Wireless Growth & Wireless Interference

windowpain writes "An article in Monday's Washington Post says "The explosive growth of the mobile phone industry has crowded and tangled the nation's airwaves to such an extent that wireless company signals are increasingly interfering with emergency radio frequencies used by police and firefighters, public safety agencies said." Wifi is not a problem, evidently. Understandable, given its short range."

Comments

How can this happen?

[rokzy]

if the FCC (or whoever) made reasonable specs and if everyone's following the specs how can this mess happen.

=> fire people at FCC or sue mobile companies.

Re:How can this happen?

[Anonymous Coward]

Because while most countries alot 50-70% of their frequencies to public and commercial use, the USA only alots about 7%. About 10% is used by Police and Fire Depts., and a whopping ~83% is reserved for the Military.

Re:How can this happen?

[Phreakiture]

NEXTEL is the prime culprit. The problem stems from what NEXTEL is.

a NEXTEL phone appears to be a cell phone, but legally it is a handheld, trunked two-way radio with full-duplex capability and access to a phone patch. While this accurately describes a cell phone, the evolution differs.

"Cell phones" that follow the CDMA, GSM, TDMA or even AMPS (analog) standards are all using standards that were originally developed for use as telephones. They are licensed as telephones, and use portions of the spectrum that are reserved for telephones.

Cell phones that follow the iDEN standard (this would be NEXTEL) use a protocol that evolved from a half-duplex digital trunked two-way radio system.

What NEXTEL used to be in the business of (under a different name, which I can't recall at the moment) was providing nationwide fleets with trunked digital two-way radio service. Someone in NEXTEL had this brilliant idea that if you could add full-duplex capability and a massive banked phone patch that you could compete with the cellular companies, and this is what they did.

The problem with this is that NEXTEL, being an operator of a Commercial Land Mobile Radio Service (CLMRS), gets to license portions of the spectrum intended for Land Mobile Radio Service users.

...and there is the rub. Police, fire departments, ambulances, tow trucks, taxis, and just about every business except sometimes aviation and maritime businesses, are users of the Land Mobile Radio Service. Like NEXTEL, they license portions of the spectrum that are there for the Land Mobile Radio Service (there is no distinction in spectrum allocation between a commercial and a private LMRS license).

The biggest pain here is involved in that NEXTEL does not operate on the same frequencies everywhere. They go from location to location, licensing 5-20 frequencies in the 800MHz LMRS band (and I think sometimes in the 900MHz LMRS band) in any given location, but the frequency that they allocate in one city may be the same frequency that a police department in another adjacent city is using.

Yes, the FCC probably should have prevented this. However, their allocation scheme was good enough for systems that had a low duty cycle, and it worked for decades. Introduction of a high duty cycle system such as NEXTEL broke the system.

Incidentally, I strongly advocate that police, fire and other services should not move to 800MHz trunked systems if they don't have a serious need to. The fire department in my home town of Selkirk still uses the same frequency (46.06MHz) with no repeater, no trunking, nothing except for a remote base (which communicates to the three firehouses on 460.6375 and 455.6375MHz) and provides complete coverage for the fire district with minimal vulnerabilities, including immunity from interference from NEXTEL.

BTW, WiFi is not a problem because the frequencies are nowhere near 800MHz.

Understandable

[rednaxela]

"Wifi is not a problem, evidently. Understandable, given its short range." Yah, that and the fact that the public safety networks operate in the 700 and 800 Mhz bands, whereas 802.11a and b currently operate primarily at 2.4 Ghz.

Locally even worse...

[Shoten]

I'm not surprised it's the Washington Post who is reporting on this. I live in Washington, DC as a consultant, and travel a lot to other cities. I have had cell phones with three carriers, and with all of them I've had much worse trouble here than in cities elsewhere, without exception. On one hand, we have some serious density of equipment here, but on the other hand there are zones where one knows one will lose their cell phone signal; the worst two zones are adjacent to the CIA's facility and the Pentagon, for example. This sort of thing can't be helping the matter.

Re:Locally even worse...

[che.kai-jei]

"but on the other hand there are zones where one knows one will lose their cell phone signal; the worst two zones are adjacent to the CIA's facility and the Pentagon, for example"

yeah but can't you guess why that is?

Re:Locally even worse...

[Creepy Crawler]

I wouldnt be surprised for one bit if the GOVT blocked consumer celluar signals near crucial buildings.

Re:Locally even worse...

[H8X55]

yeah, i was going to mention the same thing. notice you never have good reception (or any at all) near military bases? I watch as i get closer to quantico, my phone slowly.... loses... signal.. and then as i pass it, and continue up the interstate it gets better.

Re:Locally even worse...

[Anonymous Coward]

Don't cell towers only have an approx. range of 1k? Once you take into account that most bases are huge empty area used for ranges and movment practice, if your a smart company wouldn't you set your towers so the signal ran out just inside the military base? Otherwise you would be building cell towers with large portions of coverage that don't produce income. I don't think many grunts yak on a cell while setting up an ambush.

Re:Locally even worse...

[H8X55]

you're right, but at the same time consider this. I live in fredericksburg, va, 50 miles south of wash. dc and 50 miles north of richmond. va. All three cities are connected by Interstate 95 and US Route 1. No matter what, if you travel from wash to rich, you will lose signal in military base areas. in addition, there are some very desolate spots in this 100 mile stretch where if you exit the interstate or travel too far off east or west of Rt.1 you will lose signal, but if you stay on 95 or 1, you will have a strong signal for the entire trip. Not so around quantico, i will go from four bars of service to one bar to none, in a matter of 5 secs.

Police chatter in the classroom

[awtbfb]

This problem has been around for some time in the hearing impaired community.

Re:Police chatter in the classroom

[antdude]

This is true for most cellular phones as well. I wear a headband conduction hearing aid. Very annoying. I haven't had any incidents with wireless network though.

some obvious questions that

[sbma44]

I don't know the answers to:

• If spectrum is reshuffled, could this be an opportunity for the wifi-friendly (but still evil) Powell to allocate more unlicensed spectrum?
• Alternately, does the emergence of the interference bogeyman in such a prominent publication imply a tough PR road for more free spectrum?
• Could this be an opportunity for the cell carriers operating in the 800MHz band to switch to different technologies, or roll out newer services?
• Could these sorts of problems be used to get some federal money thrown toward development of software radio? would multi-format devices help avoid this problem?

As the submitter mentions, these emergency bands don't appear to be abutting or affecting the 2.4GHz range, but I imagine these issues will be relevant as wifi expands and interference problems begin cropping up.

wifi = different spectrum, limited deployment

[aldousd666]

Most of the WiFi I know of uses 2.4 Ghz band (I know there are others out there though) And there doesn't seem to be a whole hell of a lot of it out there yet. Maybe it will one day be a bigger problem as it's deployment picks up. I'm looking forward to the seeded frequency hopping options rolling out in the long term, which won't cause very many people problems (at least for more than a fraction of a second) The frequency hopping method also adds the 'security by obscurity' slant to the signal as well -- if you don't have the seed and the algorithm by which the frequency is determined, you may have trouble sniffing out the data on the signal. All really neat stuff. I don't see the wireless stuff, at least the internet variety, as being all that useful, (pardon me all of you blackberry junkies) I think wifi has the solution to it all, conceivably (someday) even cellular could just use some of that bandwidth. It's just a matter of changing the infrastructure -- something that seems to take forever with folks like Verizon involved.

Exactly how crowded?

[Faust7]

As organized, the spectrum, which is a limited resource, simply can't accommodate everyone...

There are several wireless companies operating in the 800 megahertz band...

How finely split can the spectrum be before there's danger of overlap? Is it possible for, say, one phone to send signals at 800.0001 MHz while another does so at 800.0002 MHz? Where is the precision cutoff for neighboring frequencies before things start to interfere?

Re:Exactly how crowded?

[aldousd666]

the cutoff you describe is actually dependant upon the ability of the receive to discern between the signals. Has absoulutely nothing to do with anything on the broadcasting end. More expensive receivers may be able to tell the difference between 800.0000001 and 800.0000002, but you regular cell might static up from both of these trying to receive on 800.

Re:Exactly how crowded?

[mt2mb4me]

Well on top of that 800Mhz is for "cell" technology. the Digital, CDMA, TDMA, and PCS are all in the 1Ghz+ range, I can no longer remember, I used to see cell ohones when digial was new. So maybe the awnser is to kill the old network, and offer free upgrades? to a digial phone to subsribers with an old brick for a phone.

Re:Exactly how crowded?

[Pedrito]

It's a combination of frequency, bandwidth, and a few other factors. For example, when something is at 800mhz, that's the center frequency. It will have a bandwidth associated that may be a few khz or a few mhz, depending on the application.

Then you get a combination of different technologies. Someone else mentioned TDMA, CDMA, GSM, and some others. Some of these operate in the same bandwidth without significant interference because the way they operate is so different. CDMA for example, doesn't suffer from the same kind of interference that TDMA, GSM and a few others do.

There's a technology called Ultra Wide Band that may take off over the next decade. While it covers a very large bandwidth (as the name would imply) that would normally interfere with other technologies, because of the way it operates, there's little, if any interference. The nice thing about it is, you can operate a number of different devices, each paid of sender and receivers has a "code" that allows them to communicate. Nothing else will interfere and many devices with different codes can operate in proximity.

Most technologies operate on the basic principal of sending a signal via the actual radio wave. They do this by modulating some part of the wave, usually frequency or amplitude (hence FM and AM radio).

UWB doesn't do that. Instead, it sends pulses very quickly (in picosecond lengths) and the pattern created by the pulses (think binary, on and off), are what are used to create teh signal.

Another advantage of UWB is that it works at a lower power (given the same range). Also, unlike the higher frequency applicatons (particularly when you get up to 5+ ghz), it doesn't have the same issues penetrating walls and such. Nice technology. Hope it makes headway.

Re:Exactly how crowded?

[mgs1000]

FYI

There are two sets of frequencies for cell phones in the U.S. The orginial band at 800 Mhz(actually uses 825-895Mhz, IIRC) and the "PCS" frequencies at 1900MHz(1850-1990Mhz, once again, IIRC)

You'll find CDMA and TDMA networks in both bands. GSM uses mostly just the PCS bands, and Analog uses the 800Mhz exclusively.

Re:Exactly how crowded?

[billy_troll]

that really depends on the bandwidth you need.
but, short answer, is no.

if you have a bandwidth of 1Hz, as you are suggesting, this is much too small to transmit any useful inf at any speed. for analogue voice, you need at least a bandwidth of ~30 Khz, and thats without the overheads of initiating communication, etc. now, GSM uses digital communication, which has a higher bandwidth, but means that you can multplex (which is essential, othewise you would just run out of spectrum).

add on top of that more BW for housekeeping and you can see why you are running out of space.

spread spectrum, now theres a thought.......

Re:Exactly how crowded?

[just fiddling around]

In fact, if you cut up the frequency in more chunks, you have to sacrifice the bandwidth of each chunk.

Each chunk (AKA band) of electromagnetic spectrum has a theoretical maximum throughput for information, which is related to the width of the chunk. Hence, we use the term "band-width" to describe the capacity of a link supported by electromagnetic phenomena.

Re:Exactly how crowded?

[dougmc]

for analogue voice, you need at least a bandwidth of ~30 Khz

No, you don't need anywhere near that much to send a reasonably clear voice. Police and HAM radios and such do it on 3-6 khz. Of course, you don't hear police radio advertising that says you can hear `a pin drop' either :)

(As an added bonus, they don't walk around going `Can you hear me now!')

Re:Exactly how crowded?

[billy_troll]

ok, sorry. normal telephones use 22KHz
thats what i was basing it on.

Re:Exactly how crowded?

[dougmc]

ok, sorry. normal telephones use 22KHz

No, it doesn't. A POTS (plain old telephone system) land line phone delivers about 4 khz of bandwidth.

This allows it to reproduce frequencies up to 4 khz or so ... your ears can hear much higher than that, but it's not really needed for speech.

Re:Exactly how crowded?

[KC7GR]

The closest channel spacing currently available in the civilian realm (I can't speak for the military, since their latest communications tech is classified) is 6.25KHz. The separation you describe is far too narrow to be practical because you're not taking the bandwidth required by the modulating signal into account.

Any time you modulate an RF carrier, you increase the total bandwidth or spectrum width required by said carrier. To use standard FM as an example, assume that Bw=bandwidth, Pd=Peak deviation (offset from the center frequency caused by modulation), and Fmax=the maximum audio frequency used to modulate the signal. Your formula is:

Bw = 2 (Pd + Fmax)

To clarify this further, let's assume you're talking about an FM broadcast station with a maximum transmitted audio frequency and peak deviation both at 20KHz. You're looking at 2x40KHz, or 80KHz bandwidth required. This means that your minimum theoretical spacing would be 80KHz, while minimum practical would actually be what the broadcast standard is today (100KHz spacing).

Detailed discussion of the above may be found at this link.

I'm not yet well informed on digital modulation schemes, so I'm unable to go into detail about their bandwidth requirements. I do know that they're a lot tighter than the example I used above.

Hope that helps to answer things for you.

73 de KC7GR

Re:Exactly how crowded?

[jstott]

How finely split can the spectrum be before there's danger of overlap? Is it possible for, say, one phone to send signals at 800.0001 MHz while another does so at 800.0002 MHz? Where is the precision cutoff for neighboring frequencies before things start to interfere?

It depends how much noise is present in the communications channel and how much bandwidth (i.e., what data rate) your application requires. Look at Claude Shannon's classic paper A Mathematical Theory of Communication for a complete treatment--he developed information theory to answer precisely this question.

-JS

Re:Exactly how crowded?

[KWSN-MajorKong]

There are several factors at work in the cell-tower/public safety interference issue. There is the bandwidth issue (a narrow-band FM voice channel takes about 10KHz). One cannot pack channels closer than that. This limits the number of channels available in any given area, and they have to be shared by the various competeing services. Much of the time, there is some dead space between the channels so that the reciever does not have to be as selective (hence less expensive). A common spaceing in this area of the spectrum is 15 KHz. This means that the channels are something like this:

800.000MHz 800.015MHz 800.030MHz... etc.

Due to the popularity of repeaters, these frequencies are paired. On each channel, there is a transmit freq. and a recieve freq. The 'repeater' listens on the 'remote unit transmit frequency' and when it detects a signal, it rebroadcasts the signal on the 'remote unit recieve frequency.' The repeater is usually WAY higher powered than the remote unit, and has a MUCH better antenna, thus allowing everyone in a wide area to hear the signal of the remote unit (usually a handheld or vehicle mounted radio). The dispatcher usually has a direct tie-in with the repeater and uses it to listen and transmit.

But, other than just terrain effects in the 'dead zones' (such as being down in valleys, or in the RF shadow of various structures), the most likely cause of the interference to public safety radio systems in the dead zones is due to the nearness and power level of the cell phone transmitters.

Public Safety radio systems are (nowadays) usually 'trunked'. This means that the system is allocated a number of channels with one used as a 'control' channel. When a remote unit (police handheld, for instance) wants to transmit, it sends a request to the central system on the control channel, which handshakes with the remote unit to negotiate a frequency pair to use. This allows several related 'services' such as police, fire, and various other agencies to 'share' the same set of channels, with the services radio nets and subnets being programmed into the trunked system controller. One particular channel might be used for police one transmission, and for the fire dept. the next. Fewer channels needed -> More services can share the same spectrum.

The problem is that high powered (relatively speaking) transmitters in close proximitiy (in both frequency and position) to the handheld/vehicle radio will tend to block the control signals from the trunked network in several ways, preventing the handshake and rendering the police/fire radio USELESS at that location. One cause of blocking is that FM recievers reject all but the strongest signal, and if the cell phone transmitter is putting out adjecent channel interference, and the police radio is close, it will only 'hear' the cell tower. Another cause is that FM recievers tend to desensitize around transmitters in the same general frequency band. Even if the other signal is too far removed in frequency to produce adjecent channel interference, the FM reciever effectively 'puts its fingers in its ears' and again, the unit will not handshake with the trunking system. These same factors can also interrupt conversations in progress. Both causes are due to the way that FM radio works, and are very difficult to design around.

About the ONLY solution is to put these two services in TOTALLY different bands, but, due to enormous investment in equipment for both cell phones and police/fire radio, and the inherent scaraceity of spectrum space, this either is not gonna happen, or its gonna take quite a number of years....

The only OTHER solution is to put both cell phones and public safety/local govt. onto the same system (migrate the public safety stuff onto the cell phone network, with a suitable preempt of the cell traffic in case of public safety need). This MIGHT work, but again would require capital expenditure from both parties.

All in all, this is a sticky situtation that is only going to get worse as our society becomes ever more dependant on wireless communication.

Short Range?

[Gaewyn L Knight]

*Laughs at /. editor as he types this over a 6.1mile 802.11b link*

Really... the main reason for the interference between cell and emergency is that they are all moving into the same spectrum space. 800mhz is a busy place these days.

True wireless (802.11a,b,g) are all in FAR different spectrums so other than cordless phones and some radar equipment interference is a moot point.

Correction

[Faust7]

802.11a and b currently operate primarily at 2.4 Ghz.

Actually, only 802.11b operates at 2.4 GHz. 802.11a operates at 5 GHz.

The emerging 802.11g standard is intended to deliver the same data rate as 802.11a but on the 2.4-GHz band.

Re:Correction

[Penguin2212]

Regardless, his point is valid in that WiFi operates well out of the range of Emergency frequencies.

Re:Correction

[rednaxela]

Yup. Sorry. Pre-coffee comment. Bottom line is that they operate in unrelated bands.

Re:Correction

[Cyno]

I think the FCC acknowledged that 802.11g will not be as fast as advertised.

Re:Correction

[Anonymous Coward]

The FCC has nothing to do with the rate of transmission on an unregulated frequency.

Open Spectrum

[schnarff]

Seems like we need to point the FCC towards Open Spectrum (mentioned in a previous Slashdot article). After all, why go to all the trouble of reshuffling the tiny spectrum we've got now when the proper devices and management can give us more bandwidth than we know what to do with yet?

Not the only problem...

[Anonymous Coward]

I get WiFi interference from the emergency frequency my microwave uses...

Re:Not the only problem...

[aldousd666]

I don't think this is offtopic, actually I think it's kind of funny.

Re:Not the only problem...

[Anonymous Coward]

Hmmm... parent poster has a point in that household appliances do cause a lot of interference across radio frequencies. If a microwave can cause disruption of a wi-fi connection - which doesn't immediately spring to mind - then imagine the kind of interference a housing estate might be pumping out through 'leaky' frequencies that will cause problems for emergency services communications. Vacuums, microwaves, TV's, power drills - everything that can generate a frequency has the capacity to cause problems, and it isn't just the frequency, but also multiples of it, and other atmospheric effects that can cause the signal to split, bounce and travel much larger distances than expected - I have managed to pick up my local FM station over 300 miles away because of weather conditions.

So, microwave and wi-fi - offtopic, but it's the first effect that a member of the public is gonna notice.

NYC Wi-Fi

[TrippTDF]

I do some home and small business tech support in New York, and I've seen some interesting things happen with Wi-Fi.

The most interesting being the fact that I found 20 open and join-able networks on the corner of 20th and Broadway last week. I'd say some people need a lesson in security...

Also, I've seen people name their network things like "get the fuck off my shit" when neighbors try to join their network.

I'll also be interested to see if Wi-Fi networks effect piracy at all- what if the RIAA manages to crack down on piracy to a point that it cripples internet file trading (it wont happen, but come with me on this for a sec) and the only safe way to trade files is by sharing them on a local AirPort network. Then people can run around with their Wi-Fi PDAs with 1GB or greater flash cards, and download stuff from open networks they find on the street.

Re:NYC Wi-Fi

[Anonymous Coward]

"...to see if Wi-Fi networks effect piracy at all"

affect, verb trans.: to cause an effect in something
effect, noun: something that inevitably follows an antecedent (as a cause or agent)

Although you can use effect as a verb, it means to bring about something or to cause to come into being, neither of which fit the context of the sentence. And don't try to make excuses that you're British either, and that that's how it's done on that side of the pond...

Re:NYC Wi-Fi

[raju1kabir]

I'll also be interested to see if Wi-Fi networks effect piracy at all

Apropos misuse of "effect"... Until recently I allowed all comers to use my Wifi connection (albeit at reduced bandwidth, about 128K shared between all strangers) for anything except connections to port 25. I'm adjacent to a nice outdoor public space with comfy seating so I figured it would be the neighborly thing to do.

I noticed that a few people were on there non-stop around the clock (including when said public area was observably deserted), using every bit of bandwidth they could squeeze through. Turned out to be Kazaa traffic. Scared me a little bit with all the RIAA subpoenas flying around, so now I've restricted them to ports 22, 80, and 443. Usage has dropped to a couple people per day.

Re:NYC Wi-Fi

[iso]

I was recently on a business trip to NYC and stayed at the Hilton on 6th and 52nd (IIRC--around there anyhow). I was on the 30th floor and was able to find five wireless networks, three of them open. I'd just like to thank the owner of "linksys-g" for letting me avoid paying the Hilton $12.95/night for Internet access.

I've started taking my PowerBook+airport card and a WiFi scanner with me on business trips. It's rare that I can't find an open wireless network when I need one. On the same trip near Reagan airport I was able to find four open networks from my hotel room on the 9th floor. What was frightening is that the one I used ('wireless' from a netgear router on Verison DSL) was considerably faster than the Wayport wired access available in the room!

It never ceases to amaze me how many people plug in a wireless router and don't even think to secure it. As long as the major manufacturers don't ship with any security on by default, it will only get easier. I'm going to take advantage of it while I can, because this just can't last forever.

here's a thought ...

[dlasley]

> Meantime, Fisher said many colleagues on the Anne Arundel County police force have found their own solution: They carry cell phones in case their radios go dead.

so that being an option, why not have the cell companies take over administration of all the systems in these ranges, subcontracted thru local government offices? have the companies then manage (and be liable for) the need for non-overlapping frequencies separating critical traffic from the general public. especially the ones that shouldn't be driving and yapping on a cell phone as they hurtle through traffic (you know who you are!!!)

and yes, i know this puts a consumer-driven entity in charge of systems that are depended upon by lcoal/stage/regional entities, but hasn't private enterprise often proven its ability to manage complex systems a tad better than the government? could be a re-birth of telecom spending and contracts (which could mean jobs, more bottom-line investing, more attention on the nation's eroding infrastructure ...)

Re:here's a thought ...

[sbma44]

but hasn't private enterprise often proven its ability to manage complex systems a tad better than the government?

Like, say, with energy deregulation in California? Or privatized public schools? Or the free market for health insurance, which has left millions uninsured?

Bureacracy will inevitably develop when such systems reach a certain size, and bureacracies will inevitably suffer inefficiencies. But I don't think adding profit motive to the equation necessarily makes the systems run any better -- it just shifts the mandate of the organization to revenue generation by system maintenance, rather than purely system maintenance -- and consequently opens the door to Worldcom/Enron-type problems.

I would rather not have market forces detemining whether I get 911 service on my cell, thanks

Re:here's a thought ...

[Liselle]

No, private enterprise has fallen flat on its face in this regard. Aparrently your power didn't go out last week, 50 million other people weren't so lucky!

Re:here's a thought ...

[cybercuzco]

Richard Dean Anderson: Now you can dial 911 for the low low rate of 10c a minute, with no connect fees or startup costs ever. Just dial 10-10-1976-2348-7206-459821 and then 911, because you families safety is depending on you remembering that number...

Re:here's a thought ...

[balister]

Haven't you heard what happens to the cellular system during a larger crisis? It collapses when everyone tries to call at once. Didn't you see the articles about people seeking out the remaining pay phones to get in touch with friends and family.

Cell phone networks are designed to work most of the time. Cell phones would be outrageously expensive if the network was designed to handle peak capacity during a crisis.

Public safety types need an effective communication system seperate from public systems. During a crisis, they use discipline to prevent swamping their communication network.

Philip

Re:here's a thought ...

[LittleStone]

Like, say, with energy deregulation in California? Or privatized public schools? Or the free market for health insurance, which has left millions uninsured?

All three examples shouldn't be considered as problematic free market. California never really deregulate it's energy market (the blackout is a result of misalign of incentive when price of hydro is freely adjusted). Privatised public schools endure as much as the shortcoming of public schools, while it may save some taxpayer's money for other issues (not to mention some interest group intentionally misalign resources after privatisation to create an illusion of failed privatisation). For those who have no health insurance under privatised market, how many of them would buy health insurance even if the government give them enough money to buy it?

No, free market is not the answer to everything. Just as asking the government to run everything isn't the answer to the same problem. Worldcom/Enron-type problem isn't problem of free market. It is the problem of WILD market. The actual solution is not to pseudo-analysis which label all private enterprise as greedy, but to study how each incentive arrangement (or institution) would give what outcome.

You may not be aware of it, how you get 911 service on your cell is determined by market forces already. The mobile system is not capable to handle a sudden increase in 911 call volumn at the same time. This is not a problem of free market. Either the regulator knows that and explicitly rate cell service as non-essential, or the regulator did a half ass job and did not specify clearly the cost of using the public spectrum.

Is the FCC doing its job?

[Brahmastra]

I thought one of the reasons the FCC exists is to allocate frequency blocks precisely to avoid problems like this? Aren't wireless devices supposed to be certified by the FCC?

Re:Is the FCC doing its job?

[nchip]

IANARFE (Radio Freqency Enginerd)

I think this is caused because the FCC specs say that a device X must use frequency y, and less than backgound noise on other frequencies.

Now, every RF transmitter is a analog device and as such it's impossible to transmit perfectly in the mandated frequency, so some leak to the neighbouring frequencies is inevitable. Still OK, a single device X will not produce noticable interference, as it produces less than background noise.

Now, instead of one device, you get thousands of RF transmitters, and the interference starts adding up, finally surpassing the background noise.

I believe this is also the reason why people are worried of UWB tech as well.

Instead of measuring interference produced by a single device, FCC will probably have to move measuring the interference created by an operational, full network.

Ofcourse this begs to ask why FCC allocated 800Mhz for mobile phones while knowing public safety departments use it AND the whole fucking entire other world uses 900Mhz. Thus making american phones incompatible with the entire other world for a long time...

WiFi Is Not Near Public Safety Bands

[Goody]

Wifi is not a problem, evidently. Understandable, given its short range.

Ummm, no. Here's a frequency chart of the radio spectrum. People seem to think everyone uses all frequencies and it's one big radio spectrum blob. Radio spectrum resembles IP space, except in this case you can't create more of it. Print out the chart on a big plotter, hang it on the wall and impress your techie friends, and consult it when a wireless article is posted on Slashdot. :-)

K3NG

Backup measures

[GillBates0]

Meantime, Fisher said many colleagues on the Anne Arundel County police force have found their own solution: They carry cell phones in case their radios go dead.

I would've thought emergency services always had some backup mechanisms in place to get back to the station in case of emergency.

Heck, I even thought the dash-cam that police cars are all supposed to have also relayed video back to a surveillance team back at HQ. Apparently it is not so. With risky, unpredictable work like emergency services at hand, I for one, am surprised they thought of fallback procedures only now.

I know transmitting live video from the surveillance cams is difficult, and multiple backup contact mechanisms would be expensive (cellphones, satellite phones et al), but then, I thought these guys would be the ones who deserve it most of all.

Re:Backup measures

[stratjakt]

Being as I live in Anne Arundel county, and work in the public safety field as a programmer, I can tell you that these podunk maryland cops are some of the most incompetent and out of date you'll find anywhere.

Think Mayberry. I can tell you from first hand experience that these guys are equipped to write speeding tickets and bust 14 year olds for posession of dope, and are some of the lowest paid in the nation. They start at like 25 grand a year.

If there was ever a major emergency around here (like 9/11 or hurricanes or whatever) there are so many military bases nearby, the feds can easily swoop in and take care of it. For example, the feds caught the DC snipers - local cops had an extremely minor role in the whole thing as a more of a press liason than anything else.

This is a mostly regional article and doesnt necessarily pertain to other regions I've been to, like LA or New York.

ip6 and TCP/IP for all wireless stuff?

[Chris Croome]

Would a medium / long term soultion of ip6 and TCP/IP all wireless stuff make sense?

Re:ip6 and TCP/IP for all wireless stuff?

[sbma44]

I don't think it's a lack of address space that's the problem, but it would be interesting if anyone knows whether these 800MHz applications are packet-switched or analog. If it's the latter, a change to digital technology would likely dramatically unburden the airwaves.

Artificial scarcity?

[heironymouscoward]

Is it possible that the FCC benefits from the artificial scarcity that its own policies impose on frequencies? There exist technologies to make much more efficient use of frequencies and to eliminate all interference, but mysteriously we still get all our services squashed into a small and crowded space where the right to broadcast commands high fees and prestige for the authorizing body (=FCC).

Re:Artificial scarcity?

[Phreakiture]

Nah, this is a technology gap, I think. What's in R&D and what's in production differ. Further, it's encumbered by backwards compatibility. If you go to the new technology, all of your old radios become basically paperweights.

There are technologies for making a smaller footprint. Comparing different signal formats for two-way, you have FM, which, when formatted for 2-way, uses 10-20kHz of bandwidth. Compare that to AM, which uses 6-10kHz, compared to single sideband (SSB), which uses 3-5kHz (usually leaning toward 3).

Also, some digital technologies are intersting. APCO-25 has two protocols, called phase I and phase II. Phase I uses 12.5kHz of bandwidth, and Phase II uses 6.25.

CDMA is the first rolled-out technology that tolerates overlap well. This protocol actually occupies several megahertz (that's megahertz, not kilohertz like I was talking about before) but allows several transmissions to go on at once on the same frequency, and can sort them out at the receiver. Verizon and Sprint PCS (and probably some others) use this technology for their digital phones.

No kidding

[siskbc]

That'd have to be some serious interference for those to be affected by wifi, huh? Do people really not know all that much about allocated use of spectrum?

Range

[jargoone]

Wifi is not a problem, evidently. Understandable, given its short range.

The reason it's not a problem has nothing to do with range at all. It's more because of the low usage when compared to other RF. If there were as many wifi users as mobile phone users, it would be just as big of a problem.

I don't recall this being a problem in the UK

[Zog The Undeniable]

Apart from a (very few) 3G phone users, we're all on GSM 900 or 1800 MHz. I understand there are two or three different mobile technologies in the US; does this use more spectrum?

Re:I don't recall this being a problem in the UK

[Anonymous Coward]

The US 800 MHz cellular bands are closer to the land/mobile radio allocations. GSM systems are in the middle 900 MHz bands, farther away, resulting in lower interference.

20th and Broad?

[StarKruzr]

I've found a crapload of stuff around Union Square but haven't gotten up to 20th yet. I did notice that it was virtually a barren wasteland up near Times Square, but have found a ton of ESSIDs in and around the Village.

It looks like companies don't use 802.11 hardly at all, rather, the more residential a neighborhood, the more prolific the access points become.

Unfortunately there's virtually nothing near the Staten Island Ferry (I live in SI) other than that damn Starbucks that closes too early and the Verizon Wireless AP that's WEPped and closed. 802.11 in the Whitehall Ferry Terminal!

Re:20th and Broad?

[TrippTDF]

I work on 11th and University, and there is nothing down here that I can connect to from my office.

I don't think larger companies have a need for 802.11 yet. Not until Wi-Fi PDAs get popular. It's still a better technology for commercial/residential use.

Re:20th and Broad?

[benzapp]

I live in Brooklyn and intentionally leave my wireless AP open, and I have friends in the east village who do the same.

I like to be able to access the internet anywhere in the city with my PDA, and I am happy to allow others like myself free access in my local hood.

Of course

[Faust7]

Duh, I'm an idiot. How else do radio astronomers operate? *slaps forehead repeatedly* Thanks.

Re:Of course

[aldousd666]

Well, I wouldn't be so hard on yourself there killer: the radio astronomers are a good illustration that such things are technically feasable, while not necessarily cost-effective.

Frequency Encroachment

[hobbespatch]

I had the opportunity to attend a few hearings on Capitol Hill this year regarding the impact of spectrum/frequency encroachment on military communications and training.

The basic argument, like the article makes for first responders, is that the military band for communications is being encroached upon by civilian use. Having fewer frequencies directly impacts the military's ability to conduct training operations and exercises.

The Navy (Department of Defense) has a page which educates visitors and range spectrum users on how to defend against civilian encroachment of DoD frequency spectrum.

The Electromagnetic Spectrum Training Chart shows military uses of certain frequencies and the competing civilian use.

Problem is.....

[winstarman]

There is only so much frequency space, and once it's been allocated, no one wants to give it up.

Do you really think that amateur radio operators will come forward and say "Hey, we've got bandspace to spare, take some of ours!"?

Maybe there should be more stringent standards on how much interference an appliance creates?

But personally I say we just eliminate the CB frequencies. Only those with too much spare time and no mobile phone use it anyways. Let it go and get FRS or GMRS for pete's sake!

Re:Problem is.....

[westlake]

There is still a need for cheap, simple, mid-range, solutions that are not dependent on repeaters; do you think that everyone lives, works or plays in line-of-sight of a tower?

Re:Problem is.....

[Anonymous Coward]

But personally I say we just eliminate the CB frequencies. Only those with too much spare time and no mobile phone use it anyways. Let it go and get FRS or GMRS for pete's sake!

Yeah, because people shouldn't be able to communicate over a mile without paying for an fcc license or cell phone service.

Re:Problem is.....

[kg9ov]

There is only so much frequency space, and once it's been allocated, no one wants to give it up.

Do you really think that broadcast TV companies will come forward and say "Hey, we've got bandspace to spare, take some of ours!"?

Maybe there should be more stringent standards on how much interference an appliance creates?

But personally I say we just eliminate the broadcast TV frequencies. Only those with too much spare time and no cable service use it anyways. Let it go and get cable or satellite for pete's sake!

Re:Problem is.....

[winstarman]

Exactly...

No one wants to give in.

KE4WLE

Darned Towers

[FooAtWFU]

In my hometown, there was a plan for a very tall tower (two or five times higher than a standard cell phone tower) to be built to facilitate emergency communication issues. Something about it being physically higher than all the low-level wireless interference... I think. Maybe not, but in any case...

The tower was set to be built just down the street from my house in the neighborhood park, which is un-fun, to say the least. Fortunately, starting a silly grassroots community NIMBY campaign proved unnecessary, because it turned out that the site was in the middle of an approach pathway for the local airport. Mwuahahaha!

You'll need both

[siskbc]

Has absoulutely nothing to do with anything on the broadcasting end.

I really don't know about that. Just as receivers have tolerances, transmitters do too. So if I have a crappy transmitter and I tell it to broadcast at a certain frequency, there will be a certain tolerance there, unless I'm using a laser. They're certainly not. Even with a good transmitter, there's a certain +/- to the frequency distribution, although presumably less.

So ultimately, his question was quite a fair one - for someone to tell the difference between two signals 800.0000001 and 800.0000002, the two transmitters will have to be good enough to send out precise, narrow signals, and his receiver will have to be good enough to tell the difference.

Re:You'll need both

[aldousd666]

right, point taken. I was too quick to say 'absolutely nothing.' Although, most of the problems described are probably still more to do with crappy receiving rather than transmitting.

Re:You'll need both

[siskbc]

right, point taken. I was too quick to say 'absolutely nothing.' Although, most of the problems described are probably still more to do with crappy receiving rather than transmitting.

Well, assuming they spend more on the transmitter than you do on your cell phone, huh? ;) Although the way Sprint's been lately, I'm not so sure.

Re:You'll need both

[elvum]

Even with a laser there's a spread in the frequencies of light it generates - it's this fact which limits the distance at which light from the laser is still coherent, a property of the device named (highly originally) the "coherence length".

Re:You'll need both

[siskbc]

Even with a laser there's a spread in the frequencies of light it generates - it's this fact which limits the distance at which light from the laser is still coherent, a property of the device named (highly originally) the "coherence length".

I'm so sad I actually look forward to getting called on things like that these days. Yes, of course you're correct, particularly if said laser isn't operating through a vacuum. Compared to an analog transmitter, however...

Re:You'll need both

[dougmc]

So if I have a crappy transmitter and I tell it to broadcast at a certain frequency, there will be a certain tolerance there

Yes, you're right. But it's not a function of a `crappy transmitter' ... it's called `bandwidth' -- and while you may have heard the term before, *this* is where it originated.

A radio signal, unless it's a pure sine wave, uses a certain amount of bandwidth. Double the amount of data pushed through, and you double the bandwidth used, all other things being equal.

For example, my R/C plane uses 72.450 mhz, and the bandwidth used is 20 khz, so 72.441 mhz -> 72.459 mhz are what's in use. Well, that's what's allocated anyways -- the radio should use a little less. The transmitter should not transmit signifigant amounts of power outside of that range, and the receiver should ignore any signals outside of that range.

(For comparison, morse code can use less than one khz of bandwidth. FM voice transmissions (such as used by police radios) take up 3-6 khz. SSB and FM use less, but the sound quality is less too. More bandwidth = more quality, though certain modulation types work better with little bandwidth. I don't know how much bandwith a commercial FM radio station (88-108 mhz) uses, but it's probably more like 100 khz (after all, it's music.)

If you start reducing the bandwidth used, you lose data. If it's an AM signal, you'll start losing the higher frequencies. I'd have to think about what it would do to other modulation types ...

The real problem here is that the FCC has done some really wonky allocations of bandwidth. For example, the 72.450 mhz frequency I mentioned earlier -- the 72 mhz band is for R/C planes, but it's also shared with pagers. So 72.450 mhz is channel 33 for planes, but 72.460 mhz is probably used by a pager company. If they put their tower near our R/C field, and I fly my plane over near that end, and a page goes out ... their signal can overpower my wimpy little 1 watt transmitter, and guess what happens to the plane? (Actually, 1 watt isn't that wimpy -- R/C plane transmitters have a range of about 1.5 miles. But the pager tower may use something like 750 watts ...)

In any event, that exact same problem has happened in the 800 mhz band -- 801.4 mhz may be allocated to a police band, but 801.5 mhz may be used by a cell phone tower ...

Re:You'll need both

[siskbc]

Yes, you're right. But it's not a function of a `crappy transmitter' ... it's called `bandwidth' -- and while you may have heard the term before, *this* is where it originated.

I have - I'm a chemist, so I do some spectroscopy. ;) I was more referring to the license situation the FCC has. It allots you, say, a 0.5 MHz bandwidth. Your transmitter is supposed to transmit in this band. However, say it sucks - the spectrum might not be as tight as it's supposed to be, and it might bleed over more.

For what it's worth, at a given frequency, it's the better transmitter or source/monochromator (that's what we'd call it) that provides the narrower bandwidth. For example, considering visible light, a good system would give you a bandwidth of a few nanometers - a bad one, ten times that.

In any event, that exact same problem has happened in the 800 mhz band -- 801.4 mhz may be allocated to a police band, but 801.5 mhz may be used by a cell phone tower ...

Yep. And if that cell tower is too powerful, or bleeds a bit, you know what you get. I think that part of the problem is that the shoulders on any transmitter's bandwidth are nearly disregarded, but they can create significant problems if you're close enough.

Re:You'll need both

[dougmc]

I have - I'm a chemist, so I do some spectroscopy. ;)

Neat! My degrees are in physics and astrononmy, but alas I don't get to use them much `in real life' ... that would kick much ass if I make my living at it!

I was more referring to the license situation the FCC has. It allots you, say, a 0.5 MHz bandwidth. Your transmitter is supposed to transmit in this band. However, say it sucks - the spectrum might not be as tight as it's supposed to be, and it might bleed over more.

Right. But people were talking about frequencies like 800.0000001 mhz ... as if you could somehow make a perfect radio that would allow lots of data to be pushed into 0.0000001 mhz, which you definately cannot do. Shannon and Nyquist and all.

But, as you've said, radios are not perfect. They should not transmit outside of their band, but of course small amounts of emissions escape anyways, and you cannot prevent this 100%. Alas, when your transmitter is 1000+ watts, the amount can become signifigant. And then there's harmonics -- 2.4 ghz should not affect 800 mhz, but the harmonics of 800 mhz could certainly interfere with 2.4 ghz (WiFi) signals.

Re:You'll need both

[siskbc]

Right. But people were talking about frequencies like 800.0000001 mhz ... as if you could somehow make a perfect radio that would allow lots of data to be pushed into 0.0000001 mhz, which you definately cannot do. Shannon and Nyquist and all.

Ah, I assumed by that he meant "800"+(lots of zeros to make the last digit nearly insignificant)+"1". But if you actually look at the number of zeros (which I'll admit I didn't do!), you are correct, sir.

And then there's harmonics -- 2.4 ghz should not affect 800 mhz, but the harmonics of 800 mhz could certainly interfere with 2.4 ghz (WiFi) signals.

That's most certainly true, and I hadn't even considered it. Do they try anything to avoid that? If I were the FCC, I would try to keep major bands from being integer multiples of each other - and that 800/2400 looks like potential disaster there.

Re:You'll need both

[dougmc]

Do they try anything to avoid that? If I were the FCC, I would try to keep major bands from being integer multiples of each other - and that 800/2400 looks like potential disaster there.

Impossible, or at least impractical. It would require losing the ability to use most of the spectrum -- because after all, whatever frequency you're talking about, there's a frequency that's 1/2, 1/3rd, 1/4th etc. of it, that could interfere with it.

Good transmitters have filters that almost completely kill the harmonics. Not quite, but almost.

There are other factors too -- for example, dual conversion R/C receivers have a 455 khz intermediate stage that can set them up to receive interference from 455 mhz away. I don't fully understand all of it, but let's just say there's lots of ways to cause interference. :)

Re:You'll need both

[siskbc]

Impossible, or at least impractical. It would require losing the ability to use most of the spectrum -- because after all, whatever frequency you're talking about, there's a frequency that's 1/2, 1/3rd, 1/4th etc. of it, that could interfere with it.

I wouldn't leave it unused, just try to shif things so that, say, two bands with extremely high power and use aren't directly "integered."

Also, I believe the relative power of the nth harmonic goes down directly with n, correct? As far as that goes, I suppose they've done OK - there are a few orders of magnitude between AM,FM,and wireless (800MHz). Maybe that's why nothing major happens at only a few hundred MHz?

According to This site you can actually listen to the harmonics of an FM radio station in the hundreds of MHz region.

Re:You'll need both

[dougmc]

I wouldn't leave it unused, just try to shift things so that, say, two bands with extremely high power and use aren't directly "integered."

Easier (and cheaper) said than done. :)

Also, I believe the relative power of the nth harmonic goes down directly with n, correct?

It probably goes down exponentially -- 2nd harmonic is 1%, 3rd is 1% of 1%, etc. Not sure.

According to This site you can actually listen to the harmonics of an FM radio station in the hundreds of MHz region.

Absolutely. I don't have a scanner that can do 72.450 mhz, but I can definately hear my R/C transmitter by tuning my 2m ham radio to 144.900 mhz. The range is greatly reduced, but I can still pick it up up to about 100 ft away.

Monitoring dead spots

[mfarver]

It sounds like most of the problem is local agencies not monitoring the QOS of their radio systems. Most organization get a contractor to install the equipment, towers and contract for basic maintainence and then expect the system to run for years with little oversight..

Cell providers (and most commercial radio operators) know that the precise interrelationships are always changing. New buildings, new transmitters, malicious/accidental interference. Most wireless carriers send drivers with GPS/inertial locators and signal strength meters wandering about their coverage areas to locate areas of poor coverage. "..hear me now, Good!"

A friend once showed a map drawn by a wireless system installer showing that with X number of towers the entire coverage area would have maximum signal strength. Most government entities would stop here, believing the installer. But this wireless company did a standard mapping of signal strengths and found inadequate coverage all over, the number of towers had to be almost doubled from the original estimate and dead spots still exist.

Good radio operation is more than just getting a license and standing up a tower. Whining that the FCC should step in just becuase you're too lazy to fix you own stuff is irresponsible.

Wireless Interference

[brain1]

To preface my comments, I am a holder of a First Class Radiotelephone FCC license, and an active Amateur Radio Operator. I have been involved in communications of all types, from TV and FM transmitters to engineering two-way radio installations. Now with that said --

What a load of crap! It's not the problem of the wireless providers, it's a problem of coverage due to poor system engineering.

Most, if not all, 800 mhz emergency service systems operate on what is call a "Trunking System". What this is, essentially, a system of linked towers that communicate with the vehicle or officer on the street, then relay from tower to tower ultimately connecting to either the dispatcher or another officer. These systems are designed to be interoperable with each service, such as police, fire, ambulance, etc, so a single dispatch facility can communicate with everyone, and all services can communicate with each other.

To work effectively, you must have sufficient towers properly placed to assure that there is no dead zones. Given the expense involved in site purchase, permits, tower erection, equipment installation, and backup generators, the bare minimum is pretty much the rule. Plus, you cannot physically survey the entire area of coverage, you use topo charts to try to make sure your engineering is sound.

To blame the wireless providers is silly and stupid. Modern 800 mhz equipment is very selective, most newer systems operate via spread spectrum digital, and the chances of interference are minimum.

Additionally, emergency services have priority so if there is a provable case of direct intererence, the wireless provider must take steps to either stop the interfering signal, or cease operations entirely.

Shannons Law Re:Exactly how crowded?

[WolfWithoutAClause]

The answer to how close you can put the channels is known as Shannons Theorem. This basically says that the radio channels can't be put closer together than a certain amount, since each channel needs a certain bandwidth.

The amount of bandwidth basically depends on you much noise there is around- you can pack much more data into a narrow channel if the channel has hardly any noise, whereas if the channel is very noise you'll need a wide channel to send the same data.

Also in practice you need a gap between neighbouring channels- the receivers need to filter out the other channels and they don't do this perfectly (although the better the receivers are, the closer you can pack them in.)

However Shannons law only deals with broadcast communications where everyone can 'hear' everyone else equally, if you use directional antennas then it doesn't necessarily apply- two communications could use the very same channels.

Good timing!

[Anml4ixoye]

How funny that this should come up. I was just in a meeting last week talking about this very thing.

I am the Assistant Chief of a volunteer department north of Tampa, Florida. We operate off of the 800mhz radio system, and for all of the money we spent on it, you would think it would work better. Instead, we have spots where we can hear but not transmit, or just plain not get reception at all.

While we were talking about this one of the Hillsborough County chiefs brought up the fact that when he is at headquarters (where our central towers and dispatch are) he can't use his little car-alarm-door-unlock-gadget-thingy unless he is almost touching his vehicle. When he is at home, he can activate it from 200-300 yards away.

We have always blamed it on the hardware and crappy company, but now I think I might have to look a little more into what other kinds of interference would cause transmission and reception problems in the areas where we are having difficuly.

Re:Good timing!

[Phreakiture]

I'm going to ask a question that may sound snide, but is actually very serious: Why did you go to 800MHz? Were you using VHF-Lo, VHF-Hi or UHF before? What was the coverage like?

I ask this because in my hometown of Selkirk, NY, they use a system that is a hybrid VHF-Lo and UHF system. VHF-Lo (46.06MHz) covers the town, and the UHF is only used as a remote base, with the base radios at the three fire companies talking on 455.6375 and listening on 460.6375. It works very well. My father, who is a member of this fire department, has been fighting a movement to "upgrade" to 800MHz.

At my current locationin Schenectady, NY, they use 460.6000. It works well and has no dead spots that I am aware of. I will fight any plan the city puts forth to "upgrade" to 800MHz.

Geek centric?

[Thinkit3]

Yeah, mobile phones are for the stupid mainstream people who just like to talk. Wifi is for the future, intelligent geeks who need high-bandwidth data streams. Geek spin going on here?

Depends how much bandwidth you want

[grahamsz]

There is a reasonably clear correlation between radio bandwidth and usable data bandwidth. So in your example, the phones will have only 1kHz of bandwidth each. In order to squeeze a phonecall in that space you'd need to use some very clever compression and encoding scheme which would likely be very intolerant of noise.

IIRC cellphone networks are divided into 64kHz channels, each of which are shared by 8 phones. So each GSM phone has around 8kHz of bandwidth to play with.

FM Radio on the other hand i think has 150kHz per station, so that limits how closely stations can be spaced.

TV is HUGE, several MHz wide.

As you move to higher frequencies of course you can have more bandwidth per channel since there's just more of it. There the same between 2.4 and 2.5GHz as there is between 0 and 100Mhz.

Unfortunately higher frequency signals dont tend to travel as far in crowded environments - a single LW transmitter can cover most of a continent whereas at 5GHz you need a lot more wifi cells to do it.

And microwave ovens?

[hankwang]

>whereas 802.11a and b currently operate primarily at 2.4 Ghz

Interestingly, microwave ovens operate at 2.45 GHz (not listed on the spectrum allocation chart mentioned elsewhere). I wonder how effective the shielding of a typical microwave oven is. If they leak 1% (for example when loaded with a very small amount of food), then that's still 8 W, compared to 1 W for a GSM mobile at 1.8 GHz. I don't know either within what tolerance MW ovens are; for the food inside it doesn't matter very much whether it is 2.2 or 2.6 GHz.

Re:And microwave ovens?

[CoyoteGuy]

Interestingly, microwave ovens operate at 2.45 GHz (not listed on the spectrum allocation chart mentioned elsewhere). I wonder how effective the shielding of a typical microwave oven is.

That's it!! The moment my WiFi interrupts with my microwave, it's going out the window!!! How dare WiFi interrupt me cooking my Pizza Pockets!!

Re:And microwave ovens?

[once1er]

I dont think the shielding is all that effective. I used to have a TV service like Direct TV, but it was suppose to use microwaves or some junk like that. In a weird strange sequence of events the antenna was placed on my roof, a story above my microwave oven. Every time we microwaved, our TV would go out. Took me about 2 months before I realized what was happening. So I'm thinking those little buggers leak waves like crazy.

Re:And microwave ovens?

[anjrober]

I posted this story a bit ago, but it's relevant yet again. A while ago I added the Home Media option to my Tivo. It connects to my Linux box via 802.11b. So I'm streaming an MP3 from my linux box one day to my tivo->stereo and the Mrs. puts something in the microwave. The music stops. The microwave counts down, 3...2...1 and the music starts. Very cool. Clearly the microwave has been interfering for years with my web connection, but the latency and design of the web allow for intermittent connections and thus I don't normally realize the microwave is killing my connection. But you sure do when an MP3 cuts out. Very cool.

Re:And microwave ovens?

[hankwang]

> I dont think the shielding is all that effective.

According to this microwave oven safety page, the legal limit of leaking radiation for a new oven is 1 mW/cm2 at 5 cm from the oven (for a used oven, it's 5 mW/cm2). For a complete oven measuring 25x40x30 cm, that would be a legal limit of 70 W if it radiates uniformly in all directions. (The page mentions that it's rather rare for an oven to exceed the limits). For a used oven, I'd guess that the door seal is the most critical place, which would correspond to about 1 W for 1 mW/cm2 at 5 cm from the seal.

Intermodulation

[wikki]

I'm a ham radio operator and I can speak first hand of this type of interference. Hams and Police/Fire/Emergency all use similar bands. 140Mhz 440Mhz The problem is when you get near and area with a high concentraton of cell phone towers you hear tons of digital interference. It's gotten to the point, where I live, that I dont' even turn on my radio when I'm in downtown. There are supposed to be filters around which can block out this interference but they are expensive and I've never met anyone that owned one.

Re:Intermodulation

[balister]

My suspicion is that the cell phone companies and public safety people are sharing channels within the
800 MHz band. In this case, additional filtering won't help as the receiver needs to hear signals in the band of interest.

I am guessing that when a public safety radio is near a cell tower, the radio does not have enough dynamic
range to hear the desired weak signal when there is a strong interferer nearby. The AGC limits the gain of the public safety radio based on the undesired cell signal. This would create dead zones near cell towers.

For ham operators bothered by intermod, look at one of these (for 2 M users):

http://www.durhamradio.com/dci-intermod-filters. ht m

I just did a quick google, not an exhaustive study.

This sort of interference/intermod problems is a results of wide range receivers. One of the great things about old Radio Shack 2m equipment is that since they only covered 2m's they are very resistant to out of band interferene. TANSTAAFL or however it is spelled :)

Philip

Re:Intermodulation

[brain1]

These are FM transceivers, not AM or SSB. There is no AGC. Intermodulation interference, as defined, is when two or more signals produce mixing products across non-linear devices in the receiver's front end. These can interfere with communications by actually appearing on a frequency of interest. Interference can be mild, such as inability to squelch the annoying signal, to severe enough to capture the FM receiver preventing the reception of the intended receiver. What you are describing is actually desense. It occurrs when the front-end is saturated with an near-band signal that cannot be rejected by the filters. This is not new. This has been a problem in the past when Police and Fire services used the 150 MHz band and suffered interference from taxi, mobile telephone (152 mhz), and business service. That, plus the interoperability problem between departments, fueled the move to 800 mhz trunking. Besides - most wireless phones operate in the 1900 Mhz band using CDMA, TDMA, and GSM. This has zero interference potential to 800 mhz.

Re:Intermodulation

[bandy]

This has zero interference potential to 800 mhz.

Provided the transmitters, especially the amplification stages, are properly sheilded from the outside world. The frequencies aren't generated at 1900MHz, they start at lower frequencies and mix their products to acheive the final result. Given that a perfect Faraday Cage is difficult to acheive when you need to poke holes in it in order to get things such as power and data in, as well as final signal out, you get weird leaks, causing interference to other services.

Re:Intermodulation

[brain1]

Most PCS systems use direct synthesis rather than a multiplier chain thus eliminating complexity and cost. However, even in the case of a synthesizer and multiplier chain exciter, these submultiples rarely get any distance due to the low energy levels involved. Most likely problem interference scenario here is a cell phone on the officer's belt alongside his patrol radio. That could cause potentially cause problems, but it would require a tri-mode phone that covers 800 mhz as well as 1900. If the phone is transmitting on 800 mhz then desense could happen to the patrol radio's front end. The 1900 mhz PCS system would not readily radiate these emissions due to the high VSWR the antenna would have to these lower frequencies, thus the system has a natural high-pass filter. However, if the phone (or even the tower) has a defective or poorly designed power amplifer, then the potential for generating spurious parasitics exists. The parasitic oscillations could fall within the patrol radio's band and cause interference.

Airport zoning

[nonameisgood]

This is precisely one problem the FCC cannot control. The FCC's stated "plan" for the regulation of towers (particularly HDTV) is to allow local officials and zoning deal with it.

In Texas, even the state cannot do it - our community (pop. 7700 and shrinking, about 100 miles to any other population center) - had to create an airport zoning board, adopt a state-recommended rule restricting construction in the airspace for the airport, meet three times, etc. before we could prevent such things.

While the FCC has abdicated on this, the FAA was absent, and the local, non-pilots were up for anything that brought in revenue.

Nextel is to blame

[Anonymous Coward]

In most cases, it is Nextel's somewhat unusual way of getting spectrum. They bought out a whole bunch of SMR operators and cobbled together spectrum, and made a cellular phone company out of it. It is Nextel in most cases that causes the problems.

What they are proposing is moving public safety to different bands, and giving themselves some more prime spectrum for free. They offered $850 million to move public safety into different bands, which is of course, not anywhere near what the spectrum that Nextel gets is worth.

Nextel has set up a site, which is of course, pro their plan. But don't be fooled, it's wholesale theft.

Nextel is the problem

[Anonymous Coward]

Nextels expansion into additional spectrum
next to the public safety band is causing
"splatter" from their poorly controlled TDMA
signal. Nextels operating practices and sloppy
installation specs - on purpose to allow remote
reshuffling of freqs between base stations
and tower locations increases the problem.

THIS IS THE ONLY CASE THE FCC HAS ALLOWED A
PREFERENCE TO THE CAUSE OF THE INTERFERENCE.
It pays to have the right lobbist, law firm,
campaign contributions, and a Powell running
the FCC who might want to be a Senator.

Re:Nextel is the problem

[Anonymous Coward]

Nextel was there first. It was public safety who expanded into Nextel's spectrum.

Re:Nextel is the problem

[Anonymous Coward]

nextel expanded into additional channels
that it aquired long after 800 Mhz public
safety trunking was allocated and in most
cases constructed

OMG!!

[Anonymous Coward]

None of the companies is doing anything wrong, FCC officials said. As organized, the spectrum, which is a limited resource, simply can't accommodate everyone.

I'm stunned that someone in the FCC finally gets it. Utterly amazed. I already know that the commercial industry understands. Now if only emergency services could figure it out, we could move foreward.

The wireless companies haven't done anything wrong. The FCC fucked this up. They shouldn't have allowed public safety to be allocated so closely to commercial space.

What's good is that someone there finally admitted it, even if only implicity.

As allocated the spectrum cannot accommodate everyone. Yeah, well who allocates it?

Ah... now it makes sense...

The FCC was created to manage this public resource to keep conflicts like this from arising. They fucked up. It's not the wireless carriers. It's not public service. (Whom techinically shares more of the blame anyway... Wireless carriers were there first; public service intruded. It would be like moving next to an airport and then complaining about the noise... though it never should have happened in the first place thanks to, I don't know, the zoning commission?) It's mis-allocated spectrum.

In the end, both users of spectrum space will have to work with the FCC to fix this problem. Whining and bitching about it aren't going to solve anything.

Re:OMG!!

[KWSN-MajorKong]

Wireless carriers were there first; public service intruded. It would be like moving next to an airport and then complaining about the noise...

Actually, it was the other way around.... Public Service was there first... The wireless carriers intruded...

Contemporarizing spectrum regulations

[shakuni]

There is an interesting whitepaper by people from MIT and harvard here http://cyber.law.harvard.edu/ltac98/whitepaper.doc

This brings out the need for changing spectrum regulations from auctions to open spectrum access for three specific reasons:

Technical: The system of allocating a particular frequency band to a single user is based on outdated technology. Early receivers and transmission schemes were such that we needed to be concerned about the possibility of interference. The development and implementation of spread spectrum and digital radio technology allow us to use receivers and transmission schemes such that interference is not a significant concern.

Economic: An auction acts as barrier to entry to small firms and unproven technologies, which are often the sites of innovation.

Constitutional/Legal: By allocating spectrum licenses, the government is essentially picking who can talk and who cannot.

It would be interesting to see if the technical reasons stand the test of these problems of interference. On another note: With so much electromagnetic transmissions, I wonder if the health hazards are being evaluated and addressed systematically and strategically.

Shut down the "Emergency" networks

[aminorex]

The obvious solution is to stop using the "Emergency" frequencies altogether.
Give ever cop and fireman a cellphone,
and the problem is solved, public funds
are conserved, and the traffic becomes
protected and routable with QoS guarantees.

Re:Shut down the "Emergency" networks

[Phreakiture]

Ah yes, let's throw more technology, ergo more complexity and the accompanying greater chance of a breakdown into the picture.

I have a better idea. Let's abolish networks altogether for emergency, and go back to simplex on the VHF band. It works really, really well.

The FCC can't move mountains (or hills)

[Catbus]

Interference is a relatively new part of this problem. For decades, public safety has had "holes" in coverage caused by hilly terrain blocking the signals. Same thing happens with cell phones, but the cell phone companies can afford many more transmitter sites to fill in the gaps. Put another way, public safety coverage will never be as good as cell phone coverage, for reasons that have nothing to do with interference. Reallocate the spectrum all you want -- it won't cure this problem, but it could help with the newer interference issue. What might also help is independent third-party review of public safety radio system designs that are originally drawn up by the same companies that are selling radio systems to the public safety agencies. Some agencies have no idea of what they ordered or what performance to expect or, moreover, the tradeoffs between cost and performance.

Technical Standards

[Detritus]

Are 800 MHz licensees using equipment designed for the best performance using current RF technology? The FCC has an obligation to require licensees to use well designed equipment that takes advantage of modern technology. That allows the most efficient use of the radio spectrum. There is a big difference between the performance of cheap and expensive VHF/UHF equipment. A cheap transceiver is vulnerable to front-end overload, intermodulation and desensing. A well designed transceiver can reject strong signals that are off-channel with minimal performance loss. Public safety agencies should be required to design communications systems with adequate coverage and link margins, and buy high quality radios, not those from the lowest bidder. Cellular sites can install cavity filters to drastically attenuate off-channel noise and spurious emissions.

Although the FCC hasn't always been consistent about it, the general rule has been that the operator of a properly maintained and licensed transmitter is not responsible for the communications problems of other users when those problems are caused by poorly designed receivers.

Re:Technical Standards

[brain1]

I agree. Other than the basic requirement of transmitter type acceptance, there are no rules as to how well the receiver performs. The closest is Part 15 compliance to make sure the receiver itself does not cause interference. There should a minimal set of engineering specs of a planned system that have to be met before a license can be granted. It comes down to money. Cell operators can afford the best equipment - otherwise they lose customers. And they maintain it. But once a public service is locked into whatever bill of goods is pawned off on them, then they are stuck. Nothing like a tight budget and a mayor bent on re-election to put the screws to a department.

Fear BPL!

[602]

The thing to worry about is broadband-over-power-lines (BPL). This would wreak havoc on all kinds of transmitted signals. For more info refer to the Amateur Radio Relay League (www.arrl.org).

Re:You'll need both, and proximity matters

[swordfishBob]

Transmit power and proximity matter. A receiver will have a certain selectivity to filter out other close frequencies, but the closer the frequency the harder to filter, and transmitters of close frequencies can interfere with each other.
FM Radio transmitters that are close in frequency must have a certain physical distance between them.

Also consider off-carrier interference, particularly GSM mobile phones interfering down into the audible (kHz) range. It doesn't just affect hearing aids, but much in between.

Their thinking is limited and short term.

[HeadHits]

To solve the problem, the FCC is considering reshuffling channels in the 800 megahertz band. The idea is to separate the wireless companies from the public safety departments, so they inhabit different ends of the band. None of the companies is doing anything wrong, FCC officials said. As organized, the spectrum, which is a limited resource, simply can't accommodate everyone.

Reorganization would be rife with all manner of contention and expensive reconfiguration. But as painful as it may sound, this is only one dimension of needed improvements for 800MHz band usage. Also sorely needed is more efficient modulation modes and more consistent allocation and constraint in both RF spectrum and geographical extent of the services on the band. They should emphasize improving the spectrum efficiency of the equipment living in the 800 MHz world. Just moving frequency allocations around, even if it makes the allocation chart look cleaner, is merely playing 8-puzzle with the band. Using CSMA-style systems where everyone is code-divided, some point-to-point cell phones and some "common channel"-type equipment functionally identical to current two-way radios would solve more problems than simple reallocation.

This would also erase the decades of creeping allocation and other idiocy which put the band into such a region of intermodulation.

(And while they improving the efficiency of the band, give us back 824 through 848 MHz on scanners. The "cell block" law is just moronic, ineffective, and today an utterly unnecessary reminder of how laws are bought to cover up for insufficient design.)