Pete said:
From memory is USB not limited to 3m generally? this is using a
standard cable and not twisted pair so if it was limited to 3m due to
noise interference then you may be onto an idea, though I'm not sure how
much further you would get.
You can however get HDMI to CAT5e/6 converters so if you could find a
camera with an HDMI/mini-HDMI output that could be an option.
Can't really speak for software engineering but with Mech Eng we
generally try to accommodate legacy parts where-ever possible. I don't
get to play much with final software builds much as by the time these
are finished we are generally getting ready to ship!
You could take a step back in tech terms and get something non-IP and
run the coax cable from the camera to a video capture card or IP encoder
in your house. Can't see the PCIe card costing much, the IP encoder
however could be quite costly.
I briefly looked into a few cheap IP cams last year and I got the
impression there were plenty on eBay that served a webpage and could
even push data out onto network storage... it's been so long you have me
doubting myself.
http://www.usb.org/developers/usbfaq/#cab1
"Cables and Long-Haul Solutions
The cable length was limited by a cable delay spec of 26ns to allow
for reflections to settle at the transmitter before the next bit
was sent."
At 5.2ns/meter propagation delay allocation for the cable,
that is a 5 meter limit (16 feet). (You can get more details from
the USB_20.pdf document, which is 650 pages long.)
"The best solution is ... a ... cable that had a one-port bus powered
hub in the middle."
That would be a thing like this. Some cables are available which include
more than one of those repeaters in the cable. There is a limit to how
many hubs can be strung in a row. Power quality at the end of the
chain is likely to be poor, so a wall wart powered hub near the destination
is required for power-hungry devices. A mouse might run off the end of the
cable, without additional power. (Another note - when buying these,
*always* read the reviews, and make sure that some user has verified
that a long chain of cables works. Some of these repeater products,
it turns out they don't successfully concatenate. Only buy ones that
chain nicely, and deny the loser designs a place in the market! It's
unclear why some of these don't daisy-chain properly.)
http://www.newegg.com/Product/Product.aspx?Item=N82E16812816021
So that's an example of a cabling solution, that only uses USB protocol.
The USBFaq also states:
"I really need to put a USB device more than 30 meters away
from my PC. What should I do?
Build a USB bridge that acts as a USB device on one side and
has a USB host controller at the other end. Use a long-haul
signaling protocol like Ethernet or RS-485 in the middle. Using
cables or short-haul fiber, you can get ranges upwards of a
kilometer, though there's no reason why the long-haul link
in the middle of the bridge couldn't be a pair of radio
transceivers or satellite modems. Embedded host solutions
capable of doing this already exist."
In other words, use a bridging protocol different than USB, for
the traversal of the longer-than-five-meter cable. Not USB
protocol itself. If you do find a USB bridge product, they're
unlikely to go into any technical details about how they do it.
When I show a product from this company, it's to show technical
capability. They hardly ever are price competitive. But if you
have pockets stuffed with cash, this'll solve your problem.
http://www.gefen.com/kvm/dproduct.jsp?prod_id=5529
Note also, that naive implementations, they can claim the
product is USB2, but when you check the specs, the specs
only mention 12 megabits/sec. USB2 is 480 megabits/sec, so
I don't understand the labeling on this one, at all. If
you see some cheap Ebay product, they could pull the same
trick. I've seen this in the past, for some Laplink cables,
where some idiot claims they are selling a USB2 product,
only to discover the actual chip in the product is one
of the inferior USB 1.1 chips.
http://www.gefen.com/kvm/dproduct.jsp?prod_id=2178
Paul