Steven J. Crowley, P.E.
Consulting Engineer
Recently
What’s 4G? Whatever you want.
On October 21 the ITU announced it had deemed two mobile broadband technologies as IMT-Advanced. They’re LTE-Advanced and WirelessMAN-Advanced (WiMAX Release 2 based on IEEE 802.16m). At the same time the ITU called these “true 4G technologies,” leading some in the trade press to say that marketing of some existing services as 4G is misleading.
No entity is the arbiter of the designations 2G, 3G, or 4G. Even the ITU generally uses these terms parenthetically. It recognizes the lack of consensus on their meaning, and does not say others are wrong. The ITU calls the IMT-Advanced technologies “true 4G,” and they are insofar as ITU has a view of 4G and the two technologies are truly consistent with that view.
In some of these articles, Sprint and Clearwire, to name two examples, catch flack for calling their current WiMAX offering 4G. I’d call it 3.5G, but I can’t say they’re wrong. An ABI Research analyst agrees with Sprint and Clearwire; his view is that everything TDMA is 2G, everything CDMA is 3G, and everything OFDMA is 4G, regardless of data rate. It’s a reasonable argument based on backward compatibility, but I’m not ready to agree. If that or something else ends up being the industry consensus, I won’t object.
Below is a comparison found on the ITU website. It’s generally consistent with my view.
Don’t like these classifications? Make your own. Within reason.
________
UPDATED 11/04/2010
ABI analyst link updated with more current post.
The Spectrum Famine
Mitchell Lazarus writes in the October 2010 issue of IEEE Spectrum on The Great Radio Spectrum Famine — the challenges of getting more spectrum for mobile broadband. He notes the importance of incentives that can lead users to optimize or reduce spectrum usage on their own. There is some great historical information as well. It is another informative and thought-provoking article from Mr. Lazarus, who can also be found from time to time on the CommLawBlog.
A comment of mine was posted along with the article today. I copy my comment below:
Thank you Mitchell for sharing your insights and providing an important historical perspective. I had almost forgotten about the assignment by industry in two-way. Another scary tale if I run out of ghost stories this Halloween.
I, too, have followed the decline in terrestrial TV viewing for years. Today, though, I see more people “cutting the cord” and dropping cable or satellite. They use a combination of internet and over-the-air viewing. I wonder if this downward trend is ending.
In the Capitol Hill neighborhood in Washington, the local wireless internet service provider is now offering a service called “free TV.” You pay them $300 to put an antenna on your house and hook it up to your TV. Twenty channels with no monthly fee. They market this as an innovation, and it sells.
We see these large projections for future mobile broadband demand; a lot of that is video. Mobile operators have been working about 10 years to engineer and make a business case for broadcasting. Technically, they can make it work but it’s not a business yet, anywhere in the world, in part because of the system capacity tied up in providing the service.
In the hierarchy of mobile needs, I’d say voice is most basic closely followed by text. Video is nice, but less essential to my life. I’ll use it on my unlimited data plan. If a per-byte tiered-pricing plan is put in place, I may pass. I’ll download it at home and move it to my phone. Or wait until I’m near a Wi-Fi hotspot.
I do believe in the next few years we’ll see some significant technologies that allow for more efficient spectrum use. Source coding technology will be one of those areas. There will also be better networking protocols for heterogeneous radio networks what will allow for more seamless handover among 3G, 4G, Wi-Fi, and even Ethernet; this will make it easier for operators to move off crowded mobile broadband spectrum whenever possible. Applications can be more tightly coded.
None of the above is meant to deter the search for more spectrum for mobile broadband. I like the FCC’s plan for the voluntary transfer, using financial incentives, of TV allocations to mobile broadband. As you suggest, incentives help. I like the suggestion involving a non-profit group. In addition, I wonder if spectrum can be made more like property; when people own something, then tend to take care of it and maximize its value.
One comment here suggests femtocells as part of the solution. Femtocells are a good fix to fill in coverage holes. From what I have seen so far, however, they won’t be much help with the spectrum crunch. Interference issues preclude ubiquitous placement.
netBlazr: Unlicensed Wireless versus Business Broadband Incumbents
Unlicensed wireless may disrupt the broadband service market in Boston. Communications engineer Brough Turner and his partners have founded netBlazr, a new business broadband service that takes advantage of cheap metropolitan fiber and uses unlicensed wireless technology from Ubiquiti Networks that can operate with various Wi-Fi protocols, and a proprietary one, in a mesh configuration. In the netBlazr network the devices operate at an aggregate data rate of 100 Mbps with 50-500 meters per hop. Throughput per user is 10 Mbps.
In Boston, Verizon charges about $2,200 per month for symmetrical 10 Mbps business service via fiber; netBlazr charges $189 per month for equivalent service using its hybrid fiber/wireless system. Less-expensive shared plans, including one that’s free, are available for businesses with lower-priority traffic.
In the Transition to 4G, Voice will Stay on 3G
Most cellular operators are beginning the transition from 3G technologies to 4G LTE. That will mainly be for high-speed data and not voice, support for which will come years later. Communications engineer Jim Murphy describes some of the networking issues that make voice impractical on LTE on day one. To his comments I’d add that the preferred LTE voice solution is VoIP on the IP Multimedia System (IMS) framework, which has had slow adoption in the mobile community.
Jim’s example is from the 3GPP family of technologies, of which LTE is part. Most 3GPP2 operators, such as Verizon Wireless, are also planning a migration to LTE. They migrate from a different family of technologies, those specified in 3GPP2 and which include cdma2000 developed largely by Qualcomm. In Verizon’s transition, high-speed data on cdma2000 networks, now provided by EV-DO, will migrate to LTE. Voice will stay on cdma2000 1x for several years. That’s not a problem, because that technology keeps improving. The latest version, called cdma2000 1x Advanced, has achieved a 50x increase in voice capacity over the predecessor IS-95 system launched in 1993.
Both 3GPP and 3GPP2 3G voice feature a soft-handover (make-before-break) scheme that allows for, essentially, a circuit-switched connection. Yes, you’ll be able to run Skype and other VoIP services on 4G LTE data networks, if you want. In terms of the total user experience, however, most will prefer the seamlessness of standards-based integrated voice capability.
VoIP is the future. For 4G voice, though, it’s just great in theory at this time.
Progress & Freedom Foundation Closes Doors After 17 Years
One of the original digital-economy think tanks, the Progress & Freedom Foundation announced it is shutting down after 17 years of operation. Broadcasting & Cable points to lack of funds for the closing.
FCC Finalizes White-Space Rules
The FCC finalized its white-space rules today, acting on petitions for reconsideration of its earlier decisions. It issued an 88-page Second Memorandum Opinion and Order that explains its decisions and includes the final white-space rules. A much-shorter press release was also issued.
At least one FCC observer has noted an uncharacteristic level of hype in today’s announcements. The FCC calls it “super Wi-Fi,” and adds the “potential uses of this spectrum are limited only by the imagination.”
Over two years ago, Google called it “Wi-Fi on Steroids.” It was later picked up by the popular press. Not all agree; it’s “Wi-Fi on Crutches” according to one who dares to consider the realities of physics and economics.
I’ll call it “Wi-Fi on Caffeine,” at least with respect to better range and coverage — if not data rates — compared with current Wi-Fi equipment. This is partly due to operation in the UHF-TV band instead of the 2.4 GHz band. In major markets and their suburbs, there will be few or no channels available for white space use. In rural areas and other less dense areas, the technology will be a good fit with Wireless Internet Service Providers (WISPs) and other longer-distance applications.
Cellular operators would like some of the white space on a licensed basis for backhaul in rural areas. They didn’t get it today, but the FCC is actively considering it and we may hear more on that by the end of the year. No way are all these vacant channels going to be occupied by internet services in the most rural areas, so the proposal of the operators makes sense.
In IEEE 802, Working Groups 802.22 and 802.11 are working on standards that can be used by equipment in these applications; 802.22 may be the one with longer range. Working Group 802.19 is trying to facilitate coexistence between the two. Now, there are asymetric interference effects, which is causing friction between the two groups beyond the normal competition. (802.22 takes the harder interference hit.)
There will be other standards and equipment as well. The white space concept is international, but unique to each area of the world.
Equipment is not easy; it’s challenging to develop sufficiently-broadband power amplifiers and antennas, and to meet the emission mask in a cost-effective manner.
Another challenge is developing a business plan when 120 MHz of TV spectrum could be taken away under the National Broadband Plan.
Staff at the FCC: How Many Lawyers, Economists, and Engineers?
The FCC’s professional staff is dominated by lawyers, and it wants to increase its numbers of economists and engineers. How does the FCC’s staffing compare to telecommunications regulatory agencies of other countries?
Economist and former FCC Senior Adviser J. Scott Marcus makes this comparison in a new paper. He cautions that his results are indicative, rather than conclusive. It’s hard to compare some agencies because, depending on the country, the same agency that regulates telecommunications may also regulate, say, trains and the postal service.
Regardless, his analysis makes for some interesting comparisons. Figure 5 from the report, below, shows the distribution of professions among professional staff, and shows the predominance of lawyers at the FCC.
Figure 6 is a similar comparison among senior managers where, at the FCC, engineers and economists are basically non-existent.
Marcus observes that the percentage of lawyers in these agencies generally tracks that of the countries as a whole. US society is litigious, so it may be rational for the FCC to have many lawyers.
This report is also noteworthy for it being the first opportunity for some of these agencies to compile these figures.
Mixed Messages on IPv6
The Institute of Electrical and Electronics Engineers (IEEE) is an international engineering organization. It has a unit, IEEE-USA, which was formed in 1973 to support the public policy interests of IEEE members in the US. It is respected on Capitol Hill, which sometimes turns to IEEE-USA for input on technology policy matters. That’s fine, if the input is sound.
One problem is that participation in IEEE-USA tends to be dominated by those not in industry, such as government employees, consultants, and academics. This can sometimes result in unbalanced policy positions that are unfavorable to industry.
Take the transition to IPv6. Most of the networking industry agrees that IPv6 is the future. We’re running out of IPv4 addresses. There is no panic, but organizations should be planning for the transition. These notions are generally supported by Google, Cisco, Scott Bradner, and many others.
In its August 2009 White Paper on IPv4 address exhaustion, however, IEEE-USA says that it “discusses IPv6 only as an example replacement. This paper is not intended to exclusively endorse IPv6 as the sole replacement structure and strategy for IPv4. Indeed, the lack of adoption for IPv6, which is an aging alternative, may indicate that preventing premature IPv4 exhaustion is another viable strategy.”
It’s a significant backpedaling from the industry position, plus a dig at IPv6 for having been around a while. This is supported by the dues of industry’s IEEE US members. Industry can lobby government on IPv6, but when a staffer looks up the IEEE-USA view, some of that effort is wasted.
IEEE-USA positions are largely driven by volunteer input. If industry volunteers don’t participate, industry’s view will be inadequately reflected. If a few networking companies could participate more actively in IEEE-USA, it would help tighten up these positions and help the group be a better force all of IEEE’s US members.
Experimental Radio Applications at the FCC
This summarizes a selection of applications for the Experimental Radio Service received by the FCC during August 2010. These are related to radar, military communications, mesh networking, unmanned aerial vehicles, satellite services, biomedical telemetry, aircraft telemetry, safe-driving systems, geophysical sensors, electronic warfare, smart grid, and antenna testing.
- INOVA Geophysical Equipment Limited filed an application (with supporting exhibits) to test a proprietary mobile radio system in the 30-36 MHz and 150-174 MHz bands. The radio links would be used to control remote geophysical seismic recording equipment, which INOVA manufactures. At the end of testing, INOVA plans to put the radio equipment into production and lease it to customers.
- Fortress Technologies filed an application for experimental license to test several of its secure mesh-networking products developed for military applications. Several exhibits are included but they are not publicly available due to a confidentiality request. Operation is to be on 4.9425-4.9875 GHz.
Ultra-Wideband: How Regulatory and Standardization Delays Slowed a Wireless Technology
The FCC recently issued an order denying reconsideration petitions in its ultra-wideband (UWB) proceeding. That effectively ends the 12-year UWB rulemaking process. Mitchell Lazarus recounts how UWB became bogged down at the FCC and in a failed standardization attempt in IEEE 802.
UWB, as authorized by the FCC, operates across 3.1 to 10.6 GHz, with very low power at any one frequency; its tendency to cause or receive interference is very low.
IEEE 802 attempted to create a UWB standard in IEEE 802.15.3a but did not, as neither of two competing proposals reached the necessary voting threshold for approval. One of the competing proposals, Multi-band Orthogonal Frequency Division Multiplexing (MB-OFDM), has since seen some consumer success in Wireless USB, which is based on a platform maintained by the WiMedia Alliance; data rates are up to 480 Mbps at a range of about 10 feet.
UWB was eventually standardized in IEEE 802.15.4a, where it exists as an alternative physical-layer to standard IEEE 802.15.4-2006, a standard for very low power, low data rate devices. (The IEEE 802.15.3 family is for higher data rates with higher power consumption.) It uses what was the other competing proposal in 802.15.3a, Direct Sequence UWB (DS-UWB). This standardized form of UWB has been commercialized for asset tracking and other location services, but not yet for consumer applications.
Read the rest of this entry »
Cultural Barriers to Federal Spectrum Reform
As the FCC searches for more spectrum for mobile broadband services, its National Broadband Plan points to federal spectrum as a candidate. Since the National Telecommunications and Information Administration (NTIA) is responsible for allocating federal spectrum, the FCC can’t do much more. Still, the FCC’s recommendations are good. One is for the FCC and NTIA to “develop a joint roadmap to identify additional candidate federal and non-federal spectrum that can be made accessible for both mobile and fixed wireless broadband use, on an exclusive, shared, licensed and/or unlicensed basis.” In support of that, the “FCC and . . . NTIA should create methods for ongoing measurement of spectrum utilization.”
Variations of these proposals have been around for decades, formally and informally. Once in a while, progress is made. In 1995, NTIA suggested the changing the 3650-3700 MHz band from federal-only to mixed-use (federal and non-federal). That happened, and in 2005 the FCC adopted rules that resulted in the creation of the IEEE 802.11y standard. (That allows high-powered Wi-Fi equipment to operate on a co-primary basis in the 3650-3700 MHz band in the US, except when near certain satellite earth stations.)
So, it can happen. That, and recent FCC talk of “unleashing” broadband made me think the above recommendations in the FCC’s Plan might get some traction. I’m less sure now after following the latest writings on the topic by spectrum expert Michael Marcus.
In an August 17 post on his blog, Marcus asks why NTIA isn’t measuring occupancy of the almost exclusively-federal 225-400 MHz band. He finds that the Interdepartment Radio Advisory Committee (IRAC), NTIA’s advisory committee of federal users, is concerned that measurements in major cities – where spectrum is most needed – will show low occupancy because the band is primarily used by military aircraft. Marcus says enough with these delays; in the new era of cognitive radio and dynamic spectrum access technology, it’s time to see some hard spectrum data so sharing options can be examined.
If you’re intrigued by that, there’s more. An August 9 post says an NTIA spectrum advisory committee “evades some major issues and pushes the parochial agendas of some committee members without trying to relate them to the broader public interests.” A May 10 post takes you inside that committee’s meeting, and observes a general effort to protect incumbent spectrum users.
It can happen, but these reports suggest the timetable will be later rather than sooner.