5G Aims to be the chameleon technology for wireless

March 2016: 5G Aims to be the chameleon technology for wireless

Executive Summary

The new “hype-child” of the mobile industry is 5G, with technology vendors, mobile operators and even Nation States, including the Japan, South Korea and the United States, vying for a leadership role in bringing it to market – even before it is standardized in the 2020 time-frame. In contrast to 4G, which is focused primarily on enabling increased bandwidth, 5G aims to be a “chameleon” technology, which anticipates the diverse demands of wireless services, including support for high bandwidth, low latency, bursty traffic, ultra-reliable services, or a combination of these. 5G also capitalizes on NFV and SDN standardization efforts, with the objective of enabling efficient network resource slicing, with core networking standards that are agnostic to the radio technology that is used.

There are numerous use-cases that are being applied to 5G, and many more are likely to emerge before standards based products are developed. Since 5G is a transformative technology, it is prone to missteps and errors in its development, standardization and deployment. To avoid these missteps, we believe that the industry must pay careful attention to the market dynamics for the various use-cases targeted by 5G. This report investigates five use-cases that are proposed for 5G, including narrow-band sensor networks, public safety networks, the Tactile Internet, autonomous vehicles, and high resolution media applications. Each of these use-cases are assessed relative to the 5G attributes they intend to leverage, which includes ultra-low device energy consumption, device complexity and connection latency, and ultra-high device density, data rates, network capacity, coverage and reliability, and ultra-high mobility and security, see Exhibit 1.

Based on the assessment of the use-cases the following observations are made:

  • Prior to 5G availability, narrow-band sensor networks will benefit from 4G innovations, namely LTE-M1 and M2. Since these innovations provide adequate performance gains for many of the sensor networks deployed today, we believe that 5G will be targeted towards new service categories that have more stringent requirements, particularly in terms of the cost and complexity of the connectivity modules. Notable examples include wearables, and low cost disposable solutions. The current 5G proposals for achieving the necessary functionality for narrow-band IoT solutions (including sensor networks) are varied, some use techniques similar to those associated with unlicensed LPWA solutions proposed by companies like SigFox and Semtech (LoRa).
  • 5G proposals incorporate priority access functionality directly into its radio modulation scheme for mission critical applications, such as public safety. However, we believe that it will take many years after 5G is initially deployed before its network coverage is sufficient to reliably support public safety applications. In the interim, a great deal of effort is being directed to 4G innovations that are targeted towards public safety and other mission critical applications.
  • The Tactile Internet is a transformative solution that occurs at the intersection of IoT, cloud services, augmented reality and artificial intelligence. In addition to the audio and visual functionality that is provided by the conventional Internet today, the Tactile Internet essentially incorporates information to encode physical interaction (i.e. tactile information). To function effectively, the Tactile Internet cannot tolerate connection latencies in excess of 1ms and has high bandwidth requirements. While the Tactile Internet is a compelling use-case for 5G, we believe that it will take many years after the inception of 5G for the Tactile Internet to gain meaningful market momentum.
  • Autonomous vehicle technology has benefited from tremendous innovation in recent years, with automotive manufacturers and companies like Google investing significant R&D resources to its development. While industry experts have diverse opinions relative to the timing for mass market autonomous vehicle adoption, we believe that this is likely to occur after the 2025 time-frame, with “Limited Self Driving Automation” (Level 3) being implemented. In the context of 5G, we believe that when Level 3 automation is implemented, there will be demand for low latency vehicle-to-vehicle connectivity to enable vehicle platooning.
  • We expect that high resolution multimedia will be enabled in 5G using millimeter wave technology, with advanced antenna technologies to enable non-line-of-sight (NLOS) connectivity over the coverage range of a typical small-cell. While there are a variety of technical, regulatory and operational hurdles that need to be overcome before millimeter wave can be used as an access technology, we believe that it will drive the early use cases for 5G.

Between now and 2020, when 5G standards are expected, the wireless mobile industry dynamics are likely to change dramatically and many new use-cases emerge. As this occurs, it is crucial that the industry remains nimble and pragmatic in its approach towards aligning 5G priorities with market demands.

Exhibit 1: Aligning 5G Use Cases with its Functional Attributes
Source: Tolaga Research 2016


Since its inception, the mobile industry has feverishly innovated to extract increased bandwidth efficiency, capacity, data speeds and service advancements from it network technologies. This has culminated in second, third, and fourth generation (a.k.a. 2G, 3G and 4G-LTE) technologies, which each have had their intended role in advancing mobile services. 2G delivered standardized digital technologies, including GSM, TDMA and CDMA that enabled mobile telephony services to proliferate. Buoyed by the Internet bubble, 3G was intended to herald the mobile Internet, but largely fell short of this objective and has subsequently been eclipsed by 4G-LTE. 5G is coming hot on the heals of 4G, with a broader scope to enable a unified network architecture that aims to support the various wireless connectivity demands for emerging mobile services and applications.

In essence, 5G aims to be a “chameleon” technology which can adapt to the differing demands of wireless services, whether to support high bandwidth, low latency, bursty traffic, ultra reliable services, or a combination of these. 5G capitalizes on standardization efforts with Network Function Virtualization (NFV) and Software Defined Networking (SDN) to enable unified architectures that support the dynamic allocation of network resources for specific services and applications (aka network slicing), see Exhibit 2. The stakes are high for 5G, with technology companies vying for standardized solutions that leverage their Intellectual Property and know-how. Nation States and network operators, particularly those in the United States, Korea and Japan, are competing aggressively for first-to-market status. These players are proposing to deploy pre-standard 5G solutions prior to 2020, when 5G standardization is anticipated.

Exhibit 2: Proposed 5G Functionality
Source: Tolaga Research 2016

Since 5G is a transformative technology, it is prone to missteps and errors in its development, standardization and deployment. It is also prone to an all too familiar hype cycle that under-estimates the complexities and time-frames associated with industry transformations. The mobile industry is littered with casualties of past transformations, such as mobile satellite, WiMAX, and Rabbit-CT2, which each provide useful lessons when assessing opportunities for 5G. Since 5G is intended to create a unified wireless platform, its justification is supported by a plethora of applications and use-cases, some which are already supported by 4G and others which are futuristic with complex dependencies and have yet to be supported by compelling business cases. Notable examples include:

  • Ultra-dense narrow-band sensor networks with ultra-low cost and energy efficient devices. For 5G to have a meaningful impact depends on its capabilities relative to solutions, such as LTE-M1 and LTE-M2, and unlicensed technologies including SigFox, LoRa and Ingenu are being targeted towards comparable applications.
  • The Tactile Internet requires the low latency connections (less than 1msec) to function effectively. Since the connection latencies for 4G-LTE are in excess of 10msec, the latency performance enhancements being proposed for 5G are needed for the Tactile Internet. Furthermore, the Tactile Internet is dependent on the maturation of the Internet-of-Things) IoT and virtual and augmented reality technologies, and while it has tremendous potential, it is still the domain of research laboratories.
  • For every mile driven for autonomous vehicle trials and implementation, the automotive industry inches closer to having the technology become mainstream. In addition to standalone implementations, researchers are developing technologies for “vehicle platooning”, for which vehicles travel at high speeds and in close proximity. These platoons require vehicle-to- vehicle connections that have similar latencies to the Tactile Internet.
  • Mission critical applications, such as those associated with public safety are commonly deployed on dedicated networks to ensure service reliability. 5G proposals incorporate waveforms and functionality to enable mission critical services to rapidly procure channel resources as required, from within shared network environments.
  • 5G promises to enable ultra-broadband capabilities using radio solutions that operate at 10’s of GHz, such as millimeter wave (mmWave). Successful implementations of mmWave solutions depend on many factors including advancements in radio and signal processing technologies, radio planning and design solutions and use-cases that the technology can reliably support. While technology vendors like Ericsson, Huawei, Nokia and Qualcomm are demonstrating mmWave implementations, these implementations fall well short of what is needed for commercial solutions.

This report investigates that salient characteristics of a variety of services and applications for which 5G is being targeted, including narrow-band sensor networks, public safety applications, the Tactile Internet, autonomous vehicles and high resolution media applications. It assesses the market dynamics for these services and applications and their impact on 5G demand.

Synthesizing 5G Use Cases

While the candidate use-cases being proposed for 5G are vast, the 5G design criteria can be defined in terms of a handful of criteria, including ultra-low connection latency, device energy consumption, and device complexity, and ultra-high device density, network capacity, mobility, coverage, reliability, and security.

In Exhibit 1 above, several notable use-cases are investigated in the context of the salient 5G attributes and compared to the capabilities of 4G. Based on these use-cases, some of the key 5G functionality, such as ultra-low device energy consumption and complexity, might already be adequately addressed with 4G. Furthermore, the market timing and demand will vary dramatically amongst the key 5G functions, depending on the adoption cycles of the uses-cases that depend on 5G. This is typical of a transformative technology like 5G, and illustrates the need for industry players to remain pragmatic and nimble with the 5G initiatives.

A fundamental tenet of 5G, which aligns with mobile industry transformation efforts, is to enable service agility and anticipate a shift from supply, to demand driven business models that aim to address growing customer expectations. A notable example is network service slicing, which allows for the efficient provisioning and granular allocation of network resources that are optimized for the service demands of specific use-cases. 5G is relying on NFV and SDN developments to achieve network agility, and aims to create a unified core that is radio technology agnostic. While this approach is compelling, we believe that designers run this risk of prioritizing their efforts without adequately considering end-user demands. For example, instead of implementing an end-to-end 5G solution, an operator might focus on deploying a 5G core to unify services across its legacy radio technology portfolio, including both licensed and unlicensed radio spectrum. To illustrate this point, when the iPhone was originally launched it was a wild success even though it initially operated on AT&T’s 2G network, and subsequently relied heavily on Wi-Fi network capacity. This contrasted the mobile industry mantra of the time, which had 3G as driving force for the mobile Internet.

In the past mobile operators have benefited from having the latest network technology to market to their customers – even if it lacks a compelling technical justification at the time. To capitalize new network technologies, mobile operators generally rely on relatively scant deployments until there is commercial justification for increased network investments. This is particularly true for 5G, since it is intended to complement, rather than replace, 4G for specific use-cases that can benefit from a unified and agile 5G platform. The rate at which mobile operators invest in 5G deployments has a direct impact on the feasible deployment cycles for the sample use-cases that are summarized above in Exhibit 1. In particular, when 5G network deployments are nascent, we believe that mobile operators are likely to focus on high resolution media services and subsequently targeted IoT implementations. Since V2V communications for autonomous vehicles can function with ultra-low latency 5G connectivity without necessarily requiring 5G network coverage, we believe that it is less dependent on mobile operator deployment strategies, but more so on the maturity of 5G and autonomous vehicle technologies. We expect that the Tactile Internet and Public Safety applications will have protracted time-lines relative to high resolution media service, IoT and V2V connectivity for autonomous vehicles, see Exhibit 3.

Exhibit 3: Anticipating 5G Use-Cases on Radio and Device Functionality Demands
Source: Tolaga Research 2016

Millimeter Wave Technology Fuels High Resolution Media

Virtually since the inception of mobile data services, traffic demands have exceeded expectations and been truncated primarily by service provider pricing models and service availability. 5G is proposing to dramatically increase capacity using ultra-high frequency technologies, such as millimeter wave. Wireless operations at these frequencies are challenging. Most notably, the radio design and propagation characteristics of millimeter waves (30-300GHz) are vastly different to conventional wireless mobile communications which in sub-3GHz spectrum bands.

As operating frequencies increase, so to does the effective rate of signal energy dispersion in a free space environment. Environmental objects in the path between the transmitter and receiver disrupt radio propagation. These disruptions vary depending on the physical size of the objects relative to the wave-length of the radio signal. In cluttered environments (such as urban centers) sub-3GHz waves are diffracted and reflected by large objects such as buildings and are essentially unaffected by water vapor and oxygen in the air. This contrasts millimeter waves, which are effected by water vapor and oxygen, have extremely high diffraction losses and generally cannot pass through walls. Instead the walls act as diffuse reflective surfaces for millimeter waves, and it is the reflected as opposed to diffracted signals that enable non-line-of-sight (NLOS) connectivity at millimeter wave frequencies. For background details on radio propagation please refer to our online Tutorial.

In the past millimeter wave technologies have been targeted to niche applications, largely as a consequence of design challenges. Implementations have been limited to transmission and backhaul solutions, using standardized technologies like LMDS (Local Multi-point Distribution Service), or proprietary solutions offered by companies like Aviat, Bridgewave, eBand, and Silku. However 5G leverages antenna technology advancements, which capitalize on the small antenna form factors for millimeter wave frequencies and sophisticated signal processing to enable improved signal reception with local mobility. For example, a millimeter wave system operating in the vicinity of 30GHz has an approximate wavelength of 10 millimeters. The size of the antenna elements are a multiple of the system wavelength. For example, a half-wave dipole for a 30GHz system would be 5 millimeters long. Antenna arrays are created by placing the antenna elements within one or two wavelengths of each-other, enabling large arrays in relatively small form factor configurations. The antenna arrays in combination with advanced signal processing can be used to enable massive MIMO diversity and real-time beam-forming/steering techniques so that the direction of the transmitting and receiving antennas can be optimized for signal reception.

Researchers, including Ted Rappaport’s team at New York University (NYU) have demonstrated that when beam-steering is used, a millimeter wave transmitter with 1W transmission power can achieve NLOS coverage of up to 200 meters. NYU has also open sourced a channel simulator based on its measurement results. While there is the need to refine and commercialize the advanced antenna technologies being used for millimeter wave, we believe that it is well positioned as a small-cell mobile access solution with tremendous capacity for delivering high resolution multi-media services. In the interim, the mobile industry must address the challenges in enabling large scale small-cell deployments, which is the topic of an upcoming Tolaga Report.

When licensed and unlicensed millimeter wave spectrum was originally allocated, regulators did not contemplate the notion of it being used for small-cell mobile access. As a consequence regulators must revisit their policies and objectives for millimeter wave spectrum to support 5G. Unfortunately millimeter wave discussions were not included in the ITU’s 2015 World Radiocommunication Conference (WRC15), which is conducted every four years and traditionally responsible for harmonizing global spectrum allocations. However, regulators in many countries, including the FCC in the United States have initiated consultations to advance the regulations for both licensed and unlicensed millimeter wave spectrum for 5G use.

Mobile operators in many markets hold LMDS licenses (28 and 39GHz), and others have been acquiring players with spectrum assets, such as Verizon’s announcement in February 2016 to acquire XO Communications. While we believe that millimeter wave licenses will become more valuable with 5G, we also anticipate that operators will leverage the extensive unlicensed millimeter wave spectrum resources that are already available, (e.g. 59-64GHz in the United States). In addition, some bands and allocations will offer more value than others. For example, there are specific frequencies that are less desirable than others due to water and oxygen absorption, see Exhibit 4. Furthermore, some researchers are suggesting that even higher frequencies into the Tera-Herz range might be candidates in the future, particularly if it can enable greater densities of antenna elements to capitalize on the directionality of the propagation channel at these frequencies.

Exhibit 4: Impact of Water and Oxygen Resonance on Millimeter Wave Attenuation
Source: Tolaga Research 2016

While millimeter wave technologies offer tremendous potential for 5G implementations, a great deal of research is needed to leverage its full capabilities. Currently many of these efforts are focused on developing advanced antenna technologies, radio channel models and system designs for millimeter wave implementations to enable ultra-broadband access in limited mobility environments. We expect that millimeter wave technologies will be a critical element of early 5G deployments, as mobile operators seek the ultra-broadband capabilities that it achieves. We also believe that efforts to commercialize 5G millimeter wave will benefit greatly from parallel efforts to advance millimeter wave systems that operate in unlicensed spectrum, such as 802.11ad, which has already been adopted by companies like Dell and HP for computing device docking stations, and incorporated in Qualcomm’s SnapDragon 820 System on a Chip (SoC).

The 5G Chameleon Changes its Design Priorities for Narrow-Band Sensor Networks

Sensor networks underpin many of the services associated with the Internet-of-Things (IoT) and while the lion’s share of sensor networks use local area network technologies, we estimate that there were 386 million wireless wide area network (WWAN) connected IoT devices in 2015, which will increase to 4.63 billion by 2025.

Exhibit 5: Forecast for WWAN based IoT Connections
Source: Tolaga Research 2016

Today most WWAN IoT connected devices use 2G-GSM technology and contrary to the rhetoric, we do not believe that IoT (aka M2M) has been particularly strategic for most mobile operators in the past. However as this has changed in recent years, resulting in 3GPP standardization efforts for LTE-M1 and LTE-M2 and enhanced features for GSM that are specifically designed for IoT.

Today most WWAN IoT connected devices use 2G-GSM technology and contrary to the rhetoric, we do not believe that IoT (aka M2M) has been particularly strategic for most mobile operators in the past. However, as this has changed in recent years, resulting in 3GPP standardization efforts for LTE-M1 and LTE-M2 and enhanced features for GSM that are specifically designed for IoT.

Narrow-band sensor networks typically consist of low cost sensors that are deployed over large geographical areas. These sensors typically generate small payloads of bursty and intermittent data traffic. In the context of Exhibit 1 above, narrow-band sensor networks require ultra-low complexity (cost) devices with ultra-low energy consumption, and networks that support ultra-dense device deployments, with robust radio coverage. While 5G aims to hone its capabilities to address these demands, we believe that current and emerging 4G technologies, such as LTE-M1 and M2 and other low power wireless access (LPWA) solutions that operate in unlicensed spectrum address many of the current market demands.

The salient characteristics of LTE-M1 and M2, and LPWA solutions are discussed in a recently published Tolaga Report entitled,Navigating the IoT Wireless WAN Connectivity Labyrinth. In this report it is observed that the modem complexity for LTE-M1 and M2 is 20 and 15 percent less than that of conventional LTE-Category 4 technology, respectively. At scale, we expect LTE-M1 communication modules will cost $ 8-12 USD, and LTE-M2 to cost $6-10 USD. Both technologies have reduced energy consumption with battery lives in excess of ten years. To achieve energy savings, the LTE standards enable devices to have reduced signaling activity and aggressively utilize power saving modes. These same capabilities also enable LTE networks to efficiently support a significant increase in device connections and connection density, which are expected to accompany IoT implementations.

Competitive Low Power Wireless Access (LPWA) solutions that operate in unlicensed spectrum with designs that are optimized for IoT, have been developed by a variety of companies including Ingenu, M2COMM, nWave, SigFox, and Semtech (LoRa). These solutions introduce a variety of innovations, such as those to significantly reduce device complexity, particularly in cases where ultra-narrow-band communications are used. Proposals that are being considered for 5G leverage similar innovations, and in some cases might even leverage the Intellectual Property that has been developed for the unlicensed spectrum technologies. As an example, conventional 4G-LTE depends on its connections being synchronized and orthogonal to efficiently support high traffic demands. This contrasts unlicensed solutions which typically have asynchronous operations and in some cases support as little as 100bps connection speeds.

Proposals for 5G aim to improve on the capabilities offered by 4G LTE-M1 and M2 technologies. Since we believe that these improvements will be evolutionary, rather than revolutionary, we anticipate that the 5G IoT capabilities will be developed to complement 4G with solutions that target lower cost sensor technologies, such as wearable devices and those that are attached to disposable products. For 5G to succeed in this category of IoT, new service models are needed to dramatically reduce or eliminate monthly subscription costs. It is unclear whether mobile operators can successfully implement this type of service model.

5G Embeds Reliability for Mission Critical Applications

Mission critical applications, such as those associated with emergency services and public safety, have traditionally been deployed with dedicated networks to avoid reliability issues, attributable to the user contention and coverage limitations associated with commercial wireless networks. Many of the dedicated networks for mission critical applications are outdated and are primarily limited to voice services. In the United States, the Federal Government established the First Responder Network (FirstNet), and availed it with $7 billion in funding and 20MHz of radio spectrum to deploy a dedicated LTE network for public safety broad-band services. The network deployment commenced in 2013.

As an alternative to dedicated public safety networks such as FirstNet, 5G proposals incorporate functionality to facilitate real-time priority access. Rather than continually maintaining dedicated channel resources for public safety services, it has been proposed that when dedicated channel resources are required, a signal burst is used to eliminate any active communications in the channels that are needed. The channels are then released for general usage after the priority communications cease. In addition to providing reliable service access, mission critical services require ultra-high security, network coverage, mobility and security, and ultra-low connection latency. Efforts are being made to incorporate these capabilities into 5G, however they are generally predicated on 5G being deployed with adequate network density in the areas where the mission critical applications operate. Furthermore, even if 5G is capable of delivering the requirements for mission critical applications, we believe that it will take many years to be accepted for this purpose.

The Tactile Internet Enables Star Trek Like Experiences

The Tactile Internet anticipates the tremendous transformation that occurs at the intersection of IoT, cloud services, augmented reality and artificial intelligence. In addition to the audio and visual functionality that is provided by the conventional Internet today, the Tactile Internet essentially incorporates information to encode physical interaction (i.e. tactile information). To function, the Tactile Internet depends on extremely high bandwidth and low latency (less than 1 millisecond) connections, and IoT end-point devices, such as actuators to disseminate the tactile information through machine-to-human interfaces. Achieving connection latencies in the order of 1 millisecond is challenging, but necessary for tactile applications. For example, if an augmented reality application involved the simple act of catching a ball traveling at 10 m/s, the ball will travel 1cm for every millisecond. A latency of 50ms in a typical LTE network would effectively create a 50 cm error between the visual and tactile cues. Users of services and applications that have a mismatch in excess of one millisecond between visual and tactile cues commonly suffer from nausea that is similar to motion sickness.

Researchers are targeting 5G as an enabling technology for the Tactile Internet. However it is dependent on a variety of technology break-throughs in network and service performance, the maturation of cloud, IoT and augmented reality solutions. It also depends on transformed business models to suit. We expect that these inter-dependencies are likely to create protracted time-lines in excess of a decade or more for the Tactile Internet to have a meaningful impact on 5G demand.

5G Enables “Platooning” for Autonomous Vehicles

Autonomous vehicles and intelligent vehicle systems have benefited from tremendous innovation in recent years, with all major vehicle manufacturers and other companies like Google investing heavily in research and development (R&D). While the ultimate objective of these R&D efforts is to enable driver-less cars, there are tremendous challenges in automating the human-to-machine interfaces involved. Forecasts for the availability of autonomous vehicles vary greatly amongst manufacturers and solution providers. In particular, companies like Telsa and Google predict availability in the 2018 time-frame, and other players like Ford, General Motors, and BMW more conservatively predict availability in the 2025 time-frame. We believe that this divergence in opinion is partly attributable to the definition of “autonomous”. These definitions are important, and in the context of this report provide insights into the functionality that is likely to require 5G connectivity. For example, the National Highway Traffic Safety Administration (NHTSA) in the United States defines vehicle automation as having five levels, which include:

  • No Automation (Level 0)
  • Function Specific Automation (Level 1): For which automation involves one or more specific control function, such as electronic stability control or pre-charged braking systems
  • Combined Function Automation (Level 2): For which at least two primary control functions are automated to work in unison, such as adaptive cruise control in combination with lane centering.
  • Limited Self-Driving Automation (Level 3): Where automation enables the driver to cede full control of all safety-critical functions under certain traffic or environmental conditions. The Level 3 automation system includes capabilities to bring the vehicle back to driver control in cases dictated by environmental conditions. Google’s self-driving car is a notable example of Level 3 automation.
  • Full Self-Driving Automation (Level 4): Where vehicles are designed to perform all safety-critical driving functions and monitor roadway conditions for the entire trip – for both occupied and unoccupied vehicles.

Although Level 3 autonomous vehicles are available today, the technology is expensive and regulations for its use is complicated. Most industry participants anticipate market availability in the 2025 time-frame. Initially autonomous vehicles will be deployed with standalone implementations, with the aim of replacing the human-to-machine interfaces for driving. In addition, Level 3 autonomy enables other capabilities that exceed that of the human driver. Most notably, it offers greatly improved response times, eliminates visual blind-spots, and can be integrated with information from other vehicles. In the context of 5G, researchers have been developing low latency and high performance vehicle-to-vehicle communications (V2V), to leverage Level 3 autonomy to create vehicle platoons. Trials have already been conducted for several years, and platooning is expected to follow a similar time-line to other Level 3 autonomous vehicle capabilities. These platoons have the vehicles traveling in close proximity and potentially at high speeds. Platooning has been trialed extensively for the trucking industry and creates opportunities for special vehicle lanes on highways and in congested urban environments. For these platoons to function effectively, V2V connections require latencies in the order of 1ms. Furthermore, when platoons have more than three vehicles, they become unstable unless there are meshed communication links connecting each vehicle to every other vehicle. As a consequence the number of wireless connections amongst platoon vehicles increase exponentially with the number of vehicles involved.

The long term prospects for autonomous vehicles are promising, being buoyed by strong interest from automotive manufacturers and technology providers and policy makers who are interested in addressing traffic congestion and pollution. However, since the autonomous vehicle technologies are are expensive and need to be accepted by vehicle owners, regulations are complicated, and vehicle replacement cycles are relatively slow, the time-lines for the consumer mass market adoption of autonomous vehicles are protracted and likely to be after 2025.

Finding the 5G Future

5G has a relatively protracted time-line that coincides with tremendous disruption, as the mobile industry transforms to cater for the digital economy. As this transformation occurs, 5G is likely to be confronted with diverse service demands and as a result it is essentially an umbrella standard for a variety of parallel technology initiatives, aiming to cater for ultra narrow-band and efficient IoT services at the same time as ultra-broadband and mission critical applications. While there are compelling use-cases to justify the various design objectives for 5G, we believe that there is the risk of becoming disconnected with market demands. We also believe that the mobile industry runs the risk of being lured into a false sense of security by the rapid rise of 4G, failing to realize the important role that other technologies like the iPhone, 3G and WiMAX played in incubating the mobile Internet.

Between today and 2020 or 2025, it is likely that use-cases for 5G that have not yet been conceived will emerge, however the mobile industry has been effective in identifying the salient design requirements to address the demands for those use-cases that have been identified. This report assesses several use-cases for 5G, including narrow-band sensor networks, public safety applications, the Tactile Internet, autonomous vehicles, and high resolution media. Based on these assessments, the following conclusions can be made:

  • The 3GPP has made tremendous progress with 4G standards to support IoT applications including narrow-band sensor networks. Historically these networks have long equipment replacement cycles, particularly if the equipment is adequately servicing its purpose. Proposals that have been touted for 5G aim on reducing the cost and complexity of IoT connectivity modules relative to 4G, in some cases utilizing similar functionality that is used for unlicensed LPWA solutions. On this basis, we anticipate that 5G will be well suited to ultra-low cost IoT solutions, such as wearables and disposables.
  • Public safety has traditionally been poorly served by wireless solutions, largely because they have relied upon dedicated network solutions to avoid network contention and availability issues. Recent network deployment initiatives, such as FirstNet in the United States, have been launched with the intention of modernizing public safety services, albeit with a dedicated network. Since 5G proposes to incorporate priority access capabilities directly into its radio waveform, it is well positioned to support public safety services. However, since this also depends on 5G being widely deployed to deliver adequate coverage, we do not believe that public safety service demands will incubate 5G deployments.
  • While the Tactile Internet has been researched for several years and holds tremendous promise, it will take many years to develop and is dependent on a variety of factors, most notably the maturation of IoT, transformed business models for key industry participants and advancements in adjacent fields, such as Augmented Reality. On this basis, we believe that the Tactile Internet will take more than a decade to development, with many of the initial use-cases leveraging local as opposed to wide area network technologies. When deployed over wide area networks, the Tactile Internet will require on latencies in the order of 1ms, enabled by 5G and benefit from its ability to support massive densities of interactive sensor devices.
  • With players like Google having already developed Level 3 autonomous vehicles with limited self-driving automation, we anticipate that autonomous vehicle availability will align with that of 5G. When initially rolled out, we expect that autonomous vehicles will initially be standalone, with limited requirements for 5G connectivity. However once adequate autonomous vehicle densities are achieve, techniques such as platooning will emerge, with the aim of reducing vehicle transit times, reducing accidents and improving traffic flows. To function effectively these platoons require low latency V2V communications (in the order of 1ms) for which 5G is well suited.
  • While it is important for designers to ensure that 5G supports the various use-case demands for 5G, we anticipate that demand for bandwidth will continue to increase in unison with demand for high resolution multimedia services. To achieve these bandwidth demands, we expect that millimeter wave technologies will mature and serve as low mobility access technologies. While there are many technical challenges that need to be overcome, the value proposition and associated business models for mobile operators are well understood and not dependent on complex value chain developments and business model transformations to succeed.

As is the case with any new and disruptive technology, 5G is prone to the turbulence of hype-cycles which will skew it perceived market value. We believe that this will be exacerbated if the industry fails to prioritize the develop and deployment of 5G to align with market timing with pragmatic business models.