Editor's note: Getting more spectrum is one of the two biggest issues facing utilities, Connie Durcsak, CEO of the Utilities Telecom Council (UTC), said recently. (Enhancing cybersecurity is the other.) Durcsak rightly pointed out that distribution automation applications require a lot of redundancy and have low tolerance for latency. Substantially more bandwidth is required because “it’s about transporting a vast amount of data needed to manage this complicated environment and we have gone from one-way communications to multi-way,” she said.

 

UTC is pursuing one partial solution – namely, getting utilities included in plans for the nationwide public safety network, a broadband network that will operate in the 700 MHz band. But when a small company wrote me with another solution, I asked CEO Brian Andrew to contribute the short article below.

 

Brian had previously submitted a piece asking whether free spectrum can deliver the "ultimate" network. He is not without bias, since his company is one of the few that can perform the trick described below – namely, aggregating several small channels to create the equivalent of one big one.  Whether or not you agree with this particular solution, I wanted to be sure you were alerted to the problem. Take a look, and use the Talk Back form to comment on its potential. – Jesse Berst

 

By Brian Andrew

 

Every utility I speak with is concerned about spectrum availability and achieving higher bandwidth throughputs. Today, many utilities are experiencing interference with their unlicensed installations and/or are researching or investing in their own spectrum to add to their capacity.

 

The issue presented with purchasing licensed spectrum in the 200 or 900 MHz frequency range (besides cost) is that they are typically small chunks from 25-50 kHz and are rarely, if ever, contiguous. 

 

The aggregation solution

25 kHz will give you about 20 Kbps throughput and 50 kHz top ends around 40 Kbps.  Multiple networks of 10-40 Kbps throughput may not meet the need.  For higher bandwidth throughput the only solution is to aggregate (bring together) those non-contiguous narrow band channels. 

 

Aggregation works great whether you have purchased spectrum that is noncontiguous or are thinking of utilizing FCC Part 90 secondary licenses.

 

For those of you not up-to-speed on Part 90 secondary licensees; you get protection from any secondary license holders that are licensed after you. First in gets priority. So the message here is “GET IN EARLY”.

1

Spectrum on a secondary basis

If you investigate, I expect you will find there is plenty of available spectrum in the 217-220 MHz range to use on a secondary basis. This can be beneficial if you have not purchased any spectrum or you have “some” spectrum but not enough. If you don’t have enough you can supplement your purchased, licensed channels with secondary use licenses.

 

For instance our Parseous “Aggi” radio (Aggi as in aggregating) in the 217-220 MHz frequency range, aggregates up to 40 non-contiguous channels (6.25 kHz wide) to dynamically deliver up to 200 Kbps throughput in an on-demand basis. There are 240 12.5 kHz channels in the 217-220 MHz frequency range for the radio to select from.  If you own spectrum, you can set your channels as priority channels to be used by the radio and increase bandwidth throughput capacity by adding in channels on a secondary basis.

 

To operate under the secondary licensing rules, the radio must dynamically stop using that frequency and automatically switch channels if another licensed user with priority is sensed, if necessary to avoid interference and eliminate any service interruption.

 

Consider channel aggregation to increase throughput. In addition, consider radio management of time, space, frequency and power to improve reuse and eliminate interference in a high-density network.  A frequency-hopping technique is a good choice to minimize the amount of time that any frequency is utilized. Radios should also dynamically regulate their transmit power so they utilize the minimum amount of power needed to establish solid communications.

 

Brian Andrew is the CEO of Parseous Systems, whose radio systems are deployed for power utilities, water municipalities, farming, security and government and military installations.

Talk Back to the Author

Current Comments (11)

Leave a Comment

---

We need every trick we can find

As an industry, we are just started down the path to automation and communication with devices. I would argue that what the communications needs are in 2050 will not resemble what we are thinking about today. There is no question spectrum is an issue, and it will only get worse over time as phones and other devices demand spectrum too. The governments will offer spectrum to the highest bidder, not the highest priority use. This will mean that as a industry we will have to find every trick in the book. I suspect that substations will all eventually have to have fiber installed to move enough data security - but that is an eventually - right now wireless communications will work fine for what we want to do. The important part in my mind is to think ahead and plan ahead for our future needs. To communicate those needs early and often to people who control spectrum allocation and promote new ideas on how to get more from less spectrum. I know, this sounds like motherhood and apple pie to most of us, but we don't seem to act on it very often.

Doug Houseman - 09/13/2012 - 06:33

900 MHz has large contiguous blocks

Narrowband PCS (NPCS) is 3 MHz of licensed spectrum governed by part 24 rules. There is 1 MHz at each of 901/930/940 MHz. Licenses at auctions 40,41 & 50 were in blocks from 50-200 kHz and many available nationwide. Aggregation has resulted in contiguous blocks of 100, 300 and 400 kHz being available to the utility industry and a significant amount is being used for AMI, DA and SCADA applications. Radio throughput from large and small vendors start at 64 kbps in 50 kHz and in a block of 400 kHz is significantly higher.

Jim Wiesenberg - 09/13/2012 - 07:37

Yet another reason for Power Line Carrier

European Smart-Meters use Power Line Carrier. This spectrum issue provides yet another reason to question why we aren't using it here in the US as well. Finally, Utilities continue to underestimate the public backlash against wireless. The longer we ignore it the worse it will become.

Guy AlLee - 09/13/2012 - 07:51

wired devices need wired communications, period

Wireless should only ever be a backup mode of communication on a power grid. Wired devices need to be on a wired communications network with infinite scalability and no interference with other necessary mobile use. In a wide scale emergency, the wireless spectrum is all going to be required for many other purposes, routine electric grid information shouldn't be there crowding it out and slowing it down. What Chattanooga TN, Bristol VA and other utilities that installed fibre to every building did, was the right solution. Avoid the wireless mess. Grab the revenues available for being first with fibre. Make sure that the fibre network is utility-grade (which it won't be, if a telecom company builds it). There are many ways now into the building without fibre, IEEE P1905.1 establishes a way for various powerline (IEEE 1901), coax, cat3, and conventional ethernet and wireless connections to co-operate to get perhaps as much as 3-5 gigabits into a building on existing copper wire plus the latest -n and -ac wireless short range. There are also some new technologies that use MV cables between poles to get fibre-like bandwidth. Until those mature, fibre to every transformer ought to be the rule and anything else ought to be viewed as a temporary solution. Consider appliances. Are we going to be adding a wireless mechanism to every single toaster oven or pump, or are we going to use the existing AC wire and IEEE 1901 to carry the signals to home control and monitoring systems? Consider electric vehicles. SAE J1772 plugs specify powerline networking before they begin to charge. Most utilities are not deploying now but waiting - and one of the things they are waiting for is the IEEE 1901 standard to be met by meters and gateway devices. Already, you can buy from consumer router companies like D-Link, devices that do -n, 1901 AC powerline and ethernet. All 1901 devices sold now have AES-128 encryption identical to that required for meters by the US government. Mobile devices including vehicles and worn devices are going to proliferate. The amount of emergency and diagnostic data that needs to move wirelessly in an outage or crisis is going to go up. The more routine data from wired devices that is on wires, the better. So wireless ought to be viewed only as a backup that we are installing first as an interim solution awaiting utility-grade wired fibre and AC powerline networking which will handle the big data. And will pay for itself.

Craig Hubley - 09/13/2012 - 07:56

Cart before the horse

Once again as industry we fail to understand the question. The needs for spectrum or bandwidth (or latency or reliability or redundancy; we should not mix these up) by utilities are not singular, therefore a single solution is not likely to suffice. We need to understand the current paradigm (metering, automation, physical security, corporate traffic etc), and find the solutions that meet that need. It is also important (as Doug mentions) to look to the future requirements; however, that may not be practical in the short term. Ideally, fibre to every device would be the solution, but it is cost prohibitive. I expect most places that put in fibre to the premise did so with a business case that was not solely electricty delivery. I suggest we complete the effort of defining the transactions and their communication requirements. We can then map those requirements against the transport options. Even then it will not be one size fits all but at least every utility would have the inputs to decide what would work for them today.

Eric Mewhinney - 09/13/2012 - 09:21

fire all utility execs who say

"Ideally, fibre to every device would be the solution, but it is cost prohibitive. I expect most places that put in fibre to the premise did so with a business case that was not solely electricty delivery." The phrase "cost prohibitive" assumes that a utility continues in the old sell-watts-only business model. That is no longer a viable option as it's a conflict with the businesses that monitor and reduce watt consumption. Yes, of course "most places that put in fibre to the premise did so with a business case that was not solely electricity delivery." That's exactly my point. The advanced monitoring, improved safety and (most important) access to telecom/TV revenues *is* the business case. A utility that refuses to pursue these at all is refusing to compete and dooming its customers to backwardness, lack of services, lack of safety and ultimately must charge the full cost of the network to those customers. Who should reject that utterly, ESPECIALLY AT THE REGULATOR. All utility executives who refuse to pursue non-watt revenues and try to preserve their existing industry model must be viewed as trying to continue to sell more watts and inhibit conservation while sticking their customers with more costs than necessary for a one-purpose network. Shareholders of private utilities should demand that such executives resign or be fired. Voters with public utilities should raise questions of public safety, economic development or its lack, and the negative impact of the old electricity-only and telecom-only incumbent service model on rural areas. This is not 1960. Time for a ruthless elimination of the more-watts low-rates bill-everything-to-the-ratepayer-without-seeking-other-revenues executive. Certainly when new public power districts and co-ops are formed, as most states in the US allow, they do not leave home monitoring, security, vulnerable persons monitoring, TV or telecom revenues on the table. They pursue those revenues. Often, with aggressive overprovisioning of fibre. Chattanooga TN and Bristol VA proved that even with limited take rates on the fastest (gigabit) fibre optic services, they were able to reduce outages by 40%, reduce maintenance costs, and pay loans back in record time. Plus attract new investment and build capacity in strategic industries. They did this by *discarding* the sell-watts-only and one-purpose-network business models. That should be writing on the wall to all pre-fibre executives.

Craig Hubley - 09/13/2012 - 11:10

smart use of avaible spectrum

Using spectrum allocations for multiple applications provides a smarter way to use what is available. It seems that presently each application is demanding its own special spectrum allocation. There are opportunities to leverage the spectrum available for multiple applications, but that needs greater smarts on the bearer channel, and individual vendors have little incentive to achieve this. Who is doing an inventory of spectrum and applications so as to truly deliver a bearer technology that can cope with multiple requirements? (The NIST capability framework overlooked narrowband and also shoehorns applications to wireless technology, without considering bearer adaptation) Once a full inventory of what spectrum is available, utilisation factor and application priority it may well be possible to be smarter about using the resources available and avoiding worst case choices (such as unlicensed or secondary use). (Disclaimer: I do work for a wireless vendor and perhaps have a bias on solution. Pleased to be in the debate)

John McDermott - 09/13/2012 - 13:17

Re:Fire execs

CraigI must respectfully disagree with you. The cost of fibre being too high is not related to a Watts only sales model. However, it is related to an electricity only model. Many utilities in North America are restricted by regulation from selling non-regulated services to their customers.Beyond that, even when a utility is not prohibited from offering other services, such as TV or broadband internet, the tariffs for cost recovery of power would not, and should not, support 100% of the costs, as has occurred in some jurisdictions. At the same time, i doubt many protection engineers would support SCADA signals being paired up with TV signals on the same network.It might be nice to see some innovation on the regulatory side of the electricity business, but in my opinion, under the current system, roll-out of fibre must be done with prudence.

Eric Mewhinney - 09/13/2012 - 13:27

Spectrum Crunch

Spectrum is definitely a key issue for utilities as the proliferation of IEDs beyond substations continues. Depending on your region/regulations, the deployment of IEDs may come quicker or slower. I believe California will be on the fast pace as they have a great deal of mandates to deal with that will require increased automation. Canada realizes the importance of spectrum for utiltiies and as such allotted them 30MHz to work with. There's a great deal that can be done, and is being done, with smaller chunks of spectrum but depending on the evolution of business needs, these smaller chunks may quickly become clogged up. At the rate commercial carriers are purchasing spectrum, once these small chunks are maxed out, there wont be many options other than to rely on commercial carriers (I'm sure they'd love that). Some may see commercial carriers as a positive way forward, and for some applications that is probably true, but when it comes to grid operations, utilities should be careful in just how much trust they place with them. If they aren't, we may be looking at a future blackout of epic proportion that just might push congress, or the FCC, to wake up and realize just how critical communications will be to the future of the electric grid.

Kyle Cormier - 09/13/2012 - 15:57

Why not more fiber optics?

Why not integrate more fiber optics at distribution level? The data transmission problems, security, reliability and speed could be more efficiently addressed if utilities focus more on fiber optics communication at the distribution level (good point by Doug Houseman). But this doesn't mean that they should do the wide implementation right now! Some distribution network operators in Europe have been already putting tubes for fiber optics next to their new MV network cables for years and the price of tubes alone is really negligible in the complete project price. Looking on the lifespan of distribution network, this can be a good best practice also for other network operators. Well nobody knows what bandwidth we will need in 30-40 years, but if you can make your network easy upgradable for super-fast fiber optics communication at distribution level (reaching your MV/LV substations), this might pretty much cover all your communication needs during the lifespan of the distribution assets. But the distribution network operators considering various communication options shall better consider this option now, when the additional costs are negligible compared to the cost of similar communication infrastructure in the future (which might be prohibitive expensive). Then you might end up with less reliable solutions lower bandwidth and even maybe at much higher cost. Lesson learned outside distribution business: for instance railways companies buried the tubes for fiber optics next to their new/upgrade rail tracks ages ago and nowadays they can enjoy super-fast bandwidth between cities they serve and rent remaining bandwidth to other companies.

Petr Kadurek - 09/13/2012 - 17:05

It's about wired or wireless... it's a mix and match ..

To me it's not about fiber or privat wireless or public wireless. Most likely a utility will need all of those technologies and amalgamate them to a smart communication solution. And that becomes even more a reality when you take a global perspective. All of the technologies above help the utilities to chose the right technology mix in their very specific environment constraint by regulation, geography, application requirements etc.

Ulli Feuchtinger - 09/14/2012 - 01:20

Leave a Comment

New to this Blog? Need some help?

Email: [?] We will send you an email with an appproval link you must click for your comment to appear.
Full Real Name:	[?] Real comments need a real name.
Subject:	[?] Please provide a short subject for your comment.
Comment:[?] Do not post commercial/sales messages here. They will be deleted. Try: pr@smartgridnews.com PLEASE NOTE:  All HTML codes will be removed from your comment.
CAPTCHA Security Code The numbers (0) and (1) are not used. [?] Please type the characters you see into the box above.
Remember my name? (Uncheck to forget.)
You can not edit or delete your comment once it is posted. Please check for spelling now. Your IP Address is: 23.22.212.158