new experiences for consumers, new possibilities for operational efficiency with IoT, industrial automation and new communication services. However there are many questions in the industry surrounding the unique challenges and opportunities of bringing 5G indoors.
5G introduces both evolutionary and revolutionary changes compared to previous generations, resulting in questions related to how best to bring these new capabilities to indoor spaces. These questions are in three different areas:
- What 5G use cases are applicable indoors?
- What spectrum can be utilized?
- What are the challenges and benefits of different indoor network architectures: DAS, distributed radio, small cells?
This white paper will explore these questions, compare different options available to network operators for deploying 5G technology indoors, and provided recommendations on the optimal solutions.
The introduction of 4G LTE has revolutionized mobile broadband networking. Radical changes in consumer society have been enabled by massive increases in throughput and capacity, enabling entire new ecosystems around the smartphone app economy. In the same way that it was impossible to predict the transformation enabled by 4G, the industry is on the verge of an even more fundamental transformation with 5G. In addition to advances in multi-Gbps mobile broadband, 5G will introduce fundamentally new architectures to support Massive IoT and Critical Realtime IoT applications. Ericsson has divided 5G use cases into three primary categories, of which only a subset are relevant indoors.
Enhanced mobile broadband
Enhanced mobile broadband is the step change in bandwidth coming from 5G that will enable new user experiences that benefit from that increase. While 4G LTE is evolving and offers speeds over 1Gbps with advances such as LAA and Massive MIMO, 5G will enable bandwidths up to 10 and 20 Gbps depending on available spectrum. In a building, this will enable new Augmented Reality, Virtual Reality and Mixed Reality applications delivered to wireless devices which has application in manufacturing and entertainment. 5G will also enable wireless streaming cameras with 4K and even 8K resolution. This could enable wireless security cameras, streaming of high-resolution video for training or education, or in more public spaces, upload of video to social media. These use cases are all relevant in indoor environments.
Peak data rates and capacity for mobile broadband are constrained by the amount of RF spectrum available – although 5G can use existing spectrum more efficiently than LTE in some cases, improvements in mobile broadband speeds would be constrained by available spectrum. Fortunately, national regulatory agencies around the world have responded to this need.
Massive IoT is the range of use cases that rely on broad, low cost coverage to carry very small data messages from thousands of devices. These are not time critical messages but rather insensitive to small delays, instead prioritizing cost and battery life.
Primary examples of Massive IoT are for smart meters, sensor networks and tracking. All three of these use cases could be applicable in buildings. Some use cases are identified below:
- New smart building technologies to track and manage the indoor environment - temperature sensors, motion sensors, light sensors all tied to the HVAC units, automated window shading and lighting systems
- Object tracking – a hospital, for example, tracking beds, equipment carts, individual portable machines. A stadium can track fresh beer kegs (and their temperature), lines at concessions and restrooms, and other opportunities. Other examples include an airport tracking wheelchairs, carts, and accessibility vehicles
- Tracking scanners, devices, and other equipment and logistic applications in indoor environments like factories or distribution centres
Massive IoT applications are beginning to be deployed today with 4G LTE technology. The LTE standard contains specific techniques to enable low-bandwidth, very low power devices to coexist in the same network as high-performance smartphones. 5G builds on these LTE capabilities, enabling even higher scalability of device density.
Critical IoT, unlike Massive IoT, implies extremely high reliability, exceedingly low latency and 99.999% (5 nines) availability. This is where 5G is really separating itself from 4G LTE.
Use cases being tested in this area include remote control of robotic machinery and automated vehicles. Labor regulations and industry standards place extremely stringent requirements on these safety-critical control systems to ensure worker and public safety. 5G networks will achieve end-to-end latencies on the order of 1 millisecond.
To achieve the reliability, latency and availability expected with these new use cases, networks must evolve to incorporate new technology coming with the 5G standard in 2020. Specifically, the 5G NR standard with Stand Alone core network.
Some of the low latency aspects will also come from new distributed architectures, where some control and processing occurs at the edge of the network. For an indoor application, this will require a local node either on premise or within a very short distance.
As discussed above, the nature of the 5G use case desired is critical to understand before attempting to define the technical requirements of the network. Each different use case has different implications for the spectrum required and the architecture required to support.
Several new frequency bands have been added to the 3GPP licensed spectrum over the years, and there is a direct correlation between spectrum and bandwidth available for services. To understand this, we can look at the history of cellular technology. Originally, low-band (< 1 GHz) bands dominated which were optimal for wide-area voice and SMS coverage. Operator 1G and 2G networks were universally deployed in these low bands which can be characterized as very broad coverage with limited bandwidth. Operators typically had 10 or 20 MHz of spectrum available in any given market with which to offer services.
With 3G and 4G, many mid-bands (1 – 6 GHz) were added in order to improve mobile broadband. These bands have shorter wavelengths and thus shorter reach, but larger bands of spectrum are available. Recently new spectrum sharing schemes including LAA and CBRS have been introduced in these middle bands. When operators can leverage these new bands, total bandwidth available for end-user services increases significantly. An operator with LTE Advanced service leveraging carrier aggregation across multiple licensed bands plus LAA can offer peak user bandwidth in excess of 1 Gbps.
With 5G, in addition to even more mid-bands, high bands (6 – 60 GHz) are being added which introduce entirely new challenges. mmWave spectrum is getting significant attention because of the extremely high bandwidth available to deploy services, leading to multigigabit speeds for users. The challenge with this high band/mmWave spectrum is that the wavelengths are extremely short and thus the radio waves do not propogate through external or even internal walls.
As indoor networks have evolved over several generations of wireless standards, a number of deployment architectures have been developed – distributed antenna systems, uncoordinated small cells and distributed radio systems.
In the era of 2G and 3G, the primary purpose of indoor networks was to extend voice and SMS coverage to large public venues like stadiums and convention centres. For this purpose, DAS was a good choice as it allowed existing base station and radio equipment to remain unchanged. Instead of connecting a high-powered radio to a single antenna on a tower, Passive DAS allowed the high transmit power of macro radios to be distributed across many low-power antenna points within a venue, typically via coaxial cables. This passive distribution of RF was cost-effective for single transmit/receive streams within a limited spectrum range.
As networks evolved to advanced 3G and 4G technology, the need for higher performance started to stress traditional DAS architectures, resulting in the emergence of the active DAS and uncoordinated small cell markets for indoor coverage. Active DAS replaces much of the coaxial RF cables with structured fiber and copper cabling. Rather than taking the RF signal directly from the radio and directly propagating it, active DAS captured the signals, transmitted them over a digital network within the building, then reconsituted the signals closer to the antenna points. Final leg of transmission still used coaxial cables to a distributed antenna array in a similar way to passive DAS. This solution solved many RF efficiency problems of passive DAS, but at increased cost and complexity for the network between the head-end base stations and the final antenna transmission points.
Uncoordinated Small Cells
3G and 4G networks also adopted uncoordinated small cells, similar to other wireless network deployments like Wi-Fi. Indoor small cells often have the advantages of low unit cost and simple installation with IT-grade structured cabling. Since each small cell is a self-contained network element, there can be limited opportunities to take advantage of advanced 4G performance features that rely on combining signals from multiple antenna points in a coordinated way. This ultimately enforces an upper limit on each small cell’s capacity and can require an extra layer of network coordination to ensure seamless mobility across the network.
Distributed Radio Systems
Most recently, the Distributed Radio architecture was introduced. In a distributed radio system, network functions are distributed between centralized units for flexibility and scalability, with technology-agnostic low power radios ensuring high performance throughout the venue. Like DAS and small cells, distribute radios ensure high RF signal dominance throughout a venue due to placing many low-power transmitters close to the users. Like uncoordinated small cells, they can often use low-cost IT-grade cabling for both signaling and power, and have low unit cost. However since the complex signal processing functions are in a centralized location, capacity can be flexibly shifted between antenna points and advanced coordination features of 4G can be utilized.
With such a variety of use cases, technology choices and deployment architectures, it can be bewildering to sort out which are the optimal arhictectural choices for introducing 5G indoors.
In an indoor environment there are unique considerations. Ericsson recommends a design and deployment approach tightly tied to the intended use cases for the indoor deployment. We take each use case category below.
Mobile Broadband use cases
Although Mobile Broadband is only one of the use case categories supported by 5G, we expect it will be the dominant requirement for initial 5G indoor deployments. In order to enable 5G mobile broadband use cases, there must be sufficient end user bandwidth. Ericsson recommends the following for indoor 5G mobile broadband deployments.
- Wide channel bandwidths (up to 100 MHz)
- Higher order MIMO (e.g. 8x8)
- Coordinated RAN with outdoor network to reduce indoor dominance requirements; indoor and outdoor networks complement each other
- High SINR with “good RF” high throughput
To achieve these network characteristics,a “deep layering” approach is recommended, with multiple mid-range (1 – 6 GHz) bands layered and multiple MIMO layers per band across all available spectrum. Layering would easily be extended into unlicensed and shared-access spectrum (such as LAA and carrier aggregation with CBRS). Innovations such as Cell Border Shifting can help minimize inter-cell interference and “dead zones” of low performance. To meet these specific requirements for 5G enhanced mobile broadband (eMBB), Distributed Radio architectures have advantages compared to traditional distributed antennas (DAS) or uncoordinated small cells (femtocell) architectures.
|Features||Features Distributed Radio||DAS||Uncoordinated Small Cells|
|Cost of scaling||Low cost scaling with IT-grade cabling||High cost scaling with higherorder MIMO||High cost to scaling|
|Capacity||Flexible expansion in centralized baseband||Densification of capacity per square meter requires expensive installation of headend equipment and splitting of equipment throughout the venue||Fixed maximum capacity per area|
|Mid-band capabilities||Low, medium and high-band availability||Limited mid-band capabilities (3.5GHz)||Limited to femto chipset|
|Flexibility of reconfiguration||Flexible capacity reconfiguration||Highly inflexible and expensive to reconfigure||Inflexible|
|Evolution to mmWave||Supported||No evolution to mmWave||Inflexible evolution, capped by femto chipset limitations|
|Additional functionality||RF-agnostic out to very edge, avoids coax limitations|
Massive IoT use cases
Massive IoT (M-IoT) will be the second dominant use case required for indoor 5G networks. Indoor M-IoT will complement outdoor M-IoT architectures. Indoor massive IoT will begin with migration of existing low bands (sub-1 GHz) to NR. These will primarily be outdoor high-power macro radios. Specific improvements in 5G standards will improve cell-edge performance over 4G, which will help outdoor-in coverage to extend. This enables M-IoT deployments with 4G technology today, with a path to even higher scalability after 5G migration. Low-band outside-in massive IoT will be complemented as needed by using the layered mid-band spectrum deployed indoors for MBB, or even with low-band indoor in extreme cases.
Advantages of Distributed Radio for Massive IoT
- Enables easy selective deployment in limited areas
- Flexible capacity assignment between bands
- Ability to share common RAN capacity with outside low band coverage
Critical IoT use cases
Critical Realtime IoT is the last of the three main use case categories to be addressed. Although many improvements have been made in LTE standards to improve end-to-end latency to approx. 5-10 ms, the critical realtime requirement of 1 ms latency can only be achieved using innovations included in 5G standards. It will take some time for the required enhancements to both the radio access and core networks to reach full commercial availability.
Radio access networks for 5G will use new RF spectrum allocations that are much higher in frequency than the low- and medium-frequency bands optimal for mobile broadband and massive IoT. The addition of mmWave bands in the 20 – 60 GHz range is a critical step in the evolution of indoor 5G. Ultra-high bandwidth availble in mmWave spectrum enables massive throughput, while wider subcarrier spacing can improve critical realtime IoT latency. However, indoors, propagation will be limited to open areas with limited penetration through interior walls. These characteristics may constrain mmWave deployment to only those areas where the benefits are most required. Examples of such indoor spaces might include manufacturing assembly lines requiring extensive coordination of robotics and automated vehicles.
5G Core network advances include network slicing, enabling computation resources for critical realtime applications to be physically located at the edge of the network, reducing network transmission delays. Edge computing will be a key enabler.
Ericsson envisions indoor 5G networks of the future to be based on a deep layering of midband spectrum, complemented by outside-in low bands for M-IoT and indoor mmWave bubbles for ultra-high capacity and realtime critical IoT. Distributed Radio architectures are best suited to meet the challenges due to their low deployment cost, flexible capacity and close coordination between bands and with surrounding RAN deployments.
Network operators, enterprises and policy makers should consider the 5G use cases that will be needed in the future to ensure that infrastructure investments are not stranded. When modernizing existing in-building networks with 4G, build a foundation of coverage and capacity that will be easily expanded to include 5G capabilities in the future.
The contributor to Ericsson’s opinion on this topic is Christopher Wallace.
Christopher Wallace is a Strategic Product Manager for Ericsson’s Advanced Industries portfolio, helping service providers and industrial enterprises unleash the full potential of wireless networking in their business operations. Innovative industry leaders are embracing the explosive potential of 5G powered by Ericsson Radio System. Chris brings a unique perspective to the industry based on experience gained over more than 20 years of engineering and business development roles covering a wide range of telecommunications technologies including optical networking, Carrier Wi-Fi, and 2G CDMA wireless. Most recently, he has been helping mobile network operators around the world transform their indoor cellular networks to meet the everincreasing mobile broadband demand with Radio Dot System, Ericsson’s industry-leading system for the highest performing 3G, 4G and 5G indoor networks.