Actually, Linux on x86 hardware is just as much of a "hardware-based" solution as any proprietary OS on proprietary hardware is. In fact, Linux on x86 easily outperforms most proprietary hardware solutions up to and beyond the Cisco 7x00 range, especially with the current Intel architecture systems. Vyatta (now part of Brocade) and others have proven that over and over for the last several years.
I hope we can establish something that's vendor neutral, rather than any type of single vendor lock-in.
Michael N6MEF
----
In the performance class we are operating in, "hardware based" is not really required. A suitable box (processor and ethernet interfaces) and software (e.g. Linux plus some usermode tools) can do the same thing.
I am not proposing hardware over software, but merely alternatives. I like hardware appliances purpose built for a single task. And, it is what I know. The beauty of this community is that no solution is off the table and everything works. It is nice to have options as they all can work together.
To further cross hardware and software barriers, some Cisco switches do run Linux.
At the end of the day, all connectivity to the 44net still relies on commodity Internet service as its backbone. That is the limitation we are all faced with.
Jesse WC3XS
Actually, Linux on x86 hardware is just as much of a "hardware-based" solution as any proprietary OS on proprietary hardware is. In fact, Linux on x86 easily outperforms most proprietary hardware solutions up to and beyond the Cisco 7x00 range, especially with the current Intel architecture systems. Vyatta (now part of Brocade) and others have proven that over and over for the last several years.
I hope we can establish something that's vendor neutral, rather than any type of single vendor lock-in.
Michael N6MEF
In the performance class we are operating in, "hardware based" is not really required. A suitable box (processor and ethernet interfaces) and software (e.g. Linux plus some usermode tools) can do the same thing.
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net http://www.ampr.org/donate.html
Here's one project -- http://hamwan.org
You might also want to keep an eye on http://nwdigitalradio.com (I am affiliated with this company.)
------------------------------ John D. Hays K7VE PO Box 1223, Edmonds, WA 98020-1223 http://k7ve.org/blog http://twitter.com/#!/john_hays http://www.facebook.com/john.d.hays
John,
Thanks for the links. The first one may have potential. The second, I'm greatly underwhelmed though I wish you well with it! Amateur radio, especially in the realm of data communications aka - bits streamed over the ether seems to have an issue with NIH (Not Invented Here) syndrome and constant reinvention of the wheel. While innovation is a good thing, what good is it if no one adopts and uses it. 56K while still faster than 1k2 is still reasonably slow. If you want to start with something that the masses will find interesting start at about 768k (half a T1) and go up from there. Dstar - while it claims to be open to anyone, Icom is really the only major manufacture making gear for it and only amateur radio uses it. Dstar is an island unto itself. The question I pose is how is Dstar or a 56k radio modem either an advancement of the radio art or an augment to what we have presently.... especially at the proposed price point. Hams are generally cheap. Where is the mass adoption of a 56k radio modem (the fact it hangs on ethernet was a good choice but why bother at such a bitrate....) coming from. no one adopted previous 56k modems either. also 70cm is to low in frequency. We have what a meager 30Mhz of spectrum there that's filled with repeaters, ATV, weak signal, EME, various radar QRM (and in some areas 70cm isn't even permitted for amateur radio use due to radar being primary), and contention by other users of the band. Compare this to the amateur microwave bands 13cm (upper part 2390-2450 - 60MHZ widely unused also paired with 802.11b/g), 9cm (3300-3500 - 200Mhz widely unused, just below current 802.11y spectrum with some radios tunable there), 5cm (5650-5925 - 275Mhz widely unused, overlaps 802.11a), 3cm (10-10.5Ghz - 500Mhz of which a couple beacons run here and there, a few khz of sporadic EME and weak signal activity occasionally happens, and a couple khz more of it gets used contest weekends, really close to existing commercial gear in frequency useage), 1.25cm (1.5Gb/s commercial gear fits squarely in this band and as hams we could repackage this tech for much improved EIRP & thus longer core network backhauls.). The 5Ghz and below stuff has multimegabit link speed, a price in the 100-300usd range at retail (same price as a decent HT or mobile rig), and is plug and play. The 10 & 24 Ghz stuff that would sit on high towers and mountaintops supporting the network core prices out to 1500usd/endpoint which is cheaper than a decent voice repeater and could be sponsored/supported the same way. really the cost factors here are quite disruptive and well within the reach of hams if we'd just DO. Based upon what the bell system guys were doing 50 years ago, it seems that 45Mbps over a 70-100km radio link appears it ought to be doable.... but really, 10Mbps over 50km would be a good start. Let's raise the bar a bit and get to building a RADIO BASED NETWORK!
Eric AF6EP
On Fri, Jun 28, 2013 at 4:28 PM, K7VE - John k7ve@k7ve.org wrote:
(Please trim inclusions from previous messages) _______________________________________________
Here's one project -- http://hamwan.org
You might also want to keep an eye on http://nwdigitalradio.com (I am affiliated with this company.)
John D. Hays K7VE PO Box 1223, Edmonds, WA 98020-1223 http://k7ve.org/blog http://twitter.com/#!/john_hays http://www.facebook.com/john.d.hays
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net http://www.ampr.org/donate.html
Is anyone running the buckmaster cd anymore? If so what's your 44-net address please? Thanks in advance.
Based upon what the bell system guys were doing 50 years ago, it seems that 45Mbps over a 70-100km radio link appears it ought to be doable.... but really, 10Mbps over 50km would be a good start.
We don't all have mountaintop microwave sites....
We have homes and ham stations in compromise locations with city lots, restricted antennas and big RF absorbing trees... 25W of 440 MHz will probably get through where less power at higher frequencies won't. The UDR56K is a good compromise real world solution that is probably the last best hope for AMPR data... The speed is good enough for IM, email like this list, megabyte files (I know first hand that's possible...) and facebook clones. That's way more than what we have now... The price per node (about $600) is high and will be hard for many to justify but it beats the competitors - ID1 is $1+K for 120K raw data with 10W at 1.2 G or D710 with 50W of 9k6 data UHF or VHF for $700.
Let's raise the bar a bit and get to building a RADIO BASED NETWORK!
A harder problem than the RF is the WHY.... I can download Zombie videos at 40MB (or ? - I haven't checked lately) from anywhere in the world - why should I futz around with silly HAM regulations and bureaucracy to do limited Ampr data... (FWIW I know the answers to that...)
Eric AF6EP
73 Bill, WA7NWP
I agree with you about not having the perfect locations for microwave frequencies. We have some nice sites on mountains and some not-so-nice sites with trees. So I was originally excited to read about the UDR56K.
*BUT* I, for one, am really disappointed that the UDR56K is also a linux machine. I think they screwed the pooch on that (although I understand about 300 pre-orders would disagree with me). Why do I say this? Because I already have a linux machine running JNOS (and a bunch of other stuff). I was excited about the prospect of switching our current backbone from 1200 to 56K. I was hoping to just plug it into another serial port on my existing linux machine. But it has no serial ports for my other TNCs/radios. (and adding multiple usb-serial adapters and a powered usb hub is just trouble waiting to happen). And I'm betting it doesn't have the horsepower that my existing linux system does, which runs an amprnet gateway, mail gateway, lots of iptables rules, spam and virus checking, etc., plus some automated network testing stuff. Whatever processor it does have, will be aging along the processor aging curve (speed doubles every 18 months) and to upgrade, you would throw out the radio, too - a bit of the baby and bathwater approach. In fact, forget about upgrades. If anything fails -- radio, modem, linux system memory, linux system storage -- I've got to replace the whole thing. All things fail at some point. So this just seems strange to me.
I wish they had just made the radio/modem and provided a serial port so I could connect it to whatever machine I want. I would have bought one for each of our machines, plus a spare (7). But I don't know what to do with what they built - namely a fast modem and radio hard-wired to a very limited, non-expandable linux platform. But again, I guess I'm in the minority because I understand they have about 300 pre-orders.
Oh well. Michael N6MEF
-----Original Message-----
We don't all have mountaintop microwave sites....
We have homes and ham stations in compromise locations with city lots, restricted antennas and big RF absorbing trees... 25W of 440 MHz will probably get through where less power at higher frequencies won't. The UDR56K is a good compromise real world solution that is probably the last best hope for AMPR data...
While I can't speak for the company making the UDR56k and I have stated previously where I think it missed the mark. putting a new RF modem on an ethernet connection is a good thing. It's something kantronics should have done when the 9612 was introduced to the world. The advantage of placing it in ethernet is that one simply assigns it an address, plugs it into a (poe) hub, and it works with little to no config on the host side. many if not most who might be interested in this probably would rather not spend a huge amount of time futzing with the operating system of whatever device takes bits to or from RF. personally, I'd like to see routing functions completely divorced and separated from host services functions. if you wish to offer application layer services to amprnet then fine, put them up on a host of their own. that host should not generally also route. That just seems the be good practice. The reality is that these ethernet to RF boxen will likely intigrate into your subnet just fine and be more reliable than either serial or usb (never mind I fail to find an actual hardware serial port on ANY PC built in about the last 5 yrs....) The reality is that these don't need much in the way of computer onboard and likely never will. I don't see their intent as being something that is intended to host services and the computer needed to move bits from ethernet to 56k wireless modem should not add much cost, I'm guessing 15-20usd/unit while adding much utility.
Eric AF6EP
On Fri, Jun 28, 2013 at 11:32 PM, Michael E. Fox - N6MEF n6mef@mefox.orgwrote:
(Please trim inclusions from previous messages) _______________________________________________ I agree with you about not having the perfect locations for microwave frequencies. We have some nice sites on mountains and some not-so-nice sites with trees. So I was originally excited to read about the UDR56K.
*BUT* I, for one, am really disappointed that the UDR56K is also a linux machine. I think they screwed the pooch on that (although I understand about 300 pre-orders would disagree with me). Why do I say this? Because I already have a linux machine running JNOS (and a bunch of other stuff). I was excited about the prospect of switching our current backbone from 1200 to 56K. I was hoping to just plug it into another serial port on my existing linux machine. But it has no serial ports for my other TNCs/radios. (and adding multiple usb-serial adapters and a powered usb hub is just trouble waiting to happen). And I'm betting it doesn't have the horsepower that my existing linux system does, which runs an amprnet gateway, mail gateway, lots of iptables rules, spam and virus checking, etc., plus some automated network testing stuff. Whatever processor it does have, will be aging along the processor aging curve (speed doubles every 18 months) and to upgrade, you would throw out the radio, too - a bit of the baby and bathwater approach. In fact, forget about upgrades. If anything fails -- radio, modem, linux system memory, linux system storage -- I've got to replace the whole thing. All things fail at some point. So this just seems strange to me.
I wish they had just made the radio/modem and provided a serial port so I could connect it to whatever machine I want. I would have bought one for each of our machines, plus a spare (7). But I don't know what to do with what they built - namely a fast modem and radio hard-wired to a very limited, non-expandable linux platform. But again, I guess I'm in the minority because I understand they have about 300 pre-orders.
Oh well. Michael N6MEF
-----Original Message-----
We don't all have mountaintop microwave sites....
We have homes and ham stations in compromise locations with city lots, restricted antennas and big RF absorbing trees... 25W of 440 MHz will probably get through where less power at higher frequencies won't. The UDR56K is a good compromise real world solution that is probably the last best hope for AMPR data...
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net http://www.ampr.org/donate.html
Hmmm. Well, if it works as a bridge, I guess that would be o.k. It would be like the old "TransLAN" bridges from the 1980s (for any of you who remember back that far) or, I guess, an Icom ID-1 today.
One thing I do appreciate is the option for separate transmit and receive connections so I can use an isolator on the TX side at some heavy RF sites. That's a nice touch.
But I still think they could have left off the underpowered linux part and reduced the cost.
Michael N6MEF
-----Original Message----- While I can't speak for the company making the UDR56k and I have stated previously where I think it missed the mark. putting a new RF modem on an ethernet connection is a good thing. It's something kantronics should have done when the 9612 was introduced to the world. The advantage of placing it in ethernet is that one simply assigns it an address, plugs it into a (poe) hub, and it works with little to no config on the host side. many if not most who might be interested in this probably would rather not spend a huge amount of time futzing with the operating system of whatever device takes bits to or from RF.
BTW, beginning some months back, when this email list was moved to a new machine (or some type of move took place), a weird thing started happening. More than half of the emails I get are now blank (such as below), but have three attachments:
"Untitled attachment #####.txt", which is the message text "Untitled attachment #####.htm", which is the message text in html "Untitled attachment #####.txt" which is the list footer.
This is very annoying.
I've been traveling a bunch so I wasn't paying much attention to the list and didn't report it before now.
I presume the list server is doing something wrong with the MIME format. But I'm not a mailman expert. Does anyone know why this is happening and, more importantly, how to fix it?
Mail client is Outlook. And, of the 50+ email lists I'm on (most of which are Yahoo groups, some of which are mailman), this is the ONLY one that does this.
Michael N6MEF
-----Original Message----- From: 44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu [mailto:44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu] On Behalf Of Eric Fort Sent: Saturday, June 29, 2013 1:33 AM To: AMPRNet working group Subject: Re: [44net] hardware vs. software
(Please trim inclusions from previous messages) _______________________________________________
BTW -- the modems in the UDR56k-4 are DSP to I/Q modulation/demodulation and run on the included Linux card. The intent is to have flexibility in modems, protocols, and applications and provide an open source environment for the experimenters in the user community.
The modems, and protocol stacks, will be available on socket interfaces, so protocols and applications may run either on the radio's embedded system or via interconnection to another host.
By having on-board processing we also are looking at very low tx/rx turnaround time, well below what could be done on a USB or serial port.
One application is using the device as an 'Ethernet Bridge' within a given protocol, modem, and data rate.
Oh, and the price is a 3rd less than Bill posted :)
------------------------------ John D. Hays K7VE PO Box 1223, Edmonds, WA 98020-1223 http://k7ve.org/blog http://twitter.com/#!/john_hays http://www.facebook.com/john.d.hays
On Sat, Jun 29, 2013 at 1:33 AM, Eric Fort eric.fort@gmail.com wrote:
(Please trim inclusions from previous messages) _______________________________________________
While I can't speak for the company making the UDR56k and I have stated previously where I think it missed the mark. putting a new RF modem on an ethernet connection is a good thing. It's something kantronics should have done when the 9612 was introduced to the world. The advantage of placing it in ethernet is that one simply assigns it an address, plugs it into a (poe) hub, and it works with little to no config on the host side. many if not most who might be interested in this probably would rather not spend a huge amount of time futzing with the operating system of whatever device takes bits to or from RF. personally, I'd like to see routing functions completely divorced and separated from host services functions. if you wish to offer application layer services to amprnet then fine, put them up on a host of their own. that host should not generally also route. That just seems the be good practice. The reality is that these ethernet to RF boxen will likely intigrate into your subnet just fine and be more reliable than either serial or usb (never mind I fail to find an actual hardware serial port on ANY PC built in about the last 5 yrs....) The reality is that these don't need much in the way of computer onboard and likely never will. I don't see their intent as being something that is intended to host services and the computer needed to move bits from ethernet to 56k wireless modem should not add much cost, I'm guessing 15-20usd/unit while adding much utility.
Eric AF6EP
On Fri, Jun 28, 2013 at 11:32 PM, Michael E. Fox - N6MEF n6mef@mefox.orgwrote:
(Please trim inclusions from previous messages) _______________________________________________ I agree with you about not having the perfect locations for microwave frequencies. We have some nice sites on mountains and some not-so-nice sites with trees. So I was originally excited to read about the UDR56K.
*BUT* I, for one, am really disappointed that the UDR56K is also a linux machine. I think they screwed the pooch on that (although I understand about 300 pre-orders would disagree with me). Why do I say this? Because I already have a linux machine running JNOS (and a bunch of other stuff). I was excited about the prospect of switching our current backbone from 1200 to 56K. I was hoping to just plug it into another serial port on my existing linux machine. But it has no serial ports for my other TNCs/radios. (and adding multiple usb-serial adapters and a powered usb hub is just trouble waiting to happen). And I'm betting it doesn't have the horsepower that my existing linux system does, which runs an amprnet gateway, mail gateway, lots of iptables rules, spam and virus checking, etc., plus some automated network testing stuff. Whatever processor it does have, will be aging along the processor aging curve (speed doubles every 18 months) and to upgrade, you would throw out the radio, too - a bit of the baby and bathwater approach. In fact, forget about upgrades. If anything fails -- radio, modem, linux system memory, linux system storage -- I've got to replace the whole thing. All things fail at some point. So this just seems strange to me.
I wish they had just made the radio/modem and provided a serial port so I could connect it to whatever machine I want. I would have bought one for each of our machines, plus a spare (7). But I don't know what to do with what they built - namely a fast modem and radio hard-wired to a very limited, non-expandable linux platform. But again, I guess I'm in the minority because I understand they have about 300 pre-orders.
Oh well. Michael N6MEF
John,
I'm glad you're on this list. Didn't realize that.
I've look and found no "solution" oriented documentation -- i.e. how can I use the device? All of the information about the internals, like "socket interfaces" may have meaning for some. But, since I don't write protocol internals, it doesn't help me understand the deployment scenarios.
For example, assume I have four JNOS systems that are currently connected to each other on a single subnet using a single 440 frequency. They talk to each other using IP over AX.25. I would definitely like to increase the speed. But it's not clear to me how I would deploy the UDR56K-4 to replace the existing 440 radios/TNCs. What protocol would it run, at what speed? What would the IP network diagram look like?
If this question is not appropriate for the 44-list, we can move it to your yahoo group.
Thanks much, Michael N6MEF
-----Original Message----- From: 44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu [mailto:44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu] On Behalf Of K7VE - John Sent: Friday, July 05, 2013 11:02 PM To: AMPRNet working group Subject: Re: [44net] hardware vs. software
(Please trim inclusions from previous messages) _______________________________________________ BTW -- the modems in the UDR56k-4 are DSP to I/Q modulation/demodulation and run on the included Linux card. The intent is to have flexibility in modems, protocols, and applications and provide an open source environment for the experimenters in the user community.
The modems, and protocol stacks, will be available on socket interfaces, so protocols and applications may run either on the radio's embedded system or via interconnection to another host.
By having on-board processing we also are looking at very low tx/rx turnaround time, well below what could be done on a USB or serial port.
One application is using the device as an 'Ethernet Bridge' within a given protocol, modem, and data rate.
Oh, and the price is a 3rd less than Bill posted :)
MIchael,
I don't have time right now to test, but JNOS2 compiles from source on the UDR56k-4 as does jnosinstaller. The installer configures the program just fine.
Barring any unforeseen issues, you should be able to run JNOS directly on the radio. If you have TNCs servicing local LANs (e.g. 2 meters, 220, ...), you can put USB-to-Serial interfaces on the UDR56k-4 and attach the TNCs with their current radios. No other computer would be required. We have tested TNCs attached in this manner on other applications for over a year (daily). The four UDR56k-4s would form your backbone radios. AX.25 drivers are already in place to drive the UDR56k-4 at any supported speed including 9600-baud to over 56k baud (with some steps in between). You would use a CIDR of /29 if these were the only radios on your backbone LAN.
This is an open architure/system so bring your favorite applications (Linux source or Linux ARMEL binaries) to the radio. If your application is more appropriate to run on another computer, use the UDR56k-4 as a relay device, using an IP interface (wired or wireless).
Further discussion on the UniversalDigitalRadio forum on Yahoo! Groups.
------------------------------ John D. Hays K7VE PO Box 1223, Edmonds, WA 98020-1223 http://k7ve.org/blog http://twitter.com/#!/john_hays http://www.facebook.com/john.d.hays
On Sat, Jul 6, 2013 at 9:15 AM, Michael E. Fox - N6MEF n6mef@mefox.orgwrote:
(Please trim inclusions from previous messages) _______________________________________________ John,
I'm glad you're on this list. Didn't realize that.
I've look and found no "solution" oriented documentation -- i.e. how can I use the device? All of the information about the internals, like "socket interfaces" may have meaning for some. But, since I don't write protocol internals, it doesn't help me understand the deployment scenarios.
For example, assume I have four JNOS systems that are currently connected to each other on a single subnet using a single 440 frequency. They talk to each other using IP over AX.25. I would definitely like to increase the speed. But it's not clear to me how I would deploy the UDR56K-4 to replace the existing 440 radios/TNCs. What protocol would it run, at what speed? What would the IP network diagram look like?
If this question is not appropriate for the 44-list, we can move it to your yahoo group.
Thanks much, Michael N6MEF
-----Original Message----- From: 44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu [mailto:44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu] On Behalf Of K7VE - John Sent: Friday, July 05, 2013 11:02 PM To: AMPRNet working group Subject: Re: [44net] hardware vs. software
(Please trim inclusions from previous messages) _______________________________________________ BTW -- the modems in the UDR56k-4 are DSP to I/Q modulation/demodulation and run on the included Linux card. The intent is to have flexibility in modems, protocols, and applications and provide an open source environment for the experimenters in the user community.
The modems, and protocol stacks, will be available on socket interfaces, so protocols and applications may run either on the radio's embedded system or via interconnection to another host.
By having on-board processing we also are looking at very low tx/rx turnaround time, well below what could be done on a USB or serial port.
One application is using the device as an 'Ethernet Bridge' within a given protocol, modem, and data rate.
Oh, and the price is a 3rd less than Bill posted :)
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net http://www.ampr.org/donate.html
Thanks John,
But that wasn't my question. If you re-read my question, you'll see that the JNOS machines, network, radios, TNCs, etc. already exist.
The question was how to deploy the UDR56K-4 in a 56K bridge configuration on a shared subnet to replace the existing 440 radios and TNCs. For example, some other technologies, like Icom's ID-1, only operate in a point-to-point configuration (as far as I know). That's why I asked about the shared subnet. Also, merely speeding up AX.25 to 56kbps isn't going to work unless forward error correction is added. Hence, part of my question was about what protocol would be used.
Everything I've read so far, including your answer below, indicates to me that the UDR56K-4 is really an experimenter's platform, and the end solution is left to the user to figure out. In other words, you're providing a linux hardware platform with an integrated 440 radio. That's cool. But if the solution I need is a 56K bridge, it sounds like it's up to me to find a protocol with FEC that is allowed by the FCC, then find the source code, compile it, test it, then somehow connect that to an IP routing or bridging configuration in linux. Am I interpreting the situation correctly?
Thanks, Michael N6MEF
-----Original Message----- From: 44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu [mailto:44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu] On Behalf MIchael,
I don't have time right now to test, but JNOS2 compiles from source on the UDR56k-4 as does jnosinstaller. The installer configures the program just fine.
Barring any unforeseen issues, you should be able to run JNOS directly on the radio. If you have TNCs servicing local LANs (e.g. 2 meters, 220, ...), you can put USB-to-Serial interfaces on the UDR56k-4 and attach the TNCs with their current radios. No other computer would be required. We have tested TNCs attached in this manner on other applications for over a year (daily). The four UDR56k-4s would form your backbone radios. AX.25 drivers are already in place to drive the UDR56k-4 at any supported speed including 9600-baud to over 56k baud (with some steps in between). You would use a CIDR of /29 if these were the only radios on your backbone LAN.
This is an open architure/system so bring your favorite applications (Linux source or Linux ARMEL binaries) to the radio. If your application is more appropriate to run on another computer, use the UDR56k-4 as a relay device, using an IP interface (wired or wireless).
Further discussion on the UniversalDigitalRadio forum on Yahoo! Groups.
Follow up posted at http://groups.yahoo.com/group/UniversalDigitalRadio/message/443
------------------------------ John D. Hays K7VE PO Box 1223, Edmonds, WA 98020-1223 http://k7ve.org/blog http://twitter.com/#!/john_hays http://www.facebook.com/john.d.hays
On Sun, Jul 7, 2013 at 10:20 PM, Michael E. Fox - N6MEF n6mef@mefox.orgwrote:
(Please trim inclusions from previous messages) _______________________________________________ Thanks John,
But that wasn't my question. If you re-read my question, you'll see that the JNOS machines, network, radios, TNCs, etc. already exist.
The question was how to deploy the UDR56K-4 in a 56K bridge configuration on a shared subnet to replace the existing 440 radios and TNCs. For example, some other technologies, like Icom's ID-1, only operate in a point-to-point configuration (as far as I know). That's why I asked about the shared subnet. Also, merely speeding up AX.25 to 56kbps isn't going to work unless forward error correction is added. Hence, part of my question was about what protocol would be used.
Everything I've read so far, including your answer below, indicates to me that the UDR56K-4 is really an experimenter's platform, and the end solution is left to the user to figure out. In other words, you're providing a linux hardware platform with an integrated 440 radio. That's cool. But if the solution I need is a 56K bridge, it sounds like it's up to me to find a protocol with FEC that is allowed by the FCC, then find the source code, compile it, test it, then somehow connect that to an IP routing or bridging configuration in linux. Am I interpreting the situation correctly?
Thanks, Michael N6MEF
BTW, we'll provide application notes as we get closer to and after delivery. Our first priority will be applications we are directly supporting. We invite users to provide application notes as they build additional solutions.
------------------------------ John D. Hays K7VE PO Box 1223, Edmonds, WA 98020-1223 http://k7ve.org/blog http://twitter.com/#!/john_hays http://www.facebook.com/john.d.hays
On Sat, Jul 6, 2013 at 9:15 AM, Michael E. Fox - N6MEF n6mef@mefox.orgwrote:
(Please trim inclusions from previous messages) _______________________________________________ John,
I'm glad you're on this list. Didn't realize that.
I've look and found no "solution" oriented documentation -- i.e. how can I use the device? All of the information about the internals, like "socket interfaces" may have meaning for some. But, since I don't write protocol internals, it doesn't help me understand the deployment scenarios.
For example, assume I have four JNOS systems that are currently connected to each other on a single subnet using a single 440 frequency. They talk to each other using IP over AX.25. I would definitely like to increase the speed. But it's not clear to me how I would deploy the UDR56K-4 to replace the existing 440 radios/TNCs. What protocol would it run, at what speed? What would the IP network diagram look like?
If this question is not appropriate for the 44-list, we can move it to your yahoo group.
Thanks much, Michael N6MEF
-----Original Message----- From: 44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu [mailto:44net-bounces+n6mef=mefox.org@hamradio.ucsd.edu] On Behalf Of K7VE - John Sent: Friday, July 05, 2013 11:02 PM To: AMPRNet working group Subject: Re: [44net] hardware vs. software
(Please trim inclusions from previous messages) _______________________________________________ BTW -- the modems in the UDR56k-4 are DSP to I/Q modulation/demodulation and run on the included Linux card. The intent is to have flexibility in modems, protocols, and applications and provide an open source environment for the experimenters in the user community.
The modems, and protocol stacks, will be available on socket interfaces, so protocols and applications may run either on the radio's embedded system or via interconnection to another host.
By having on-board processing we also are looking at very low tx/rx turnaround time, well below what could be done on a USB or serial port.
One application is using the device as an 'Ethernet Bridge' within a given protocol, modem, and data rate.
Oh, and the price is a 3rd less than Bill posted :)
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net http://www.ampr.org/donate.html
Just some input... on USB Adapters. As far as USB to serial Adapters.. I have two Edgeport 8's on mine* each one uses ONE USB port and gives you 8 serial ports. I have two thus 16 ports. Nice.. Ebay them you can find them for $40 at times.. New they are $400+ They do not suffer from swapping ports when you reboot Linux.. To accommodate my Rig control lines and other USB Devices I have 2 4 Port PCI USB3 cards.. All works well ... I do have three Ports (USB) that swap when I reboot the two rig control lines and my SCS Tracker, which have nothing to do with the Edgeports... Otherwise all is well...
Just found these to be superior devices to the multiple USB Adapters.. that usually stink..
73 Jerry N9LYA
* N9LYA/K9BBS http://w9bbs.no-ip.org:8080 Linbpq 6.0.2.12 Debian Linux 7.0.0 Dell_Server_2600_4GBRAM_73GBHD_Dual_Dual_Core_CPU_1.8Ghz
W9HU http://w9bbs.no-ip.org:8081 PilinBPQ 6.0.2.12 Raspbian-wheezy-7_Rpi_B_700MHz_256MBram_32GB-SD-class10
N9LYA-5 JNOS 2.0j Debian Linux 6.0.6 HP 3.4Ghz 2GB Ram 2TB Hard drive
http://www.mitchellwx.com http://www.n9lya.com http://www.kutche.net http://www.hfskipnet.net
On Fri, Jun 28, 2013 at 10:19 PM, Bill Vodall wa7nwp@gmail.com wrote:
(Please trim inclusions from previous messages) _______________________________________________
Based upon what the bell system guys were doing 50 years ago, it seems that 45Mbps
over
a 70-100km radio link appears it ought to be doable.... but really,
10Mbps
over 50km would be a good start.
We don't all have mountaintop microwave sites....
Not everyone needs a microwave mountaintop site or even a decently tall tower. bear in mind the items that would likely be hosted in sites such as these would be more of a community resource such as many voice repeaters are (and could be collocated with a club that had decent resources) The resources and hardware at such sites would be for the most part pieces of the network core, not the outlying networks that connected to it and used it's backbone..
We have homes and ham stations in compromise locations with city lots, restricted antennas and big RF absorbing trees... 25W of 440 MHz will probably get through where less power at higher frequencies won't.
some will be somewhat limited, this is true but when it comes to effective radiated power and RX gain I can easily outdo your 25W of 70cm at the antenna feedpoint with a watt at 5760 or 10250 measured at the feedpoint of an antenna of similar physical size. just because some may not live on top of a 10,000 ft mountain peak with multiple 1000ft self supporting tower structures does not mean they can't make good use of amateur microwave frequencies. such antennas get smaller and easier to put up higher as the frequency goes up. all that's needed is that you can link to the next node from you or into the backbone network. some may be so limited as to only be able to do so by internet tunneling. but that doesn't mean that those who are not so limited should not build the network. Amateur microwave at 2.4GHZ and above is the proper place to do so. It gives the performance we need with hardware most will find affordable and easy to use and implement. If you're living on a city lot, my guess is that your next hop neighbor ought to be much closer than if you're living 40miles out of town. also in more densely populated areas you will need greater network capacity and bandwidth to properly serve each user an equal experience. 20 people contending for use of a 56k link is much different than 1 user having it to themselves.
The UDR56K is a good compromise real world solution that is probably the last best hope for AMPR data... The speed is good enough for IM, email like this list, megabyte files (I know first hand that's possible...) and facebook clones. That's way more than what we have now...
on a per user basis, yes 56k may serve as a minimum bar but try sharing that simultaneously with 20 other users. get 20 monkeys together in an attempt to recreate the works of shakespere and you'll have that line saturated for an infinite amount of time (ref rfc2795) as even they can likely be trained to type faster than that. I know I can certainly read what one of them could type at that speed. That said why shoot for a minimum even if it is 5-6x faster than what we have stagnated upon for 20+ yrs. People didin't jump at 56k then, why should they now?
The price per node (about $600) is high and will be hard for many to justify but it beats the competitors - ID1 is $1+K for 120K raw data with 10W at 1.2 G or D710 with 50W of 9k6 data UHF or VHF for $700.
yes, Dstar has effectively priced itself out of the market as to costly, to little, to late while being an orphan that no one except amateur radio uses. compare that however to the cost of a 1M dish and a professional grade 802.11a 30dbm radio that goes from ethernet to RF. the 1M dish can be had for 89usd and the radio for about the same.... so for under 200usd you have a setup that can link that even in a somewhat rural setting ought get you to your nearest ham neighbor. Compare this, faster and cheaper to the UDR56k. I think people would go for the faster, easy to set up 200USD solution.
Let's raise the bar a bit and get to building a RADIO BASED NETWORK!
A harder problem than the RF is the WHY.... I can download Zombie videos at 40MB (or ? - I haven't checked lately) from anywhere in the world - why should I futz around with silly HAM regulations and bureaucracy to do limited Ampr data... (FWIW I know the answers to that...)
Why is actually easier to answer than it might seem. We'll start with 47CFR97.1 with which you and everyone on this list and with access to AMPRNET ought be familiar:
§ 97.1 Basis and purpose.
The rules and regulations in this part are designed to provide an amateur radio service having a fundamental purpose as expressed in the following principles:
(a) Recognition and enhancement of the value of the amateur service to the public as a voluntary noncommercial communication service, particularly with respect to providing emergency communications.
Building such infrastructure enhances our ability to serve the public especially in providing emergency communications in time of need. one of the key uses for such infrastructure is tha ability to show up somewhere and light up communication facilities between disaster sites that are capable of moving voice and computerized data rapidly and efficently which is also usable by those with little or no tech training. i.e. they attach their laptop to the network at the disaster site and are online with all other disaster sites they need contact with and have simultanious voice and data connectivity between sites. By building such a network we augment our ability to justify our existance.
(b) Continuation and extension of the amateur's proven ability to contribute to the advancement of the radio art.
by building such a network we work to advance the radio art thus satisfying another point of why we are allowed to exist.
(c) Encouragement and improvement of the amateur service through rules which provide for advancing skills in both the communication and technical phases of the art.
by building and operating such a network we advance our skills in communication and technical phases of the art such as RF linking and TCP/IP network operation and optimization. We serve as a training ground and build our own skills along with the skills of others in the area of data communications.
(d) Expansion of the existing reservoir within the amateur radio service of trained operators, technicians, and electronics experts.
by building such network we expand the existing reservoir within the amateur radio service of trained operators, technicians, and electronics experts not only training those in technical arts they can take elsewhere but also bringing greater numbers into the hobby for which this is their primary interest in communications (not everyone is into 30second weak signal contacts or chasing DX on 20M).
(e) Continuation and extension of the amateur's unique ability to enhance international goodwill. By building such a network on an international basis we create yet another means to extend the amateur's unique ability to enhance international goodwill.
those right there ought to be reason enough to get building... but I'll add a few more:
Preservation of the precious frequency allocations we have available for our use - anyone remember when we once had 220-222Mhz available to our use and how due to perceived non-use it was reassigned for land mobile use resulting in a 40% loss of the spectrum in the 1.25M band? We have microwave bands that go largely unused and other services that would love to run us out of them. By building such a network we put these bands to effective good use
again, It serves as an avenue to bring more people into the amateur radio fraternity.
it allows us a platform upon which creative applications can be built. imaging having a minimum of 24 voice channels between each and every mountaintop or backbone node as well as reasonably high data bandwidth. that kind of trunking could be used for repeater linking or even allowing multiple conversations over a wide area to be carried on simultaniously. once voice is bits, it's just data like anything else on the network. voice is only a start, the rest of the applications that may be built are limited only by the imagination of the amateur radio AMPRNET community and their willingness to advance the technical arts.
Eric AF6EP
Eric
AF6EP
73 Bill, WA7NWP _________________________________________ 44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net http://www.ampr.org/donate.html
-
Bill Vodall -
Brian Rogers -
Eric Fort -
Jerry -
Jesse Hindmarsh -
K7VE - John -
Michael E. Fox - N6MEF