Reed Hundt's article on the benefits of municipal wireless

Registration required on that direct link. No thanks. I was able to get to it without registration from Google News

or . In part:

Congress should grant $1 billion in federal matching grants to any municipality that will pay 50 percent of the cost of such a local wireless broadband network. Local government should let competitive contracts and build city-by-city, county-by-county, coast-to-coast WiFi network.

Incredibly bad idea.

"Officials ought to reallocate a spectrum, probably in the 700 megahertz band, for a national wireless network reserved for first responders. The local WiFi networks can be used by anyone with a laptop. The first-responder network would be available only for authorized emergency services."

Um... that's what they've allocated 4.9 GHz to.

Also the way this article uses the term "mesh" is incorrect.

Steve

4.9GHz is where Motorola and others are trying desperately to prevent customers from adopting non-proprietary (802.11a) standards. See any 802.11a mentioned here? 700MHz is worse because even acts of congress and major FCC concessions cannot seem to move the broadcasters off the frequencies. Incidentally, depending upon whom you discuss the issues, "wireless data" at 700MHz often means Project 25 data at a fabulous 3600 or 9600 baud. Whoopee.

Actually, he understands mesh quite well within the frame of reference of when the FCC first mis-allocated the 700MHz band in early 2002.

No way. The whole idea behind allocating 4.9GHz was to allow the use of modified 5.7Ghz 802.11a hardware on 4.9Ghz using coordination and licensing to mitigate interference. Nobody even suggested a wide area wireless municipal mesh network on 4.9GHz in the original proposal. What we're seeing are vendors trying to lock large government equipment anti-terrorism funding into proprietary corners.

I'll spare you my rant on such a technical "solution" looking for a problem to solve. Do you remember what problem 4.9GHz was originally intended to solve? It was a rush allocation by the FCC following the WTC bombing as a solution to a rather disgusting inter agency communications interoperability problem. One doesn't do that by pushing proprietary solutions.

802.11p is for data between high speed vehicles to fixed access points at 5.9GHz, not 4.9GHz. It is not a mesh network standard. I don't see why Firetide would need or want 802.11p for their mesh. Progress is currently on draft 0.23 which is very preliminary. It's been running for about a year and there will probably need to be some live testing before the vote. My most optimistic guess is 1.5 years.

There's a pre-standard product; it will become 802.11p, probably.

Steve

Yes, the "modified 802.11a" you're talking about, formerly called "WAVE", is now called 802.11p.

I only say "probably" in the above because it's not yet adopted by IEEE.

N obody even suggested a wide

Agreed.

There may be some of that going on, but mostly it's pre-standards confusion, not a deliberate attempt at a lockin of non-standards.

Steve

Can't be done. However, we're getting off the subject. You stated that the Firetide mesh product that being advertised for the 4.9GHz band will eventually mutate into 802.11p. I stated that there's no connection, relation, or reason for Firetide mesh to have anything to do with a completely different technology around 802.11p. There's absolutely no connection. The 802.11p topology was suppose to be a moving mesh network but has more realistically slithered back to fixed access points and moving vehicles with no attempt to play mesh network between the vehicles and the roadside access points. With the extremely short times allows for transmission, 802.11a timing just isn't going to work. It won't be an adaptation but a total redesign.

Sorry. My perception of the situation is far more conspiratorial. In a past life, I dealt with quite a bit of FCC related issues. Although I'm far out of the loop these days, my experience showed that when the big vendors proclaim "open systems" or "interoperability", they never seem to quite deliver. I have lots of examples, but one of my favorites is that Motorola doesn't even pretend to be interoperable with their own equipment. Every new model requires *ALL* new accessories and batteries. Even the power connector is new requiring a vehicle rewiring. I don't think there's an antenna connector available that they haven't used. The mics are similar, but the connectors are all different. When they accidentally use a standard connector (i.e. RJ45), then they make sure that the pinout is completely different from other manufacturers and their earlier products. Many more examples if you want them. Not only is the proprietary design deliberate, but it's institutionalized by company policy.

I agree, there's no way an 802.11 MAC will run on top of the

802.11p PHY, and in that sense it will be a less-standard standard.

I do not know if Fireside uses some earlier version of that standard -- WAVE has been around for a long time, it was originally controlled by SAE ... yep, the same people who standardize motor oil.

In any case, yes nothing sold now will interoperate with anything that may be standard in the future, unless someone deliberately puts in back-compatibility.

Steve

I'm trying to visualize what TCP/IP would have looked like if DARPA had adopted the same philosophy. Back compatibility with what? The technology lifetimes are so short that it's literally not worth the effort. Incidentally, wanna try using my 802.11 (1-2Mbits/sec Teletronics PCMCIA cards? It won't work with about 1/3rd of the hot spots I've tried and does a fair job of monopolizing the ones that allow it to connect.

If you do consider backwards compatibility a desireable feature, then make it optional with a switch. Eventually, you just turn it off and leave it off. I just turned on "802.11g only" mode on a clients three access points because all the 802.11b client radios had finally been exterminated. They noticed an immediate improvement in general performance.

If the government really wants inter agency communications interoperability, they have to: 1. State that this is their intent and goal. (done). 2. Establish a shopping list of functional requirements. (not done) 3. Establish a standards selection process and deadline to insure that at least some thing useful will be done. (not done) 4. Establish a testing mechanism and criteria for standards compliance. (not done)

This is roughly the way TCP/IP, FIPS, POSIX, (not SVID), were established. In my never subtle opinion, either the government isn't serious about interoperability, is completely clueless about how it is going to achieve it, or both.

Actually I've seen (pretty large) government RFQ's for 4.9 GHz emergenry comm gear. This was about two years ago. I don't know who (if anyone) responded but Motorola would be a reasonable guess.

Eeeeek.

Do you really want the feds to be triving to drive the standards process?

(Actually, the feds to this, for Military standards... but not for anything really used commercially, except by accident.)

Again they do this for mil specs.

You will recall, in the early 90's, there was a fed-wide move towards COTS -- commercial off-the-shelf euquipment, standards, etc. No governmental efforts unless it was military or intelligence.

Steve

I haven't seen those so I don't know what's in them. Over the years, friends working for various companies have sent me bid specs involving RF and ask me the traditional "What are they asking for"? The problem is that in order to be universally fair and equitable, the specifications for wireless system cannot follow a specific vendors product specs. As a side effect, they therefore tend to be rather vague. Unfortunately, even good ideas get overdone and non-vendor specific turns rapidly into incomprehensible or just plain vague. Over the years, I've seen a few elaborate wireless government RFQ's, that either would not work as specified, require major R&D to deliver a working device, or are borderline science fiction. Since the FCC has only recently started type certifying 4.9GHz hardware, I can only speculate as to what the government RFQ was expecting.

Incidentally, I regularly read First Responder, MRT (Mobile Radio Technology), Mission Critical Communications (Radio Resource) magazines and others. Lots of stories about possible 4.9GHz applications. Up until recently, the major application seems to be video from either aircraft or surveillance devices. Lately the hype has drifted into building municipal mesh networks. Interoperability is rarely even mentioned. The few users in the business that I've asked about what they plan to do with 4.9GHz seem to think that it would nice to have a secure data link for smart hardware and wireless laptops for filling out forms. It seems like a solution looking for a problem to solve, but I may be talking to the wrong people.

Read what I wrote. I said "standards selection" not creation. That means they go shopping for available commercial standards necessary to deliver a workable product. For example "must comply with IEEE-802.11a-1999 and such". After all, the feds are both the biggest customer and the source of funding for the smaller municipal customers. If they're going to burn my tax dollars on this joke, they might was well get something that has a chance of working without spending its days occupying a warehouse.

The radios I used to design were full of MIL Spec components. There were quite a few differences between commercial components and MIL spec. Most of it was in the testing necessary to insure that they met the specs. The coax cable in your wireless contrivances is all references by Mil designations (RG-XXX and Mil-C17-XXXXX). As the military was buying more off the shelf commercial hardware, what really happened was the elimination of the compliance testing component. The military was more than happy to get the Mil Spec components they had been used to buying as sky high prices, but as commercial prices without the testing. Kinda like "kosher style" food. Same stuff but without the rabbi's inspection and blessing.

Actually, they're suppose to do that for literally everything they buy, from pencils to satellites. This step is literally the key to the whole exercise. A dozen companies can build 4.9GHz interoperable hardware that meets some federal RFQ specs. But, it's not really interoperable until someone actually tests them for compliance and functionality. Three examples of a 90% job that could have been 100% if someone had established a testing mechanism were, WDS, NWAY ethernet negotiation, and VPN interoperability. In fact, both of these are still problems today because specs and testing vaguely defined. If you want it to work, you gotta test it to be sure it's gonna work.

Yeah, I guess. Just have the city drive down to Fry's and buy a

4.9Ghz wireless system. Actually, that was the original intent and why the FCC released 4.9GHz from their clutches for the purpose. One would expect that all the 802.11a vendors would make a 4.9GHz model. In sufficient quantities, it would probably be available off the shelf from the usual emergency responder vendors.

I may be reading this wrong, but why is compatibility such an issue if we are talking about systems that are going to be primarily used by emergency agencies and so on? Isn't the history of that sector pretty much filled with proprietary solutions? Would you really want to make it trivial for someone to modify a consumer product in order to snoop on law enforcement and emergency-vehicle radio transmissions?

I see the potential benefits.

On the other hand, I guess I'm one of those skeptics that finds a certain degree of comfort in the fact that states and municipalities don't all operate as if they were one homogenous mass. From a civil liberties point of view, the idea of "National ID cards" and national citizen databases of various forms aren't too appealing.

If all the law-enforcement and emergency agencies end up having the same basic communications equipment, my cynical mind starts to worry about whether it doesn't pave the way for centralized, national control over all of these agencies.

To allow inter-agency operation, or to ensure that a system could be expanded with more units in the future and still be compatible.

Recall the 1991 Oakland fire, wherein fire crews from different jurisdictions found that, in many cases, their hoses would not fit onto the fireplugs because fireplugs are non-standard.

Steve

The original inspiration for 4.9Ghz was the muddle at the World Trade Center, where different agencies could not talk and coordinate with each other. 4.9Ghz was somehow intended to give law enforcement a common datacomm system. 700MHz will eventually do the rest. Here's Michael Powell's pitch line before a Senate committee on the grand plan for emergency services. |

"This spectrum at 4.9 GHz is part of a transfer of Federal Government spectrum to private sector use and will accommodate a variety of new broadband applications while also fostering interoperability." Note the word "interoperability" which means that anyone can talk to anyone else. It would be a fair assumption that such broadband applications would be standards based and not some proprietary protocol fabricated for the sole purpose of restricting competative bidding.

Absolutely. Motorola leads the industry in insuring that absolutely nothing will talk, communicate, or plug into other vendors hardware (or their own previous models). In addition, great efforts are made by the frequency coordination organizations (APCO) to prevent interoperability in the name of interference reduction. Even the FCC has weighed into the interoperability prevention business by refusing to type certify field programmable radios. While over the air programming is allowed for Motorola SecureNet operation, this does not include juggling frequencies and channels remotely. A good example of how an agency responds is the current use of Midland 256 channel radios by CDF. They need all those channels in case the unit has to move to another area. I've been told another 256 channels would be handy. Basically, all the tools for interoperability are available and functional. What's missing is the removal of bureaucratic impediments, standards, and the necessary relaxation of rules-n-regs.

It's more than just snoop. It's also trivial for someone to hot wire a microwave oven and turn it into an area wide jammer. If you dive into eBay, you'll find a fairly large collection of used public safety radios suitable for constructing a monitor or jammer. The FRS, GMRS, and MURS rules prevent combining these channels into public safety and commerical radios insuring a market for both. Never mind that much of the short range communications in Iraq is still being done with cheap FRS radios because the fancy military stuff didn't work as expected. In other words, the rules are intentionally restrictive in areas of cross service interoperability. Wouldn't it be nice if your cell phone had a "walkie-talkie" like function (such as FRS) that didn't require the benevolent involvment of the cellular provider?

Interestingly, the major reason that law enforcement doesn't use encryption as much as one would expect is that they also use a huge amount of cheap consumer scanners and commercial hardware that does not allow for encryption. It's also an expensive option. The issues are commonly discussed under the topic of "scanner law".

[excellent comments on interoperability and proprietary vs standards-based technology within the emergency services sector]

So here's an interesting alternative - it may have most of the benefits without most of the problems of radio standardization, and some interesting "goodies" of its own:

Well, sorta. But the real crux of this seems to be the ability to receive and transmit to all these disparate endpoints and merge everything together into a homogenous communication channel.

|

Those Telex Viper and Sharpcom thingies look interesting, but the limitation seems to be how many "ports" you have connected to the command center for each disparate system.

Nonetheless I guess that these things are more-or-less precursors to the Cisco product. It also makes me wonder - if these devices are really in use out in the field, then what's the beef when it comes to interoperability in a big disaster? Once again, it seems that the limitation is that you can only funnel a limited number of channels through the "big mixer". Perhaps some other geographic issues too, like having to backhaul a conversation from a fireman and a cop standing 1000 feet apart from each other to a command center 100's of miles away, for example.

It looks like the Cisco thing will have a lot more bells and whistles, not too surprising there. In a way, it's rather philosophically compatible with Cisco's roots - designing multiprotocol network routers that understand all the various protocols in deep detail and getting them all to co-exist and run over the same infrastructure.

I would agree there.

I dunno. I just think it's a lot harder to implement than something that assumes that you've got working towers to communicate with. The options are far more limited, especially for long distances.

Why? Because it would encourage agencies to stick with their existing radios? That's true, but then again, I don't see why you couldn't plug-in new-fangled infrastructure-less radios into such a system like anything else, so it could offer a certain sort of "bridge" in that respect as well, from old to new.

Basically, that's a pitch for VoIP for dispatch. That's already being done with a wide variety of related products. One part that's conspicuously absent interfaces for Nextel, Telco, POTS, lease lines, P25, conventional FM, and other forms of incompatible communications. There are products that glue these together which include VoIP.

|

| However I have a different view of emergency communications. It largely reflects the thinking of Andy Seybold at: | a real disaster, the infrastructure always breaks down. The more extensive the disaster, the more extensive the damage. If communications is reliant on infrastructure such as cell sites, repeaters, internet connections, and wired backhaul, then it will fail. What's really needed is infrastructure-less communications. That's something the major manufacturers detest because it frees the customer from their proprietary clutches. In this light, I see the Cisco emergency "system" as a giant step backwards.

Satellite communications may be your best bet here, or, possibly, HF although that doesn't scale as well. These two have been favorites of military communicators for decades. Very few natural disasters or human adversaries have the ability to knock out a satellite (although ground stations are potentially vulnerable), and disrupting HF requires a procedure known as "pumping up the belts" which involves the use of nuclear bombs.

Steve