In Communications, "More" and "Faster" are the Watchwords

Key wireless technologies continue to evolve to meet changing communications needs and opportunities.

April 3, 2016

12 Min Read
Telecom cables
Hywards, Thinkstock

By Lou Frenzel

You can never have enough data speed. And you can never have enough memory, network connections, bandwidth, or longer range. More is always better. Technologies that offer more of anything continue to drive communications development. You probably already know that these technologies are including 5G wireless, the Internet of Things, LTE, and Wi-Fi enhancements. Here is a summary of what is happening now and what is coming in the years ahead.

5G Wireless

Fifth-generation (5G) wireless technology research and development projects are now well under way. This worldwide effort will eventually lead to new standards for cellular and broadband wireless. With 4G LTE still not fully deployed, you have to wonder just why there is a need for a new standard. As usual, the desire is simply to increase the subscriber capacity and to offer higher data speeds. Faster mobile broadband is needed to support the ever-increasing need to stream video and transmit photographs.

In addition, there is a need to be prepared to handle the glut of messages that will eventually come from the emerging Internet of Things (IoT) and machine-to-machine (M2M) markets. Billions of new connections are projected. Then there is an eventual need for a highly reliable low-latency service to handle applications such as vehicle-to-infrastructure (V2I) communications in the forthcoming intelligent transportation system (ITS).

The features and specifications of 5G will eventually form a new standard that most worldwide carriers will adopt. This R&D is being done in universities, company research labs, and international organizations such as the Third-Generation Partnership Project (3GPP). The standard will in due course be adopted by the International Telecommunications Union (ITU) as IMT2020.

While the details are still in flux, we do know some of the general characteristics of the 5G system. Here is a brief summary.

  • Data rate. The targets are a data rate of at least 1 Gb/s and as much as 10 Gb/s peak with an average user rate of about 100 Mb/s.

  • Spectrum. The millimeter wave (mmwave) bands in the 6 GHz to 100 GHz range will be used. Governments are just now allocating spectrum for 5G. In the U.S., the Federal Communications Commission recently proposed the following band assignments for 5G: 27.5-28.35, 37-38.6, 38.6-40, and 64-71 GHz.

  • Modulation and access. Multiple formats are being explored including standard OFDM, filtered OFDM, and a non-orthogonal method called generalized frequency division multiplexing (GFDM). Other variants include universal filtered multicarrier (UFMC), filter bank multicarrier (FBMC), and others.

  • Network configuration. Short-range small cells overlay existing LTE networks with self-organizing networking (SON) software for automating provisioning and deployment.

  • Virtualization and a cloud-based architecture. Software-defined networking (SDN) and network function virtualization (NFV) techniques are expected to be used to establish and manage the network.

  • Massive MIMO. Configurations of 4 × 4 and 8 × 8 will be common to help achieve the desired data rates and reliable links in the mmwave bands.

Many technical issues are yet to be resolved. Dealing with mmwaves is a major problem because of the high channel propagation loss. These signals experience diffraction, scattering, and reflections. Penetration loss of buildings, trees, and other obstructions is also extremely high. These problems will be dealt with by the use of higher power and MIMO. The higher power will come from agile beam-forming directional gain antennas. Phased arrays widely used in radar are very small at mmwave frequencies and offer an excellent solution.

Another looming challenge is putting all this mmwave hardware inside an already fully packed smartphone. Space for MIMO chips, filters, Pas, and antennas and the resulting higher-power consumption are problems yet to be solved.

5G is a work in progress. Research, prototyping, and trials are ongoing. Research is focused on things such as modulation methods and mmwave channel models that are critical to the success of this advanced wireless technology. Initial deployments are not expected until 2020 and beyond.  Cellular infrastructure provider Ericsson predicts that 5G will be adopted faster than 4G with as many as 150 million subscribers by 2021. In the meantime, we have a very successful 4G LTE system in place that is doing a fine job and it, too, is a work in progress.

4G LTE

Most cellular operators are still rolling out their LTE networks while still maintaining their 2G and 3G connections. Sprint is finally shutting down its 4G Wimax network in favor of LTE service. A few carriers are beginning to deploy LTE Advanced. LTE-A uses carrier aggregation (CA) to  provide wider bandwidths (up to 100 MHz) to deliver higher speeds approaching a peak of 1 Gb/s. Again, Sprint’s latest network expansion uses some carrier aggregation and other LTE-A features to boost speeds. It will take a while for full LTE-A rollout. This will provide much higher speeds until 5G comes along.

Another enhancement beginning to show up in existing LTE networks is voice-over LTE. VoLTE is the compressed voice service that is expected to replace the 2G and 3G voice service still used by most carriers. VoLTE is far more efficient in terms of spectrum and should significantly improve voice capacity and performance while lowering costs. AT&T and T-Mobile have begun implementation, but Verizon, Sprint, and others lag in this deployment. Juniper Research anticipates an estimated 2 billion VoLTE connections by 2020, up from 123 million connections in 2015.

Another LTE development is LTE-U for unlicensed. Another name for LTE-U is Licensed-Assisted Access or LAA. LAA is a method of using the 5 GHz unlicensed Wi-Fi spectrum for LTE cellular data. The big question is will LAA interfere with Wi-Fi access points (APs) and hot spots, disturbing millions of Internet connections. So far, there is an ongoing battle between the cellular carriers and the Wi-Fi interests over the potential deployment of LTE in the unlicensed spectrum. However, it is an interim solution for the lack of new spectrum for high-speed LTE.

The Third-Generation Partnership Project (3GPP), the organization that develops the next cellular standards, is working on LAA right now to be incorporated into the forthcoming LTE Release 13. Companies like Ericsson and Qualcomm already have products under developments. The FCC seems to be staying out of this fight as long as current regulations are met. Some carriers say they will try to implement LAA in 2016 as part of their LTE-A small cell system using carrier aggregation. If it works, it is just an LTE stop-gap.

LTE will be with us for many years to come. Its enhancements will keep it useful and competitive until 5G arrives.

Wi-Fi

There is always something happening with Wi-Fi. This wireless technology has become like an expected utility service to us all. We use Wi-Fi at home, at work, and on the road to connect our smartphones, tablets, and laptops. It is everywhere and mostly free as the FCC is fining hotels and convention centers for charging a fee. Because of its ubiquity, Wi-Fi is expected not only to be there but also to become faster and more reliable. New versions and enhancements are always in the works with the IEEE standard body and the Wi-Fi Alliance.

Right now, most networking organizations are still upgrading their networks from 802.11n to the faster 802.11ac standard. The 11ac version operates only in the 5 GHz unlicensed band, but is faster. The Wave 1 version can achieve up to 1.3 Gb/s peak under ideal conditions. Most new smartphones, tablets, and laptops already incorporate Wave 1 chips. Wave 2 versions are showing up in routers and some advanced access points (APs). Wave 2 802.11ac uses 80 and 160 MHz wide channels to achieve peak rates of over 6 Gb/s. Wave 2 also uses multi-user MIMO to allow up to four users to transmit data at the same time.

As both Wave 1 and Wave 2 versions are rolling out in the months and years to come, network operators are considering whether to upgrade their wired Ethernet connections from 1 Gb/s to 10 Gb/s to keep up with the wireless speeds. Ten-gigabit Ethernet is available, but still expensive. Some are using multiple 1 Gb/s connections as an interim solution. There are 2.5 Gb/s and 5 Gb/s Ethernet versions being developed. Called NBASE-T, this new version of Ethernet is expected to be ratified later in the year, but companies like Cisco already have routers and APs available.

The next generation of Wi-Fi is 802.11ax. It is under development in the IEEE’s 11ax task group. The 11ax version will use advanced MIMO and orthogonal frequency division access (OFDA) to offer greater capacity and higher speeds. OFDA-like OFDMA in LTE divides the OFDM spectrum into smaller clusters of subcarriers so that multiple users can be accommodated simultaneously. Peak data rates could top 10 Gb/s, but a typical user can expect rates of hundreds of Mb/s up to 1 Gb/s. The 802.11ax standard is not expected to be ratified until 2018.

Others Wi-Fi versions of interest are 802.11af and 802.11ah—802.11af is a version of Wi-Fi that uses the unused television white-space (TVWS) channels in the 54 to 698 MHz range. These VHF/UHF channels promise much longer-range communication than the traditional 2.4 and 5 GHz bands. Typical range with 100 mW is 1 km or more depending upon frequency and antenna height. Using 6 MHz wide channels with higher-level QAM data rates to 24 Mb/s are possible.  The use of the TVWS requires cognitive radio techniques to avoid interference to TV stations and wireless microphones. The 11af standard is available now, but has not been widely adopted.

The 802.11ah standard is another <1GHz wireless technology that uses the 902-928 MHz unlicensed band. Using 1, 2, 4, 8, or 16 MHz wide channels 11ah can provide data rates of 100 kb/s to as much as 40 Mb/s, depending upon modulation method and coding. A range of many km is possible. This makes 11ah flexible to accommodate a wide range of applications. The 11ah standard is expected to be finally ratified this year. Its intended use will be in IoT activities and could replace M2M is some situations.

The Internet of Things

Probably the most hyped technology in 2015 was the Internet of Things (IoT). It seems as if every electronic-related company is gearing up to enter this business that appears to offer something for everyone. But does it really?  Maybe companies are basing their IoT plans on some of the early overly optimistic predictions of 10 to 50 billion connected devices by 2020 and trillions of dollars in new revenue by 2025. Similar projections seem to be way off base.

UK market study firm Beecham Research recently indicated that “…these numbers to be unrealistic and potentially damaging to the industry, if they are believed, and companies are building their business plans and funding expectations on such false promises.” Furthermore, Beecham also indicates that currently there are less than 1 billion connected devices, not including smartphones and tablets. While several predictions indicate that many of the new connections will come from ome automation, Beecham indicates there is no evidence that the connected home market is taking off in a big way. If not the home market, what will be the largest initial segment of the IoT movement?

Taking in all of the latest articles, research reports, white papers, product brochures and other sources, here is a summary of current and expected outcomes.

  • The industrial Internet of Things (IIoT) may be the largest beneficiary of initial IoT efforts as there are plenty of real needs and buyers for remote monitoring and control applications as sensor and actuator networks in factories, process control plants, oil and gas pipelines, utilities, and the smart grid.

  • No one wireless technology will emerge as the clear winner in IoT applications as Wi-Fi, Bluetooth, ZigBee, 802.15.4, Z-Wave, and a few others will all find their niche. In fact, one company, Systech Corp. has a line of gateways that are full routers for Internet connections that can accommodate a mix of modules for any of the above-mentioned wireless technologies, including cellular (see photo).

  • Longer range low power wide area (LPWA) networks using newer wireless technologies such as LoRa, Sigfox, Weightless, 802.11f/h and others will greatly expand the usefulness of IoT.

  • Longer-range low-power wide-area (LPWA) networks will also probably impact the M2M segment of this movement. M2M typically uses 2G or 3G cellular technologies for remote monitoring and control and some Internet access. LPWA technology will capture some of the M2M business. However, M2M will continue to be competitive as a new version of LTE comes on line in the coming months. Called LTE-M, it is a modified version of LTE in Releases 12 and 13 of the 3GPP standards that reduces power consumption, bandwidth, data rates, and electronics costs to accommodate large numbers of devices over the cellular networks. This could lead to more rapid phase-out of the older 2G systems.

  • No one software platform appears to dominate the IoT landscape. Many are available including AllJoyn, Thread, Brillo, MQTT, IPSO, JSON, IoTivity, HomeKit, and multiple others. As with wireless technologies, these multiple platforms will find their markets and niches.

  • Security is still a key factor in all segments of IoT. All hardware devices incorporate encryption, but service providers will also face the need to provide additional authentication and security features to meet the demand.

  • IoT and M2M produce massive amounts of data. Some studies indicate that as much as 60% to 90% of that data is never used. To make IoT worthwhile, there is a huge need for data analytics to make useful sense of all that big data. This is a complex, but potentially rewarding opportunity for some software companies.

  • Node costs can be an issue for some applications. Many simple sensor nodes just do not need a 32-bit processor with an operating system to be effective. Simpler focused software on a dirt-cheap 8-bit processor is what is needed.

  • What is really needed is a good business case for many application areas.

Article was originally published on Electronic Design.

Sign Up for the Newsletter
The most up-to-date news and insights into the latest emerging technologies ... delivered right to your inbox!

You May Also Like