5G 2015⇒2019 – Incremental Evolutionary Development and Standardization of 5G Technologies for the New Era of Wireless Internet of the 2020’s and 2030’s: An Outlook as of January 1, 2015
- Extending the cloud-centric model to a multitude of resources which are distributed close to the users with research concepts like fog computing and edge-centric approaches
- Millimeter wave radio, shared spectrum, small cells and beam-forming for Enhanced Mobile Broadband (Enhanced MBB) with multi-Gbps data rates and ms latency
UPDATE, MWC 2017: The 1st remotely controlled car demo by Ericsson and Telefonica
- Much higher levels of mobility, reliability, scalability, coverage, energy efficiency in networks, and reducing power demands in devices by striving to achieve multi-year life-spans without recharging which is particularly critical for sensors
- Highly advanced use cases such as traffic safety, self-driving vehicles, mission critical remote surgery, robotics, drones and network operators to drive the advancement of Internet of Things (IoT) and wearables markets during the period
- Players from academia and industry joining in multi-year pilot development alliances around lead companies such as Ericsson and Qualcomm
- Absolutely new industry players joining the core network equipment scene such as Foxconn Electronics (Hon Hai Precision Industry)
Ericsson’s perspective on core network technologies:
Qualcomm’s perspective on core network technologies:
The details are presented in the following sections:
- State of the research
- Current state and market forecasts
For the previous stage of 3G to 4G LTE evolutionary development and standardisation see my Mobile Internet (Aug’11) analysis.
<Links to the Year-end Update Posts will come here>
Intel’s perspective has been presented elsewhere as:
October 29, 2013: A Look at Tomorrow’s Tomorrow: We Will Live Inside Computers – Brian David Johnson, the resident “futurist” at Intel talking at The Feast social innovation conference in NYC
- As we approach the year 2020 the size of meaningful computational power, the size of the chip that goes into our computers, devices and our gadgets — the size of that approaches zero. … Five years ago, the average computer chip was approximately 22 nanometers across. Now, it is 14 nanometers across on average. By 2020, the computer chip will be 5 nanometers across, projects Johnson. A chip that is 5 nanometers wide is 12 atoms across.
|See: “By 2020 Compute Moves to Zero” leading to the question “Could we turn everything into a computer?” in the below much longer talks
– Intel’s Futurist Brian David Johnson Looks Out To 2020-2025 [33d Square, Dec 4, 2012]
– Intel futurist Brian David Johnson focuses on humanity [OSU College of Business blog, Dec 12, 2014]
– Dean’s Distinguished Lecture – Brian David Johnson, Futurist at Intel [Oregon State University College of Business YouTube channel, Nov 26, 2014]
- We will be surrounded by computational power. So for average people, for average users and consumers, you will be living in a world where you are essentially living in a computer, where you are surrounded by intelligence.
- On the micro side, computers will be inserted in all manner of health-care related devices. … We can actually monitor our health on an hour-by-hour, minute-by-minute basis.
- From a macro perspective, Johnson says that the computational powers available a decade from now will allow us to build more sophisticated, nuanced and larger cities – what Johnson calls megacities. … We can actually put computation in the infrastructure of the city so that we can make the city not only greener but more efficient. … Our cities could become computers and when our cities become computers we can tune them.
December 5, 2014: New Intel® Quark™ SoCs to enhance the Intel Gateway Solutions for IoT (“Moon Island”) portfolio in a year from now article on this same blog to show the chip roadmap for both the IoT and the wearables
December 10, 2014: Wind River: Intel’s core technology provider for IoT article on this same blog to present the newly introduced Intel IoT Platform
December 11, 2014 – January 3, 2015: 2014: A bleeding edge effort in chips (SoC), COM and software technology (inc. Wind River) for already marketable wearables by Intel has enabled its overall 2015 strategy in IoT and allows to expand its wearable ecosystem with an ‘art of the possible in technology’ article on this same blog to present Intel’s much more advanced state of its 5G effort (albeit not named as such) than either Ericsson’s or Qualcomm’s
Note: Ericsson—however—has already launched an end-to-end IoT effort on November 3, 2014 in its services space. See also the complete placement of that into the whole service portfolio at http://www.ericsson.com/spotlight/services/#internet-of-things. They were also soliciting a book From Machine-to-Machine to the Internet of Things: Introduction to a New Age of Intelligence (Academic Press, April 30, 2014) by 3 authors of their own and 2 from Imperial College in London, and 1 from SAP. Their M2M effort was first declared to move to an IoT one in 2012 by publication of their Internet of Things – Propelling the Networked Society brochure.
1. State of the research
December 18, 2014 – The road to 5G: Charting the next generation of connectivity by Matt Grob, Executive Vice President, Qualcomm Technologies, Inc. and Chief Technology Officer AND Erik Ekudden, Vice President, Technology Strategies at Ericsson
In a North American first: 50 professors, engineers and researchers met November 12 to 14 in Monterey, California to discuss and share their work on topics related to 5G. They took deep dives into new use cases, system-level orchestration and advances in radio technology needed to deliver the promise of 5G by 2020. The conference was sponsored by Qualcomm and Ericsson. In the videos linked below, five of the attendees [in addition to co-authors of this briefing, Ericsson’s Ekudden and Qualcomm’s Grob, the other three are from: Ovum, University of California San Diego and the New York University] discuss the status of 5G research and the progress they are making in their work.
One of the main themes of the presentations in the conference was that the future 5G network will provide an integrated, distributed platform for computing, storage, network resources as well as connectivity. One of the main improvements that this platform will provide will be much lower latency, even to the point of millisecond latency, and an improved cost and energy efficiency. Use cases like autonomous vehicles, remote medicine can all benefit from lower latency and new levels of reliability.
In delivering this distributed architecture, research concepts like fog computing and edge-centric approaches were also discussed. Presenters also pointed out the need for new, distributed ways of connecting everything, such as device-to-device and multi-hop communications. It is no longer enough to rely on a cloud-centric model and there must be a multitude of resources distributed close to the users. The emergence of 5G will enable a new class of applications and services that can take advantage of this distributed platform.
In the radio domain the theme was “do not be afraid of the future, we can shape it together.” The discussion of spectrum, millimeter wave radio, shared spectrum, small cells and beam-forming all indicated that researchers were ready to address the challenges of uniform user experiences and exponential increases in network capacity. Several of the presentations showed that progress has already been made in the areas of gigabit transmission over the air. The future challenges revolve around mobility, reliability, scalability, coverage, energy efficiency in networks, and reducing power demands in devices. The decrease in power will be particularly critical for sensors in the 5G network that may require multi-year life-spans without recharging.
There were many opportunities for cross-domain collaboration among researchers when the discussion turned to broad use cases such as traffic safety, self-driving vehicles, robotics, drones and network operators. The panels often mixed industry and academia, users and network operators. The varied people at the conference reflected the technology as 5G will need to connect everything across an extreme variation of types of services and use cases—from entertainment to emergency services. All of this variation can be unified by design: spectrum, services, and deployment models.
The event showed that 5G research in North America is increasing in priority across the public and private sector and distributed across a wide variety institutions and companies. It also showed that these researchers are prepared to collaborate and address the biggest challenges to making 5G a reality. Many of the people attending left with new connections, perspectives and inspirations in pushing for the boundaries of 5G research.
Jan 7, 2015 – Qualcomm CTO, Matt Grob, explains his view of what 5G is at CES 2015 and talks about the role that Qualcomm is playing in setting the standards.
5G is not just a new generation, but a new kind of network:
- Enhanced Mobile Broadband ⇒ Enhancing the foundation
- Massive number of things ⇒ Connect everything
- Proximal and aware services ⇒ New ways of interacting,
- Mission-critical services ⇒ Failure is not an option
So the requirements for 5G should be defined according to the following criteria:
- 5G is scalable and adaptable across an extreme variation of use cases
- 5G is user-centric to bring connectivity, computing and content close to the user
- 5G is a unified 5G design across spectrum types and bands for the next decade and beyond
- 5G is a unified platform for expanded connectivity needs for the next decade and beyond
- 5G is providing simultaneous connectivity to leverage 4G investments
Qualcomm in addition is emphasizing:
- User-centric Connectivity for providing edgeless connectivity (i.e. devices are no longer end-points, they are integral parts of the network + access nodes, but also devices, can act as relays and hubs etc.)
- Significant improvements over 4G
- Drive 4G to its full potential (i.e. continue to evolve 4G LTE in parallel with the advent of 5G)
December 11, 2014 – #GLOBECOM: Keynote—Future of Wireless by Rajesh Pankaj, Senior VP Engineering, Qualcomm Technologies: He is discussing the future of Wireless and the latest technology advancements Qualcomm is working on, including 5G, LTE-unlicensed and small cells, in an effort to solve the 1000x mobile data challenge and improve network efficiencies and the user experience. Qualcomm aims to be a leading force in bringing 5G to life. Rajesh will speak on the latest developments with 5G and how it is about enabling new services and devices, connecting new industries, and empowering new user experiences. He will delve into how broader dimensions of improvements will drive new 5G services and how a unified 5G design is scalable and adaptable across extreme variations of use cases.
⇒ Technology Invention Site – Envisioning the Future with 5G [Qualcomm, Nov 26, 2014]
⇒ Whitepaper sponsored by Qualcomm – LTE Broadcast [IDC, September, 2014] in which the following is written: “LTE Broadcast holds the promise of a strong link connecting the current 4G /LTE networks and the 5G networks of the future”
⇒ Blog post – Standing on the edge of a new frontier for LTE Advanced [Qualcomm’s OnQ Blog, August 20, 2014]
⇒ From Research Project Incubator – Up and coming research projects [Qualcomm, June 3, 2014]
The world is evolving quickly, with increases in mobile devices, more connected things, and applications demanding new experiences and forms of communication. It is therefore, important that we start planning how we can scale and support billions of connected things and enable instant experiences. To support the expanded connectivity needs for the next decade, we need to start the 5G definition now in order for it to be ready around 2020. Qualcomm Research is actively working on defining this unified 5G platform. We envision this platform to be about enabling new services and devices, connecting new industries, and empowering new user experiences. To accomplish this, Qualcomm is developing a unified 5G design that is scalable and adaptable across an extreme variation of use cases and that supports all spectrum types and bands from sub-1GHz to 6GHz and higher mmWave bands. In parallel, 4G LTE will evolve to reach its full potential with 5G seamlessly leveraging investments in 4G and Wi-Fi.
As such, Qualcomm believes 5G is not just a new generation, but a new kind of network.
⇒ From Technology Invention Site – Envisioning the Future with 5G [Qualcomm, Nov 26, 2014]
A new kind of network
A new generation has to provide significant gains over the previous generation. 5G will enhance today’s and emerging service with improved cost and energy efficiency. But 5G has to look beyond today’s trends, it needs the ability to virtually connect everything and go beyond human perceived requirements with emerging services like mission critical applications and bring awareness. We believe this requires a new kind of network.
5G needs scalability and adaptability across extreme variations in use cases.
- Connect massive number of things while scaling down cost and energy consumption
- Scale up to enhance the foundational mobile broadband service
- New mission critical services to remotely control vehicles, medical procedures with ultra reliable links
- Proximal and aware services to empower and optimize new user experiences
5G needs a user-centric approach to scale for billions of connected things:
- One that will bring content, connectivity and computing close to the user
- For connectivity, it means that devices are no longer just end-points
- This distributed approach also lowers latency and improves cost and energy efficiency
5G needs to be a unified platform designed for expanded connectivity needs in next decade:
- A unified air interface across all spectrum; licensed and unlicensed, below 6GHz with the addition of higher bands above 6GHz such as mmWave.
- Support new service through broader dimensions of improvements
- Scale across business dimensions like subscription and deployment models from traditional wide area to enterprise, hotspots and residential deployments
5G needs to meet expanded connectivity needs for the next decade in broad areas such as:
- Uniform fiber-like broadband everywhere—not just higher peak data rates
- New services like mission critical applications to remotely control medical procedures, vehicles and industrial processes
- Proximal and contextual awareness to empower new user experiences and optimize
- The ability to connect everything from simple sensors to complex robots
The user centric connectivity—as apposed to a spoke and hub model—will provide edgeless connectivity. With device-to-device communications and multi-hop, devices are no longer end-points, they are integral parts of the network. Access nodes, but also devices, can now act as relays and hubs. Integrated access and backhaul helps to solve challenges with installing fixed backhaul.
Simultaneous connectivity to leverage 4G investments
5G will unify all that has been added to 4G, like broadcast and unlicensed spectrum. A new design needs to go much further with cost, energy efficiency, ultra low latency and reliability for new services.
Drive 4G to its full potential
We are committed to drive the best possible 4G LTE, which will continue to evolve in parallel with the advent of 5G.
October 6, 2014 – 5G: new mindset meets next generation networks by Ericsson CTO, Ulf Ewaldsson
December 30, 2014 – Foxconn said to partner with Ericsson for 5G R&D from DIGITIMES
Foxconn Electronics (Hon Hai Precision Industry) reportedly is preparing to invest NT$4 billion (US$126.16 million) and a team of 200 technicians for 5G communication technology development and is considering forming cooperation with the largest telecom carrier in Europe Ericsson, looking to join the establishment of 5G standard during 2016-2018 to compete against Samsung Electronics and Huawei, according to a Chinese-language Commercial Times report.
Foxconn has recently started seeking assistances from Taiwan government units and related corporations and is looking to become the leader of 5G technologies, ahead of other Taiwan-based telecom carriers and equipment makers, the report noted.
Since Foxconn already has product lines for manufacturing telecom equipment and has license for running telecommunication business, sources familiar with Foxconn’s plan said that Foxconn is looking to join the establishment of the global 5G standard to help it become a core network equipment player worldwide, similar to China’s Huawei, the report added.
October 20, 2014 – Tech Talk: The Journey to 5G by Sara Mazur, Vice President and Head of Ericsson Research
November 24, 2014 – From 5G Concept presentation by Ericsson Research
2020: small cells and 5G
– impact on Transport, sync, security?
LTE will become the dominant technology < 6 GHz with backhaul, midhaul & fronthaul transport5G: “Seamless wireless Internet”
– “10 Gbps” throughput on air interface
› More bandwidth (100G → 1T) needed on backhaul
› e2e transport + radio solutions
– “1 ms” latency on air interface
› Lower latency on backhaul and fronthaul
› More localised X2 routing over midhaul (including IPsec)
5G target architecture
– Ultra dense, 10 m cell range
› Use any available transmission for backhaul, midhaul & fronthaul
› Cloud-based security infrastructure
› ”Intelligent” transport (e.g., SDN, SON)
– New Licensed/unlicenced spectrum
› Available for wireless backhaul
↑ transport Evolution: ↓latency, ↑BW, ↑connections, options ↑
5G Technology areas
What is 5G?
November 22, 2013 – Tech Talk: 5G by Erik Dahlman, Senior Expert in Radio Access Technologies within Ericsson Research
⇒ Blog post – 5G METIS conference, permeated by consensus [Dr. Olav Queseth, Ericsson Research and METIS Project Coordinator on Ericsson Research Blog, Nov 24, 2014]
⇒ Blog post – Ericsson hits 5 Gbps in 5G labs demo [Johan Furuskog, Ericsson Research on Ericsson Research Blog, July 2, 2014]
⇒ All posts on “5G” in Ericsson Research Blog
⇒ White Paper – 5G Radio Access – Research and Vision [Ericsson, July 25, 2013]
⇒ METIS 2020, a €27M and 2,5 years long EU co-funded research project announced (and initiated) in November 27, 2012 by Ericsson: The 5G Mobile and Wireless Communications system [Dr. Afif Osseiran, Ericsson, METIS Project Coordinator, November 21, 2013]
Project’s end date: 2015-04-30
4. Current state and market forecasts
December 15, 2014: from What’s HOT at Globecom 2014? communiqué by Lou Frenzel from Electronic Design Magazine
The greatest emphasis, in my opinion, was on 5G cellular. 5G is still in the development stage but there was lots of discussion about what it is and when it might come on line. The consensus seems to favor millimeter wave bands (28, 38 and 72 GHz) using small cells and massive multi-user MIMO. But as they say, the devil is in the details. There is still lots to do in defining, developing and deploying 5G. A good guess is that we won’t see this until 2020 and beyond. Most carriers haven’t even started LTE Advanced which is next up for 4G.
Along with 5G were multiple sessions on heterogeneous networks and small cells. The small cell movement has been around a while but few have been deployed. It is finally happening but there apparently are lots of options and some critical issues to solve. Self-organizing networks (SONs) are a part of that effort.
Another hot topic was the internet of things (IoT). This trend is already taking off in many forms. Yet there is still some confusion over multiple standards. Many wireless technologies are used including Wi-Fi, Bluetooth, ZigBee, ISM band and others. The new Bluetooth 4.2 and ZigBee 3.0 standards will boost their use.
One major theme covered was software-defined networking (SDN) and network function virtualization (NFV). Multiple sessions attempted to define each and show how they will impact all future networks. SDN/NFV seems to be one solution to the security problem as well as implementing 5G. The transformation is just beginning but this appears to be a major shift in networking. A fully programmable network is in our future. Some other topics with multiple sessions were cognitive radio, Wi-Fi, sensor networks and fiber optics.
As for exhibits, there were few. Only about a dozen companies showed. The largest exhibit was Austin- local National Instruments. They had the most significant announcement at the show. NI introduced their LabVIEW Communications System Design Suite. Dr. Truchard, NI CEO briefed me on this amazing new development platform uses a combination of hardware and software for prototyping software-defined radios and other new wireless systems. It is already being used to test new modulation algorithms, 5G approaches and large MIMO systems. Look for it to expedite 5G.
December 12, 2014 – #GLOBECOM: National Instruments CEO on 5G Prototyping video in which Dr. James Truchard, CEO and co-founder of National Instruments is highlighting NI’s new LabVIEW Communications System Design Suite, which combines software defined radio hardware with a software design, according to the company. The product helps engineers prototype 5G systems.
December 8, 2014 – 5G Standardization Begins As Companies Lay Royalties Groundwork press release from ABI Research
Work on new technologies that will form the foundation of 5G standards has been occurring for the last few years; however, the formal standards process has just begun. A range of major vendors are working on all aspects of 5G including; Alcatel-Lucent, Ericsson, Huawei, and Nokia Networks and device, semiconductor, and IP vendors including Intel, InterDigital, Qualcomm, Samsung, various mobile operators, academic bodies, and start-ups as well. “These companies are all waving their 5G flags, although 5G definitions and visions remain very vague,” commented Research Director Philip Solis. “But this is not merely marketing. These companies are most certainly putting a stake in the ground with regards to contributions to 5G that will leverage their work, competitive strengths, and most crucially, patents.”
Some highly influential companies, such as Qualcomm, have remained quiet until recently about their vision and plans for 5G. Meanwhile, more companies, previously not very involved with standardization efforts are putting their hands up. Apple’s involvement with the NGMN 5G Initiative is a perfect example, as is Google’s acquisition of Alpental—even if Google might only use a 5G or 5G-like air interface to augment fiber-to-the-home deployments with a combination of fiber-to-the-curb and 5G.
These companies are working together so the standardization process can hit the ground running. They are doing their own work, forming alliances with universities and other companies, and hedging their bets by partaking in different research projects that focus on different parts of the network and air interface, in an effort to dictate the direction of 5G.
“Expect efforts to get intellectual property into standards to be fiercer than with 4G, but naturally much of the existing IP will be in play as well,” added Solis. “More companies learned the importance of having a fair amount of IP with 2G and 3G, so the 4G playing field evened up a little. This trend will continue with 5G.”
Companies should also move beyond sometimes vague marketing and generalizations around 5G and the IoT and create more definitive messaging around how technology will improve specific applications. They need to better describe how waveforms and modulation schemes best apply to increasingly mixed-use traffic. This can only help them with more brand building and influence in the standardization process.
December 8, 2014 – 5G in 2020 Will Be Rare; Over 100 Million Subscribers by 2025 press release from ABI Research
According to a new Market Data forecasts from ABI research, it will take more than 5 years for 5G to reach 100 million subscriber mark—2 years longer than 4G. 4G subscriber growth was much faster than with previous generations, fuelled by the capabilities of increasingly powerful smartphones and the availability of 4G devices. 5G subscriber growth will likely be a bit more muted at first due to the increased complexity of 5G cells and networks, but will pick up in 2023.
“There are a number of commonalities between countries that are early builders of 5G networks. They have a large population, of which a large percentage is living in urban areas. They also have many companies pushing the envelope with IoT strategies. These countries will drive 5G subscriber volumes,” said Research Director, Philip Solis. “These are the United States, China, Japan, South Korea, and the United Kingdom in order of 5G subscribers in 2025.”
It is also important to understand the nuances around 5G to recognize where it is headed. “5G will be a spectrum of evolution to revolution—it will be an evolution of the way the core network and network topology is transforming now, but it will be clearly delineated as a fifth generation mobile air interface on which the mobile network of the 2020’s and 2030’s will be built,” added Solis.
5G will encompass spatial division as the foundation of the air interface, leveraging techniques like massive MIMO—achievable in devices because of the high frequency of spectrum that will be used—and 3D beamforming to form narrow beams that divide the space around a 5G basestation. Client devices will have links to multiple cells simultaneously for robust connectivity. Spectrum will be used flexibly and shift as needed between access and fronthaul and backhaul. The waveform and modulation scheme are the least clear aspects of 5G currently.
A 5G network will be a network of small cells and will be practical in urban and industrialized environments for the population density and the reflections in urban canyons; however, expect a scaled down version of 5G to use existing spectrum for macrocells as well in the longer term. One potentially problematic issue, however, will be regulatory issues concerning concentrated RF beams in centimeter and millimeter wave spectrum.