Harm claim thresholds for satellite earth stations

I'm reasonably confident at this point about deriving harm claim thresholds for cellular neighbors and TV receivers (see e.g. the TPRC 2012 paper http://ssrn.com/abstract=2018080). Here's a first cut (a few months old, but I'm behind on blogging...) at thresholds for satellite earth stations.

TV/cellular guard bands - second thoughts

In a recent post ("The FCC's TV/cellular guard bands don't compute") I wondered whether the guard bands between TV and cellular service that the FCC proposed in its Incentive Auction NPRM (pdf) had been designed to make room for more unlicensed in the TV bands. Having spoken to some experts, I’ve concluded that I was probably wrong about that: my new best guess is that they’re a benefit for the cellular industry.

800 MHz receiver criteria as harm claim thresholds

Bob Pavlak at the FCC prompted me to read the 800 MHz “unacceptable interference” rules in the context of an interference limits approach (root post).

The 2004 Report & Order (pdf) introduced the concept of unacceptable interference, “a term of art adopted for the limited purposes of this proceeding … that defines a bright-line test for interference protection that takes into account, among other factors, the strength of the desired signal and the characteristics of the receiver being employed” (Report & Order, footnote 8). While this sounds like a receiver performance requirement, I think it actually amounts to an interference limit or harm claim threshold. In fact, if it is reformulated as a harm claim threshold, it becomes more powerful because it does not enshrine a particular set of receiver performance parameters in the rules, leaving manufacturers and system operator with more flexibility.

Receiver Interference Tolerance: The Tent Analogy

A postcript to my testimony (see previous post) at the House sub-committee on Communications and Technology hearing on receivers: It sounded like Rep. Walden, who chairs the sub-committee, hoped my oral presentation would've mentioned the tent analogy I included in the written testimony. So since at least one person liked it, here it is:

Testimony: Harm Claim Thresholds

I was privileged to testify yesterday at the House Committee on Energy and Commerce’s sub-committee on Communications and Technology hearing on the topic “The Role of Receivers in a Spectrum Scarce World.”

My testimony (pdf; my oral testimony at time code 0:17:23 in the YouTube video on the hearing web page; the members' questions start at 0:27:00; some press here and here) tried to make four points:

1. We need to improve the ability of radio systems in one frequency band to tolerate reasonable signals in adjacent bands. 
2. Receiving system operators must bear some of the responsibility, but need to know what those responsibilities are.
3. Regulators can bring receiving systems into the mix by setting harm claim thresholds (aka interference limits or receiver protection limits), i.e. the interference levels that a service needs to tolerate without being able to bring a harmful interference claim.
4. Congress can play a role by keeping up the pressure, allowing the FCC to move ahead, and funding FCC technical investigations.

The FCC's TV/cellular guard bands don't compute

The FCC incentive auction NPRM [1] proposes 6 MHz guard bands between cellular and TV services (actually 6-11 MHz, depending on how the auction works out). The number is arbitrary, and could well have been chosen on political grounds to make room for more unlicensed in the TV bands.

The impact on interference from cellular systems into TV receivers is much the same whether the guard band is 1 or 20 MHz: most receivers will be unaffected, and for the small but significant number that suffer harm (0.5-5%?), only receiver filters will really help. The real question is: who's responsible for buying and installing those filters - the consumer or the cellular companies?

TV whitespace vs. cellular power limit anomalies

In the previous post, I considered interference between cellular base stations and TV receivers. What about interference between cellular handsets and TV? Considering this case highlights striking contradictions between the low power allowed for TV whitespace devices and the high power the FCC proposes for cellular operation: 20 dBm for whitespace personal devices but 37 dBm for cellphones, in both cases with a 6MHz guard band.

Post-auction cellular interference into TVs?

How many TV receivers will be affected by interference from cellular services as a result of the FCC’s "incentive auction" plan?  The FCC’s proposal doesn’t venture an answer; I don't think it even asks the question. Ofcom’s technical analysis in the UK’s rearrangement of the TV bands to accommodate more cellular service suggests that the number will be small, but not negligible. Ofcom therefore decided to require cellular operators to install filters on TV sets where there is a problem; the FCC has not raised this possibility.

In summary, the UK modeling suggests that TV reception will be affected in about 5% of  homes if there's a 11 MHz guard band between TV and cellular channels; the FCC's proposed guard band will be 6 - 11 MHz, depending on auction outcomes. By far the most effective way to mitigate this interference is by installing a TV receiver filter in affected homes.

Receiver regulation: Why no progress?

(Written with Madelaine Maior, Silicon Flatirons research fellow)

There’s an emerging consensus that the role receivers play in interference should be recognized in wireless regulation. But why has it taken so long to come to a conclusion about the performance of receivers?

Spectrum sharing is not a partisan issue – NOT

After his keynote at Dyspan yesterday, PCAST member Mark Gorenberg and his podium guests (and spectrum report co-authors) Dennis Roberson and Michael Calabrese were asked about the political prospects for the PCAST spectrum sharing recommendations (pdf). I can’t remember exactly who said what, but the message was that spectrum sharing wasn’t a partisan issue. Not so.

Three meanings of "spectrum efficiency"

“Efficiency” is a word of power, chanted when someone wants to bewitch an audience with the potency of economics. It’s often used in wireless policy, and I’ve realized that even when “spectrum efficiency” isn’t purely a fetish and is used to refer a ratio of input divided by output, the amounts compared depend on whether the speaker is an engineer or economist.

PCAST Report endorses receiver interference limits

The President's Council of Advisors on Science and Technology (PCAST) released its report Realizing the Full Potential of Government-Held Spectrum to Spur Economic Growth today (PDF, Administration’s blog post, webcast).

The main thrust of the report is the need for a shift from clearing and reallocating federal spectrum to dynamic sharing. As part of implementation, the report recommends that interference limits are used to include receiver considerations in spectrum management. (I was an advisor to the PCAST committee that wrote this report.)

Recommendation 3.1: The Secretary of Commerce working through the National Telecommunications and Information Administration (NTIA), in cooperation with the Federal Communications Commission (FCC), should establish methodologies for spectrum management that consider both transmitter and receiver characteristics to enable flexible sharing of spectrum. To safeguard primary Federal users, FCC should require that future non-Federal devices will be permitted to share government spectrum as Secondary Access users only if they are certified to operate within the stated interference limits for the band of interest. Initial specification of protection should be reviewed such that they safeguard new FCC assignments against harmful interference while grandfathering in existing devices and operations.

The report recommends that “[i]n order to facilitate more intensive and efficient sharing among Federal users, the NTIA should set and publish receiver interference limits using a transparent process for government assignments” It also recommends that “in the immediate timeframe, the FCC should begin the Notice and Comment cycle on implementing receiver interference limits as part of license terms for new allocations, updating old licenses to include receiver interference limits, and ex ante enforcement mechanism for non-Federal devices sharing with Federal users.” (Section 7.3, p. 77-78)

A detailed discussion of interference limits is given in Appendix D (p. 107 ff.).

From Kwerel & Williams to Interference Limits

Evan Kwerel & John Williams have proposed that future allocations should self-protect against projected adjacent band interference by assuming that they will receive only the “protections provided between flexible use bands” (Kwerel & Williams 2011, references at the end). The slide deck in Kwerel & Williams (2012) provides more detail: when a new allocation is being established next to a band likely to be repurposed for flexible use, the new allocation must (1) protect existing systems and future flexible use systems in that adjacent band, and (2) self-protect against interference from those systems, where flexible use systems is defined as “a dense deployment of base, mobile and fixed transmitters operating at fully functional power levels typical of a modern wireless cellular architecture.”

 Requirement (2) bears on the receivers of the new allocation. It resembles a qualitative interference limit based on the resulting energy from a “modern wireless cellular architecture.” A key selling point of this approach is that it doesn’t go beyond familiar parameters already used in regulation, like transmitter EIRP, compared to over interference limits that introduce probability distributions of resulting signal strength.

Thinking about a cellular deployment in the adjacent band is a very useful starting point. However, I do not believe it is precise enough to be useful in regulation, and particularly in enforcement. If one removes the studied ambiguity of the Kwerel & Williams proposal, the apparent familiarity and resemblance to existing rules evaporates, and one ends up with interference limits.

Transmitter versus receiver specifications: measuring loudness versus determining understanding

In arguing that regulators should attend to receivers as well as transmitters, I’ve may have mistakenly left the impression that they’re symmetrical, e.g. that one is a reflection of the other. For example, in “Four Concerns about Interference Limits” I observed that a communication system can be engineered to operate successfully either by improving receivers, or by delivering more transmitted power.

However, this framing is potentially misleading. It could be taken to mean that transmission and reception are two sides of a coin, that they are similar in kind, but that one is the reflection or complement of the other. They’re actually more like the outside (transmission) and inside (reception) of a black box - a black box like the human head, say.

Interference Limits for AWS-4

I have filed a comment (PDF) on the FCC's proposed AWS-4 rulemaking.

I argued that receiver performance affects the allocation of spectrum and opportunities for the creation of new services; that receiver management does not necessarily mean imposing receiver standards; that the Commission should define interference limits, an explicit statement of the radio interference level that a licensee’s system needs to tolerate, for AWS-4 licenses; but that receiver performance standards are not needed in this allocation.

Four Concerns about Interference Limits

Dennis Roberson has flagged some concerns about the interference limit approach. (The approach is summarized in http://sdrv.ms/ReceiverLimits; to trace the evolution of the idea, see the root post Receiver protection limits, and subsequent posts listed there).

His questions have helped me understand the importance of characterizing interference limits as a way to define harmful interference more precisely, rather than as a description of the radio interference environment. I probably over-played the interference environment angle in some earlier accounts in an attempt to distinguish interference limits from receiver performance standards using the slogan, “It’s not about the device, it’s about the environment.” That’s still true – the environment is indeed more important than the device – but using interference limits doesn’t require the regulator to measure and monitor the RF environment.

Dennis’s concerns relate to:

1. The regulator’s ability to adequately define the current stable environment, much less a dynamic future state environment
2. A designer’s ability to figure out what environmental limits mean from a receiver design perspective
3. The ability to test a device to ensure that it really works in the specified environment and to convey these test results to the FCC to demonstrate the adequacy of the design
4. The regulator’s ability to field an economical approach to resolving issues when a receiver does not work and the manufacturer says the environment is the issue

Markets for adjusting interference rights

One of the problems I tried to solve with the proposal “Stamps and Stewards: A third way to regulate radio operation” was negotiating adjustments to boundaries (e.g. power levels) between unlicensed bands and their neighboring bands given of the collective action challenges faced by unlicensed operators. There are other possibilities; this post explores using auction mechanisms.

Is 2.4GHz Wi-Fi the next GPS/LightSquared?

No, unlicensed devices in the 2.4 GHz band (2400 – 2483.5 MHz, operation under Part 15.247) probably won’t be the next GPS/LightSquared, where a large installed base of unlicensed devices with significant susceptibility to out-of-band interference was threatened by the deployment of a cellular service in an adjacent band. However, some similar characteristics raise concerns: tens of millions of devices, poor adjacent channel rejection, and a quiet band next door. What would happen if there were a large cellular deployment next door to 2.4 GHz?

There are significant differences to temper concerns: Wi-Fi devices don’t depend on such exquisitely low signal levels as GPS receivers; we’re not talking about safety of life applications; the RF front-ends of Wi-Fi devices are not open many tens of MHz away from the allocated band; and there already is some cellular operation nearby, at least in the US (Clearwire/Sprint’s 4G service in the 2.5 GHz band).

Still, as I’ll argue, the fact that interference has been observed between 4G service in 2.5 GHz and unlicensed devices in 2.4 GHz even with at least 10 MHz of guard band between them suggests that we’ll see interference problems to and/or from unlicensed devices if a cellular service were allocated in the fallow 2360 – 2400 band. That in turn suggests that it could make sense for the FCC to start encouraging or mandating better filtering for unlicensed devices over the 2.4 GHz band now, well before the 2.3 GHz band starts being populated with a potentially interfering service.

Update 4/23/2012: Monisha Ghosh kindly let me know that  2360-2400 MHz has been requested by healthcare device manufacturers (GE, Philips etc.) for Medical Body Area Networks on a secondary basis to Aeronautical Telemetry (OET proceeding 08-59). The June 2009 Notice of Proposed Rulemaking (PDF) provides good background information on the current uses of the 2360-2400 band.

Drones, the RF noise floor and encouraging directional antennas

According to a story in the WSJ, Silicon Valley techies are building drones for $150 to $1,000 each, and some of them provide live video feeds. The story doesn’t specify the wireless channels that are being used, but I’d guess amateur bands and 2.4 GHz or 5 GHz unlicensed.

 

The impact of broadband unlicensed transmitters up to 400 feet (the current limit for amateur operation) in the air on the RF noise environment is likely to be significant once a significant number start flying. As Carl Martin mentions in the comments to the WSJ article, applications include real estate videos, movie shoots, police monitoring and local small parcel delivery (can you say pizza?). The commercial/professional impact will be even larger once the FAA develops rules for the licensing of commercial drones by 2015 as it has been required to do by Congress. Even now, the there is a surprising range of uses; according to data released by the FAA to the EFF following a Freedom of Information Act request, drones are operated not only by the Customs and Border Protection, DARPA and branches of the military, but also by universities, police departments, and small towns and counties across the United States.

If these live video feeds are transmitted from omni-directional antennas, the signal will cover huge footprint, potentially interfering with many other devices. I wonder whether more should be done to encourage the use of adaptive antennas by unlicensed devices so that transmitters only deliver their signal where it’s needed. As I understand it, EIRP transmit power limits effectively put a ceiling on power delivered in any direction; the rules don’t contemplate allowing transmitters to concentrate their power using antenna gain. However, there could be a benefit to allowing transmitters to deliver a higher amount of power in a small area; yes, the interference there will be greater, but fewer other players will be affected.

There are constraints, of course. The directionality is a function of the antenna array size as well as processing, and even a spot beam could be quite wide at a range of a few hundred meters. There’s also cost; how does one encourage drone wireless links to spend the extra money?

Coda on Privacy.  The EFF FOIA request to the FAA mentioned above seems to have been prompted by concerns about privacy and surveillance. The obvious application is video surveillance, but intercepts of wireless networks are of course possible, too. If Google Street View could amass a trove of passwords, Internet usage history and other highly sensitive personal data by driving around, imagine how easily such an operation could scale with an airborne platform. I wonder, though, if it doesn't get harder from a drone: how well could a receiver disentangle the myriad signals from all the base stations it could see? Not too hard, perhaps, since they're not all operating simultaneously.

Interference protection for unlicensed devices

It’s generally accepted that an unlicensed device must accept interference from any source, and may not cause harmful interference to any licensed service; for example, Weiser and Hatfield explain that Part 15 “safeguards only the rights of licensed spectrum users and provides no protection to commons access users”  in their 2005 paper Policing the Spectrum Commons.

I don’t read it that way. Unlicensed devices shouldn’t cause harmful interference to other unlicensed devices, either.

Receiver interference limits presented at FCC workshop

I made a presentation and participated in a panel at the FCC Workshop on Spectrum Efficiency and Receivers on March 12-13, 2012 (Public Notice PDF). My presentation is at http://sdrv.ms/AwoAZj. My presentation is on the day 2 video starting at time code 36:00 (there is also video for day 1); my opening comments on the subsequent panel are at 86:00, and various times subsequently.

A two page summary of the proposal is available at http://sdrv.ms/ReceiverLimits 

Placebo Legislation: Doing Good by Doing Nothing

I recently sat in on a conversation at the Silicon Flatirons Center among a group of cybersecurity experts about the need and nature of government regulation, given the vulnerability of critical infrastructure to cyber-attack, and indications that the market alone may not provide the desired degree of communal action needed for defense. I concluded that the answer was “placebo legislation”: Doing Good by Doing Nothing.

Apparently the U.S. federal government has recognized the need for action, and the Administration and leaders in Congress have proposed a number of legislative solutions. However, I ended up feeling that we’d be better off if the government pretty much did nothing given that:

A number of practitioners described how regulations too easily lead to a “compliance culture,” where statutory requirements incentivize companies to check the right boxes, rather than implement better security

White hat hackers explained that vulnerabilities and exploit strategies are evolving much, much more rapidly than any conceivable government framework

Cybersecurity is a complex and messy socio-technical system, and the unintended negative consequences of any regulation could easily outweigh the benefits

A leading computer scientist observed that the incentives for appropriate behavior were already about right – and that a process-oriented standards framework was guaranteed to drive the technical experts out of the room.

The government therefore needs to be seen to be doing something – but the less it does, the better. This is not unlike the placebo effect, where a patient given a dummy treatment experiences a true improvement in their condition – without harmful side effects.

Incremental management of reception: When protection limits are not sufficient

As the growing demand for wireless services squeezes radio operations ever closer together, we can no longer afford to ignore the costs that poor receivers and ambiguous interference standards impose on society. There’s a growing consensus that radio regulation needs to attend to reception issues as much as to transmission, which has led to a clamor for receiver standards.  As I’ve argued in the June 2011 post Receiver protection limits: a better way to manage interference than receiver standards, however, the best way to manage receivers is to specify the radio environment in which they have to operate (i.e. receiver protection limits) rather than government getting into the minutiae of setting performance requirements (i.e. receiver standards).

However, while protection limits are necessary, there may be cases where they’re not sufficient. In this post I outline a progression of increasingly interventionist steps in managing reception, starting with protection limits and adding more and more requirements until one reaches full-strength government-imposed receiver standards.

Vendor representations: a solution to the decoupled receiver problem

Requiring receiver vendors to represent to buyers that their equipment is fit for purpose is a way to avoid cheap receivers from reducing the performance of coexisting systems in the “decoupled receiver” case, i.e. when there isn't a license holder to negotiate with.