Saturday, May 31, 2008

Two-meter radio instead of cell phone

A recent search term that brought someone to my blog was "2m radio instead of cell phone". I touched on this topic somewhat in a previous post. However, I didn't talk about particular topic in much detail. There's a whole bunch of differences between using a two-meter handheld for personal communications, and using a cell phone for personal communications.

First, let's ignore the presence of infrastructure. In the absence of infrastructure, the two-meter HT is a lot more useful, but only because the cellphone is just about useless. The only way I could communicate with my cell phone in the absence of service (other than throwing it like a rock) would be to use its flashlight to send messages in Morse code. The HT, on the other hand, will be useful even in the absence of any infrastructure to communicate with someone else (within a relatively small distance) who also has a two-meter HT, in what is known as "simplex mode". So in the no-infrastructure case, the HT wins easily.

But that all changes once you bring infrastructure in. The "ordinary" infrastructure that one would have access to with a HT would be a two-meter FM repeater. There's a lot of these out there, although there's no coordinated way to find out other than looking up your location in a database. Most notably, the HT won't find one on its own; the operator will have to know from a third party database what frequency and squelch tone are required to access the repeater. Also, the person you want to talk to will also have to be listening to that repeater (and not some other repeater), which means they'll have to be somewhere within that same repeater's geographic coverage area, typically a range of 30 to 50 miles. On the other hand, the cell phone allows you to talk to anyone with a phone (cell phone or not) anywhere in the world, knowing only their phone number. Neither you nor they need to know which cell tower you're using. The network takes care of all of this. There are technologies that exist in ham radio that bring the 2m HT closer to the cell phone, things like autopatches and repeater linking, but autopatches are increasingly less common (mainly due to cell phones) and repeater linking not widespread enough to come close to the capabilities of the commercial cell phone networks.

In addition to the above, there are several other major differences. The cellphone is private (encrypted with newer standards), full-duplex (both sides can talk simultaneously), and exclusive (you do not share your channel with anyone else). None of these is true of the amateur HT. All communications on a HT (whether simplex or through a repeater) are over public, unencrypted channels as encryption is forbidden (with only two exceptions) in the amateur service. Typical amateur handhelds are half-duplex only, which means you can either talk, or listen, but not both at the same time. There is no practical way to do full-duplex within a single band; the filtering required would be technologically challenging to put into a handheld device. Finally, you will be sharing the frequency and the repeater with whoever else might be using it.

In short, the 2m HT is no substitute for a cell phone in ordinary circumstances. There's only two ways in which the HT is superior: it works, at least somewhat, even when you don't have an available cell network, and you don't have to pay for a service contract in order to use it. It's the former that makes the HT extremely valuable when you're off in a remote area or during a communications emergency.

The main value of having all these hams with handhelds out there is so that when there is a disaster, there will be people out there with the equipment and the experience needed to effectively provide emergency communications. Otherwise, it's pretty much just fooling around with radios. Which, to be fair, is a great deal of what hams do.

Sunday, May 18, 2008

More on sliding dovetails

I finally got around to taking pictures of the two most recent carpentry projects.

The first is the step. The top is made of three planks of four-quarter panga panga, edge glued. The edges are rounded over above and below with a quarter-inch roundover on three sides; the side to the rear has merely had the edges broken. The sides are also cut from panga panga. The underside of the top has a double-blind groove cut to receive the front, which is correspondingly notched to fit into this groove. In addition, sliding dovetail dados are cut in the underside of the top and in the back of the front to receive the sides. To assemble, first the sides are slid in on their sliding dovetails, and then the front slid up on the dovetails cut in the front of the sides until it nestles in the groove in the underside of the top. That's it; the machining is tight enough that no glue is needed. All surfaces are finished by random sanding to 220, and then two coats (sides and front) or four coats (top) of a 50/50 mix of tung oil and citrus solvent. There is no back; however, the entire assembly rests on top of a core made of solid cedar 4x4s endcuts, glued together in a three by ten array. Nonslip shelf covering is above and below the cedar core for cushioning and to reduce slippage (which is nonexistent).

The second is the air conditioner stand, which I wrote about previously. The visible parts are made of standard-quality 12x1 pine planking cutoffs, except for the legs which are 2x2 cedar, and the stretcher at the bottom of the legs, which is panga panga. The table top, which cannot be seen in the photo, is a sheet of four-quarter melamine with a simulated pine finish. As it stands this project is still not entirely finished; the surfaces have only been sanded to 150 and no finish has been applied, the legs have not yet been mechanically attached (although it barely seems to matter), and we plan to put some sort of trim along the top to hide the relatively ugly ship-lap joinery between the sides and the front.

There's also a bunch of flower photos which I've uploaded to Flickr. Feel free to browse and enjoy.

Friday, May 16, 2008

Connecting to the Internet with Amateur Radio

Yes, a lot of people seem to want to do this. ICOM even claims that you can do it using D*STAR, although they're kinda vague on how it works. There are a few problems with doing so. The most commonly cited one is the "third party rule" (§ 97.115), but this actually isn't a major obstacle in most cases; the real obstacle (in the US, at least) are the list of "prohibited communications" in § 97.113.

The third party rule, as implemented by the FCC, only comes into play on international communications; that is, when one of the two amateur stations in the communication is in the United States (or other area regulated by the FCC) and the other station is not so located. So the third party rule would be a problem if you were in the US and trying to use a radio to communicate with a station in, say, Germany, and then from there to connect to the Internet. However, if both of the stations involved are in the United States, this doesn't come into play. The third party rule says nothing about where the third party in the communication may be located; as far as the radio regs are concerned, it doesn't matter that you're going to a website in Armenia, as long as your radio communication isn't going there.

However, the real problem is that any connection to the Internet would have to be continuously supervised to ensure that no prohibited communications took place over the link. And there are so many things that are prohibited: music; anything encrypted; any message in which the operator has a pecuniary interest other than the occasional sale of radio equipment; obscene or indecent language. Imagine how hard it would be to avoid all of these while browsing the Internet.

On top of that, there's the general prohibition in § 97.113(a)(5) on "[c]ommunications, on a regular basis, which could reasonably be furnished alternatively through other radio services." There are plenty of alternative services that can provide mobile connections to the internet, in most circumstances, at least. There's probably a few edge cases where this won't get in the way (out of the way areas, or during a communications disaster).

The problem that this creates is that the only time that the regulations really allow a ham to connect to the Internet would be during a communications disaster. But because we aren't allowed to do it under ordinary circumstances, there's not a whole lot of incentive to have the equipment and infrastructure in place to do it in the event that we do need it. And once-a-year drills on Field Day aren't really a good test of how the system would perform in an actual emergency. We've become increasingly dependent on the Internet for ordinary and extraordinary communications needs; it's almost certain that in a major disaster getting Internet access active in a disaster zone is going to be a priority, and the FCC, by not letting us do this during non-emergency conditions, makes it that much less likely that we'll be able to when it really does matter.

The other day I came across this old article from 2001 about an effort by hams to provide a backup Internet infrastructure using ham radio services. I don't think they're still around; a search for their organizational name (the Emergency Communications Network) turns up only a company in Florida that seems to be providing telephone-based emergency communication services. (Fat lot of good that'll be during a telecom emergency when the telephone network is down.) I have no idea what happened to them or their idea.

So, the long and the short of it is basically that you really can't use ham radio to connect to the Internet... but we would probably be better off if you could.

Thursday, May 15, 2008

Two meter handheld range

Here's another post inspired by a search hit. Some fine soul hit my blog today on a search for "expected range on a 2 meter handheld radio". This search would have brought our feckless reader to this post, which does not fairly answer the question. So, in the hopes that the next person to do this search actually learns something, I'll essay to answer this question.

The answer, of course, is "it varies". The amateur radio two-meter band is a VHF band, and as such is almost entirely line of sight. (Sporadic E skip, tropospheric ducting, meteor scatter, auroral skip, and earth-moon-earth are all quite difficult with an HT, so I won't dwell on them here.) So, basically, the range of any 2-meter radio is going to be limited by the horizon. However, because the atmosphere refracts radio waves considerably more than it does light, the effective radio horizon is about 15% further than visual horizon. So that establishes one of the limits on range: the receiver at the other end must be above the effective horizon. This is a function of four things: the altitude above ground of the transmitter, the altitude above ground of the receiver, the distance between the two, and the terrain between the two. The effective radio horizon for a given location, assuming flat terrain, is about 4.11 kilometers times the square root of the height of the antenna (in meters) above ground. So two stations with their antennas each one meter above ground will be in each others' horizons if they are closer than about 8 kilometers (a typical handheld-to-handheld case). If one of the stations is, instead, a repeater with its antenna 60 meters above ground, the range is nearly 36 kilometers. If one of you is atop the John Hancock Center in Chicago (344 meters), the range would be a whopping 80 kilometers. Of course, all of these are assuming flat terrain; if one of you in on top of a mountain, then that will also increase range, and if there's a mountain between you then you will be out of luck.

This isn't the end of the discussion, though. Not only do you have to have to be within each other's radio horizons, you also have to have enough power to survive path loss. Path loss represents the reduction in strength of the radio wave as it travels through space. Much of this is simply due to the inverse square law: as the wavefront grows in size, it occupies more space without having any more power, and thus has a lower power density. The receiving antenna's size doesn't change, so if it's moved further away it will receive less of the transmitted field. This is "freespace path loss", and if this is all you were facing loss would simply double twice with each doubling in distance (a loss of 6 dB for each doubling in distance, or octave). However, in real situations, other factors also contribute to loss, and in average conditions the effective loss in VHF is closer to 7 dB per octave, with a base at 1 km of about -71 dB. Most HTs will be able to just barely receive a signal at about -120 dBm, and should give suitable performance at about -100 dBm. This means that at 1 km, to have acceptable performance, you'll need to transmit a signal of -29 dBm. That's not even two microwatts. (In practice, you wouldn't get that far on two microwatts, because your HT's antenna probably sucks. More on that below.) Even at the 80 kilometers long-range case above, the path loss is only -115 dB; a signal of 15 dBm, or about 30 milliwatts, would be sufficient. Even with antenna losses, that's possibly within the typical 5 watt (37 dBm) capabilities of your average HT.

However, this discussion ignores one critical factor: the poor antennas that most HTs have. In practice the antenna of a handheld radio has negative gain in most operating environments, and you can expect to lose anywhere from 2 to 5 dB due to this issue alone. Obstacles (other than terrain) can also wreak havoc with the signal; if you're in a car, for example, expect to lose as much as 20 dB due to absorption from the car; other things that can reduce range include trees and buildings. Also, when transmitting within one wavelength of the ground (which for an HT is almost always) a significant portion of the signal is lost into the ground; this can account for up to 10 dB of loss.

However, it remains the case that in almost all cases, radio horizon, and not power, is the primary limitation on VHF range. So, in ordinary conditions, your 2m HT range is going to be about 8 km HT-to-HT simplex and about 25 to 40 km HT-to-repeater (depending mainly on the height of the repeater's antenna).

P.S. All numbers in this article may be wrong. I've tried to get them right but it's late and I may have made mistakes. If you do spot anything wrong, please feel free to hit me.

Tuesday, May 13, 2008

Too many repeaters

Is there some law that requires ham radio sites to have atrocious web design? Some of the sites I've been at recently (and I won't name names, not right now) have just horrible web design. Blinky graphics, animated horizontal rules, busy background patterns... it's Geocities all over again!

One thing that I've been frustrated by recently is the lack of useful repeater listings for Chicagoland. Most of the listings I can find just list the city or town for the repeater. This is multiply frustrating for me where I am. First, there are just dozens of little suburbs around here, many of which the names of which I forget, and scanning a list of repeaters in Illinois sorted alphabetically for the ones near me is challenging and tedious. Second, a repeater in Chicago may or may not be close enough for me to use; Chicago is a big city and a repeater might not cover the entire city if it's not high enough or is on the wrong side of the Sears Tower or something. Third, many repeaters are listed with a city/town of the residence of the operator or the trustee, not the actual location of the transmitter (which may be miles away and several towns over). K5EHX's repeater list tries to rectify this somewhat, but this database suffers from the fact that most of the repeaters have only a default lat/long of the city center of the listed city (which, as noted, may be incorrect). (Not K5EHX's fault; he can't create data out of thin air anymore than anyone else can.)

I am therefore going through the list of repeaters for my location, trying to update the information as best I can, using the web and emailing operators to the extent possible, and going on air to try to get info when I can't find the info via the web. This is proving to be tedious, but at least it will hopefully help someone.

If you know a repeater that isn't listed in K5EHX, or is incorrect, please add it/fix it as appropriate! Chicagoland isn't the only place where this data needs work.

Saturday, May 10, 2008

Toll Avoidance, and Gain Conversions

I occasionally look at my referers to see what people are searching for when they find my blog. There were two in the recent list that stuck out to me: "convert dB to gain" and "chicago route around no tolls". I've obviously touched on these before, or else people wouldn't be finding my blog so easily, but I haven't talked about either directly, so here we go.

Decibels are a dimensionless unit used to relate a measured level against a reference standard. The actual unit is the "bel"; one bel represents a tenfold increase in level as compared to the standard. A decibel is a tenth of a bel, and therefore represents an increase of about 25%, since 100.1 is approximately 1.25. To convert decibels to an absolute gain figure, one simply raises ten to the power of the gain ratio expressed in bels. If G is the gain in decibels, then g=10G/10 is the absolute gain multiplier. The reverse is done with logarithms and is left (for now) as an exercise for the reader.

As for finding a route around Chicago that avoids tolls, this is more complicated. In part, it depends on where you're trying to get and from where. The most common situation where this comes up is when someone is going east to west (or west to east) on I-90 and is routed through Chicago. (I-80 doesn't actually go into Chicago.) Interstate 90 is toll from the Wisconsin state line to just outside Chicago, and then again on the south side of Chicago; avoiding it is not easy. In addition, using the freeway part of I-90 to go through Chicago is not terribly fun; the Dan Ryan is widely regarded as one of the least pleasant freeways on which to drive.

Avoiding the the Skyway toll is easy; just enter Illinois on I-80/94 instead of on I-90. If you're already avoiding tolls you've done this anyway since I-90 in Indiana is the western end of the Indiana Toll Road. Avoiding the Jane Addams toll is much harder as there are no suitable alternative routes for most of its length. (You can avoid the east terminus toll by using I-290.) A lower-toll alternative is I-80 and I-39; you only pay one toll on the portion of I-80 that is overlaid on the Tristate and one on the portion of I-39 that is overlaid on the Jane Addams. You could avoid these by using lesser surface routes, but none of these alternates is very enjoyable. If you're heading far enough west, consider taking I-80 all the way to I-35 (in Iowa) to reconnect with I-90 in Minnesota. If you're coming up from central Indiana, consider using I-74 to I-39 (in Bloomington) or I-80 (in the general area of the Quad Cities). If you are heading up into the Wisconsin coast (Kenosha or Milwaukee), you can try using US 41 instead of I-94 from the north side of Chicago.

Hope this helps someone. If not, oh well.

This post is not about pedophilia

I'm very much aware of all the furor, and do not wish to add to it further. It seems, however, that negative public opinion of "one of the most prolific reputation and defamation engines on the internet" is spreading fast:

(Taken from the May 10, 2008 issue of Secret Asian Man by Tak Toyoshima at; link will only be good for two weeks, sorry.)

Tuesday, May 06, 2008

Ham Radio, Internet, and the Cell Phone

A recent commentator in my blog remarked that there has been a declining interest in amateur radio, citing as a reason for this an increase in the availability of computers and the Internet.

While I think wider availability of the Internet may be a factor, I don't think it's a major one. The Internet is not really a substitute for amateur radio. While both do permit relatively inexpensive long-range communications, the nature of those communications are quite different. Amateur radio permits mostly immediate, relatively low bandwidth, personal contacts between people who would otherwise likely never meet. While there are those in amateur radio who pursue packet radio and other store-and-forward technologies, even in their heyday these were not a major aspect of the hobby for most people involved. (In part, this is due to the regulations that strictly limit third-party traffic on ham radio frequencies, which has amongst their effect making it impossible to use ham radio to connect directly to the Internet.) The main appeal to most people in the avocation has always been talking to other people at a long distance, from wherever they might happen to be.

No, it's not the Internet that is killing ham radio. If anything, the Internet is helping it. What is killing ham radio, at least in the West, is the cellphone. Prior to the cellphone, there were very few options for people who wanted reliable mobile communications. There was, of course, CB, but it's only good for a few miles if you follow the rules (which most of the people using it don't, sadly) and only a few more miles after that even if you don't. And CB is also full of obnoxious loudmouths who seem to exist only to hear themselves talk (they're not interested in having an actual conversation), making it nearly useless as a personal communication service anyway. You could, of course, obtain a business class license and use "color-dot" radios, but those licenses are not particularly cheap and the radios (at least back before 1990) also not all that cheap. Practical for a business; not practical for people who just want to be able to talk to one another when they're out and about.

There was one final option: you could get an amateur radio license and use VHF or UHF handhelds. Relatively affordable and reasonably sized two-meter (144-148 MHz) handhelds were available for the amateur service long before cellphones were widely available to the public. In simplex mode a two-meter HT (handheld transceiver) has a modest but decent range; used in conjunction with a well-located amateur repeater station a group of hams with two-meter HTs could easily maintain contact with one another over the area of a mid-sized city. The introduction of the "no-code Tech" license in the United States in 1991 helped here; this license gave broad privileges on VHF and higher frequencies without requiring the licensee to learn Morse code and made it much easier to attract technically competent people to the hobby who didn't want to take the time to learn Morse (a technology which computers have largely made obsolete). This did lead to a surge in licensing, but many of those who got their licenses in 1991 have since drifted away.

Enter cell phones. While the idea of cellular radio was invented in 1947, and the cell phone itself in 1973, the early models were large, expensive, and had very limited service areas. (Even as late as 1995 a cell phone was substantially more expensive than a 2m HT, and the HT has no "service contract" that has to be paid monthly; even if you used a repeater and the repeater operator asked that you join the club and pay dues, the dues were probably less than what you'd pay for a cell phone contract.) It wasn't until the late 90s that cellphones became widespread, and not until the early 2000s before they became ubiquitous. Now, for a relatively small fee, can get the same thing using a cellphone, and on top of that call anyone else they might care to talk to as well. Given a choice between a VHF HT or a cellphone, the cellphone wins nearly hands-down for this sort of communication in almost all situations now.

That's not to say that there's nothing interesting going on in the amateur radio world; it's not just a bunch of nerds in their garages beeping at one another in Morse code (although, to be fair, there still is some of that going on).

One big thing right now in radio seems to be software-defined radios (SDRs), and this is an area where free software can and should be getting involved. Traditional radios have used discrete, specialized circuits designed specifically for the application at hand, which means a given radio is good for, say 2m FM reception but not much use for anything else. "All mode" radios can receive and transmit multiple modulations, but they do this typically by having physical (or electronic) switches that gate the signal to the appropriate modulators and product detectors, which introduces the possibility of loss and makes the radio much more complicated (and therefore expensive). SDRs are different.

In its ultimate form, an SDR is a analog-to-digital converter (ADC) attached directly to an antenna. As there are few ADCs that can process signals at the frequencies used for many forms of radio (the current limit is around 40 megahertz), and there are other practical limitations, what is done instead is a traditional front-end is used to select a portion of the radio spectrum to receive, which is then preconditioned using (typically) a superheterodyne mixer to produce an intermediate frequency signal suitable to be shoved into the ADC. Everything else is done by the ADC. Transmitting works the same way, in reverse; a digital-to-analog converter (DAC) generates an IF signal that is then either merely amplified and put directly on the antenna (for sufficiently low frequencies) or fed into appropriate mixers or transverters for VHF and up. Since all the digital signal processing is done by the computer, the radio can have all of its functionality, other than its fundamental RF receive and transmit frequency capabilities, reprogrammed merely be changing the software load. Since virtually all computers out there today have ADCs and DACs in them, in the form of a sound card, this is something that most anyone with a computer can do with minimal investment. For somewhat more investment, there are hardware solutions (such as the Universal Software Radio Peripheral) that offer more capability by providing more powerful ADCs and DACs than are found in your average sound card, along with a FPGA solution to allow for faster processing. And, of course, the opportunity for homebrew solutions is huge here, too.

There is already an open-source project working in this area: GNU Radio. One of the reason it's important for open-source advocates to get involved in this now is that SDRs are quite likely the direction the radio industry is going. And, of course, the solutions being created by the commercial manufacturers are locked-up-tight proprietary you-can't-touch-this approaches that limit freedom just as much as any other proprietary solution. We need for there to be open alternatives so that people will be inclined to adopt open approaches that let us use the hardware we pay for the way we want to. There are some freakingly neat things being done by the GNU Radio people.

There's also some very fascinating work in weak signal developing codes and methods that can punch a message through noise, which should appeal to those with interests in compression, information theory, and data reconstruction. Morse is surprisingly good for this purpose, but there are better codes out there. The computer has really revolutionized this area of the hobby as well (although not without some controversy).

I haven't mentioned Echolink or IRLP, either; both of these are fusions between ham radio and the Internet.

In any case, I don't think computers and the Internet are killing ham radio. The cellphone, while it certainly has had an impact, is not the real problem, either. I think the real problem is much more complicated and has to do with attitudes toward engineering and science in Western society more than anything else. Ham radio is very much alive in Japan and in much of the developing world, after all.

And we definitely need more hams; with our increasing dependence on telecommunications for everything we do, we need to have ham radio operators equipped, trained, and ready for when disaster strikes and the grid falls down. (Which is also why we really do need to keep Morse: it's just about the best way to punch a message through noise that doesn't require complicated equipment, just a trained operator and something that you can pulse on and off. All these other fancy digital modes are basically useless without a computer.) It's for this reason as much as anything else that I carry my HT with me most of the time.

The FCC last year removed all Morse code requirements (as have most other countries) for amateur radio licensing, and this appears to have caused another upsurge in interest. It'll be interesting to see how long it lasts.

Sunday, May 04, 2008

Bike racers and amateur radio

Last week the FCC denied a STA (Special Temporary Authorization) requested by Miller Motorsports in Tooele, Utah that would, had it been granted, allowed Miller to, inter alia, temporarily use a part of the 70 centimeter band allocated (on a secondary basis) to the amateur radio service for coordination of various services related to the scheduled holding of the HANNspree Superbike World Championship at the end of May, 2008. The FCC's denial seems to have been predicated on the idea that the proposed use of the amateur radio frequencies would unduly burden amateur radio operators, even though many other races have been granted STAs under similar circumstances in the past.

In this case, though, I think the denial is based not so much on the particulars of this request (which, from what I've been able to piece together, aren't that unusual for a racing event), but rather on whose behalf the request is made. Bikers, and especially offroad, rough-terrain bikers, apparently routinely use radios certificated only for use in the amateur radio service for personal communications without bothering with obtaining amateur radio operator licenses. See, for example, this thread on a biker's board, in which the lead poster raves about the merits of the Yaesu FTM-10R for use on his motorbike. This radio is a Part 97 certificated radio and requires an amateur radio license to use. Using it without a license is simply illegal. However, when someone came along and mentioned this, another poster basically announced contempt for the existing regulations and the original poster told the poster pointing out the illegality that they weren't interested in hearing about it.

In fact, I suspect the reason Miller wanted to use the ham frequencies is because most of the racers and crews in the planned race already have ham radios and so this precludes them having to go out and buy another radio. Ham radios are the only commercially-available radios in the United States that are "frequency-agile" (all other services are channelized), and most have trivial mods to allow out-of-band transmit, making them especially appealing to people who don't plan to follow the FCC's rules.

I find it unlikely that the attitudes expressed by those two posters are extraordinary in the motorbiking community. While I haven't read the FCC's decision (I haven't found a source for it yet), I wouldn't be surprised if their decision to deny Miller Motorsports' request for a biking event wasn't at least somewhat motivated by an awareness of a widespread streak of contempt for the FCC's jurisdiction and authority over wireless communications. If there's one thing that's bound to get the FCC to really go hard on your ass, it's demonstrating a lack of respect for the FCC's authority.

I'd really like to see what the full FCC decision is. I haven't found it anywhere on the FCC's site, but there's lots of nooks and crannies there and I could easily have missed it.

Update: I found Miller Motorsports filing for the STA in ULS, as well as their current license for a 450 MHz business pool service with 2 bases and 100 mobiles, and their other current license for 450 MHz business pool service with 5 bases and 70 mobiles, both in the same location. Unfortunately, ULS does not have the "offline" STA decision letter or the ARRL's informal objection.