3rd party firmware

Is there any 3rd party firmware which can allow a WRT54G to modify the frequencies used outside of the standard selectable ones? I'm working on a ham radio project and need to move this outside of the standard channels as well as increase the power significantly.

Anyone have any ideas on this?

thanks Steve

Reply to
Steve Smith
Loading thread data ...

Meanwhile, at the alt.internet.wireless Job Justification Hearings, Steve Smith chose the tried and tested strategy of:

It's not just a matter of whether or not the UI allows you to enter some arbitrary frequency, the underlying radio hardware has to support it. And the most likely answer is "no it doesn't".

Even a software-defined radio such as a USRP2 requires different RF modules for input and output, so don't expect anything clever from a WRT54G router!

Reply to

#begin lecture; There's no way you can use either 2.4 or 5.8Ghz Part 97 frequencies with high power an not trash users of the lower channels on 2.4GHz or the ham satcomm frequencies. The occupied bandwidth of 802.11g is about 23MHz which is guaranteed to create interference on the lower channels. Keep your power low, and nobody will complain. Create an alligator (big mouth, small ears) and I'll be the first to notify your local Official Observer. If you want to play lawyer and claim that a licensed service takes precidence over unlicenced, you'll find that the inertia of perhaps 200 million wireless users, doesn't do well against perhaps 700,000 licenced hams. Even the ARRL knows enough not to even mention it. #end lecture

To the best of my limited knowledge, there is no easy way to modify the frequency range of a WRT54G below the lower end of the band. The channel assignments are by lookup table inscribed in the Broadcom chip. In addition, you'll be doing battle with a ceramic bandpass filter on some models. I've read some suggestions of using a transverter, but have not seen any indication that it's been done.

Reply to
Jeff Liebermann

Any idea if the WRT54G (of some models, thereof..) can unlock the additional 802.11 "channels"? In ths USA we're limited to full power on ch's 1 throgh 11, but can (if the equipment allows) use a lower power setting on a few others.

And, of course, there's the Rest Of The World.

Hence I'm guessing [tm] that the WRT, which is manufactured "elsewhere" and intended for a worldwide market, might have this capability but has blocked it via software.


Reply to
danny burstein

Yes. I just looked at DD-WRT v24 pre-SP2 build 14896 and found that you can set the radio to Ch 1-14.

Yep. You might want to find out what services you're jamming before you do this:

Pg 37-38. Mobile satellite, space to earth. 2483.5 to 2500.

Yeah, you're going to be really popular using those in the US.

Reply to
Jeff Liebermann

First off, higher power will not effect home users nearly as much as anyone might think at first, by utilizing point to point antennas high enough for line-of-site. There are many wireless commercial installations on the same frequencies which everyone else is on, and most people don't even realize they are there. So, this argument doesn't hold much water in any form. Second, the FCC DOES specifically state that any and all part-15 users of a shared frequency MUST just deal with interference caused by licensed services no matter what service it is. And it just so happens that hams do have "primary" use to this allocation. This does not mean that public interference would be tolerated in any way - especially if the ham knows that such interference is happening, but just that if there was any incidental interference, it would have to be tolerated. I am full-well aware of this and the intention is to not interfere with non-ham users. Geesh, just mention high power - and everyone freaks like it's a bad thing, when in fact it it could be a better thing. The higher power will cause a more narrow beamwidth emmitting from a parabolic antenna - meaning less interference to ground based units.

the power I was suggesting was not the full 1.5KW anyway - that would be really stupid if you think about it, for many reasons, not just health and welfare.

This project is for EMCOM for data comms within disaster areas to areas nearby just outside of the scene to provide live video, data, and digital voice at a much lower cost that dstar. So, the power levels would not be over mayby 5 watts with erips of maybe 25 or 30 watts. This would give much more sufficient throughput that packet and FM would do. Using the 5 Watts would require a licensed operator though, according to part15 rules.

There is no reason to jump all over someone for asking a simple question with comments not related to the question in fear of stepping on toes where there aren't any to step on in the first place.


Reply to
Steve Smith

2483 to 2500 are not licensed to part97 users either, so for a licensed ham, they aren't of interest.


Reply to
Steve Smith

Steve Smith wrote in news:i7fujc$4uu$ snipped-for-privacy@speranza.aioe.org:

IIRC, for 2.4Ghz, the FCC limits are....

4 Watt (+36 dBm) for multi-point, 200 W (+53 dBm) for point- to-point

So if you're doing just a PtP.......

Reply to

True. They are of interest to Steve Burnstein, who asked about using wi-fi channels 12-14. Channel 14 is centered at 2484MHz and extends to about 2495MHz due to the 22MHz of occupied bandwith used by wi-fi.

Reply to
Jeff Liebermann

Sorta. Take your line of sight and draw a line on a map extending in both directions between the endpoints. Does it cross any metro areas? if it does, your high power (or more correctly high EIRP) radio is going to have quite an effect on users along this line. RF doesn't magically stop when it gets to the receiving station. If you're doing this in the middle of nowhere, no problem. If you're doing this in a crowded area, there will be problems.

True. That's what a spectrum analyzer is for. Ummm... you have done a site survey at both ends of your link to be sure?

It's not an argument. It's a warning. You do what you want. Your system can probably tolerate using normal wi-fi levels, without going to high power. Have you run the numbers to see if high-power is really required?

Use 20dB for a reasonable fade margin and see if you really need "high power".

Correct. The only reason that's still in the rules is that ham operators in general have avoided any conflicts that will get the attention of the FCC. As I stated, should such a conflict come to their attention, it's highly likely that 200 million wi-fi users will prevail over 700,000 hams. Do you want to be the test case?

Fortunately, that's never been tested in front of the FCC. My prediction is that if it were tested, the rules would magically be reinterpreted to the detriment of ham radio. Do you want to be the test case?

I was involved in a minor version of this debate between a WISP (wireless internet service provider) and a ham operating satcom on

2.4GHz. Fortunately, the FCC was not involved. The solution was to narrow up the barn door front end of his receiver with a minimal BPF and LNA. At the time, someone obtained an unofficial verbal opinion from someone at the FCC, which stated that should this become an issue, the various wi-fi manufacturers would probably insist on a rules change in order to secure their position. Do you want to become the test case?

Intentions are a good things. I prefer calculations. Did you determine if high power is really necessary?

That's high EIRP, which is quite different. In a wi-fi newsgroup or mailing list, high power means exactly what it says. It means you are running power exceeding what is normally used by others on the frequencies. Up to 4 watts EIRP from and omni, or

5 watts to the antenna with 25 watts EIRP is an antenna gain of about 7dBi. Nobody makes a dish with that low a gain, but there are plenty of omnidirectional antennas with 7-8dBi gain. Somehow, I don't think you'll be running a dish or directional antenna with those numbers. If you were, a 24dBi barbeque grill dish, with 5 watts in, will yield an EIRP of 250 watts.

Using it in the Part 97 only part of the ham band, as you suggested, would also require a licensed operator. Incidentally, are you prepared to monitor all the traffic that is passed to insure that either the source and destination are licensed hams as required of all packet radio operator? It's not really being enforced, but most packet systems are making varying efforts to comply. Same with the baloney about "authentication" which emerged from an effort to clarify packet identification requirements, that backfired.

If DSTAR DD speeds are sufficient for your purposes, you might consider using frequency hopping industrial radios that will do up to about 128Kbit/sec. It uses a narrower occupied bandwidth than

802.11g with much better range and sensitivity on 2.4Ghz and 900MHz. I use these for point to point data links that need range and reliability. For example:

I'll leave it to your good judgment to determine what's best. All I've done is enlighten you as to the regulatory dangers and suggest that you do some re-thinking and calculations. Nobody is going to complain about a temporary EMCOM installation from a portable location during an emergency or a drill. However, a fixed location might be a very different story.

Incidentally, if you want decent range (and folliage penetration) you might look into 900MHz:

I don't think I jumped all over you. What I did was supply a suitable answer to your question, explain roughly why, and add a warning about the regulatory dangers of continuing with your approach. I don't think I could have been more diplomatic or tactful.

Reply to
Jeff Liebermann

Thanks for this second reply - this was more informative than the first one. To answer some of your points, no we have not done any tests yet or calculations as far as how much power we need. And I was intending to use no more than 4 or 5 watts to the antenna anyway - which would definitely reduce the possibility of interference. To me, 5 watts is high power when considering the typical home-user is running about 200 mw on most routers.

The dish idea was for point-to-point between larger areas where omni antennas of lower gain would suffice for local use in the vicinity of the point-to-point locations. This would give larger area of coverage for locals maybe from city to city for testing purposes.

Keeping non-hams out is accomplished by using frequencies not standardized by routers out of the box and with no connection to the internet at any point, there really wouldn't be anything there for a non-ham to bother with. Anyone that did happen to get into the system wouldn't find much there to keep him/her occupied very long.

Also using AMPR-IP block the IPs aren't routable on the internet anyway and most non-hams have no clue what they are anyway. Removing DHCP is one other way to help with regulating the connections to the system.

At the moment this is nothing more than an idea, and may never materialize, but I'm first looking for the firmware to see if this is even possible to accomplish.

Most of your statements were correct and right on the money - and I appreciate your links also.

Thanks Steve

Reply to
Steve Smith

Most home wireless routers run about +17dBm (50mw) to an omni antenna with an EIRP of maybe +20dBm (100mw). Most laptops are running +15dBm (32mw) with an EIRP of maybe +7dBm (6mw). There are home routers that run 200-400mw, such as the common 2wire 2710 which runs about +26dBm (400mw), but due to the inefficient antenna antenna, has an EIRP of maybe +26dBm (400mw).

I don't have a huge objection to your system if it's point to point, with directional antennas at each end and roughly identical TX power levels. The problem is when you try to use high Tx power to communicate with a much lower power radio, such as a laptop. This creates an "alligator", which is an animal with a big mouth and small ears, where your transmit range is MUCH farther than your receive range. Compound this with an omnidirectional antenna, which will obliterate more square miles of area than your receiver can possibly need to cover, and your system is no better than a jammer.

This is nothing really new to wi-fi, where the legal limits exceed the practical limits. For example, several commerical mesh network vendors offer high power (1 watt into 6dBi omni) systems. Most have had to crank down the TX power in order to prevent their own transmitters from interfering with their own users. Incidentally, don't ask me about mesh networks or you'll get another rant.

You're thinking in terms of transmit coverage. Now, switch directions and think of what a tower located omni will do for receive. If you're using non-overlapping wi-fi channels (1, 6, and 11), a high omni is going to hear lots and lots of garbage from all over the city. Sure you can cover the city with your transmit signal, but can you excavate the desired remote user from out of the QRM? Probably not. This is why most WISP operators use sector antennas. Incidentally, if you want to build your own sector antenna, Google for AMOS and Franklin wi-fi antennas:

Cheap, easy, and works well.

Every system I know has an interface to the internet, even if it's just SMTP/WINLINK for email. The problem is not use by non-hams as your use of Part 97 frequencies will require a ham operator. The problem is proscribed non-ham related traffic going through the system such as non-ham equipment or services offered for sale and non-emergency work related messages that violate Laura Smith's personal interpretations. Good luck keeping it clean.

Net 44 (APMPRNET) is problematic for various reasons. The really dumb one is lack of broadcast packet control. If you setup a gateway to Net 44, you'll find that 99% of the packets are ARP (address resolution protocol) broadcasts, where someone is trying to connect to another net 44 system, and sends a "what owns this IP?" broadcast to all over the internet. There are other problems.

Private IP selection has it's own horrors. 10.xxx.xxx.xxx is used heavily by corporate medical and government internal LANs. If you're going to use a VPN into one of these systems, you can't risk IP duplication, so 10.xxx.xxx.xxx is out. 192.168.xxx.xxx is heavily used by home and small biz systems. The chances of IP duplication is also high but controllable if you use odd Class C IP blocks. I use

192.168.111.xxx in my office, which was fine until I found a customer that had their LAN setup the same way. Your best bet is 172.16.xxx.xxx by default.

Well, you have to start somewhere. Perhaps an existing and working example might be helpful:

Note the use of non-routeable IP's, not Net 44 ham IP's. There are also some really slow IP links via old Metricom radios on 900Mhz. It's quite an extensive EMCOM network with multiple internet gateways. I don't agree with everything they've done, but it's legal and doesn't use Part 97 only channels or high power.

Bug me if you need techy help with your proposed system.

Reply to
Jeff Liebermann

You stated that all systems are connected to the internet - but that isn't true. If there isn't any connection to the internet from any of the system links, then there isn't an inter-connection to the internet. So there isn't any reason why the 44 Net won't work. And since there is already coordination for the 44 net, part of that issue is already handled and there are plenty left to work with since packet tcp/ip networks are kind of a thing of the past for the most part.


Reply to
Steve Smith

What I said is that all such systems eventually have a connection (i.e. gateway) to the internet. Here's an old list:

Note: Most of these won't respond to pings. I just ran an nmap scan of (the local Net44 IP block) and found that (wr6jpl.ampr.org) responds to pings (after multiple timeouts), and not much else. Still, it might be interesting to see who's active in your area with something like: nmap -sP for Ohio. Hmmm... not much:

Incidentally, your local AMPRNet IP address admin may not be thrilled with passing out a block of IP's for your proposed network. For Ohio:

Looks like you get /27 or 30 hosts per subnet in Ohio, which should be adequate for your project. (In California, we only get two IP's per call sign).

Checking mine: C:\> nslookup Default Server: DD-WRT Address: > wb6ssy.ampr.org Server: DD-WRT Address: Non-authoritative answer: Name: wb6ssy.ampr.org Address: (WB6SSY was my previous call sign. I guess I should fix that.)

If you don't plan on dealing with the internet connectivity problems, then it will bite you later. I don't really know how each gateway and user deals with the regulatory issues. My guess(tm) is that they largely ignore it, control spam, deal with abuse individually, and hope that nobody notices. In effect, you'll be running a small ISP, with all the administrivia and management overhead that it involves. Note that you can use for testing without getting any assigned IP's (along with,, and

Don't bother with 1980's packet technology. Look at what people are doing with Winlink and Winmor:

for a clue. There are also systems that use surplus telco hardware for backhauls, mux/demux, interface, VoIP, signaling, etc.

If it's not connected to the internet, what good is it? Most of the traffic will come to/from the internet. More simply, it's really difficult to pretend that the internet does not exist.

See my comments on broadcast packet control. Not a problem with high speed links. Big mess with TCP/IP on top of AX.25 at 1200 baud or with HF at maybe 100 baud.

Ask the SVWUX people what they're doing. Listening to the voice of experience is a great way to avoid repeating past mistakes.

Reply to
Jeff Liebermann

If this is for ham radio, then why have an internet connection? We already have those - echolink, irlp, dstar to name a few. Not sure it is necessary for a total amateur network to have internet connectivity. If I remember right - in the beginning of echolink - may hams screamed this was not really radio because of the internet connections. Now, it seems to be ok since DStar because that seems to be the backbone of this system. My point was to have a separate radio only network with digital voice, messaging, and pretty much everything packet provided at much higher speed with the protection from the internet and problems that it brings into the situation.

Winlink, I don't like for several reasons, but there is no need to get into that here. Winlink doesn't provide much anyway other than internet email, which is not the point of this system.

Thanks for all of your input though, I have been taking notes and reading your links.


Reply to
Steve Smith

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.