I have recently bourght a narrowboat. The marina has a wifi connection which I can get fine outside the boat. The boat is basically a long squarish steel tube, with windows as an afterthought.
I can get a so so reception inside if I contort myself at a window, but it is not ideal.
What I would like to do is put an external aerial out, and then have some sort of way of putting that signal inside so I can roam within the boat. There are no obstacles inside for the signal.
The PC I currently use is a laptop with IntedlPROset/wireless and would like to retain this.
What are the options? In my mind I see an external aerial, linking with coax to a wireless router, but I am sure it can't be that simple.
Simplest is to have a USB wireless dongle on a longish lead, you can stick it outside the boat every time you need to use the 'net.
If you really want to do something like this you either need a unit than can be in both client and AP mode simultaneously, or you need two units, one in AP mode to service your laptop and one in client mode to hook up to the boatyard WLAN. If you go this route, any AP with client mode and detachable antennae will be fine, you can buy a matched aerial from the same manufacturer in most cases.
Mark McIntyre suggested a USB radio which is probably a good way to go, especially if the marina has a strong wireless signal.
However, if you want to use the Intel ProSomething MiniPCI card in your laptop, you can add a pigtail and external connector. It might require drilling a hole or snaking the cable through an existing hole. I found that I modem connector hole is sometimes plugged and can easily handle a coax cable and connector.
You'll need a u-FL to SMA, TNC or N connector pigtail. See:
formatting link
formatting link
to the cable or connetor would be an external antenna of some sorts. I can't recommend anything specific without knowing something about the signal strength, connector location on the vessel, and approximate range required.
I don't think you're going to do much roaming around the boat with either USB or a pigtail antenna connector. You're tethered by the necessity of having an external connector, which is limited in both length and ruggedness.
It may be possible to setup a passive or active repeater to get the signal in and out of the vessel. I'm not a big fan of either, but they do mostly work. Passive is always hurting for more signal. Active store and forward repeaters (i.e. range extenders) have compatibility and thruput problems.
Because I'm currently a pariah here, I refrained from jumping in at the beginning.
This is exactly, without the mast, my project. Presumably, all the ways I've been vociferously assured will work on my boat will work on his. He just has a tougher hull to get 2.4G through, metals being more challenging than fiberglass.
So, how come nobody's told him? Given that I manifestly have failed in all of my attempts (I'm taking another run at it some time soon, making another attempt at divide and conquer, simplify, simplify, etc.) thus far, I'm obviously not the one to do it.
I do essentially this. I've been told it won't work, though :-)
I have a Linksys WRT54G router (which has dual antennas). One of the standard antennas has been removed and connected to a coax cable to an external antenna pointing at the house from which I get my Internet connection (you'd probably just want an omni antenna). The router is configured as a repeater (which would require, iirc, cooperation from the marina people, as it needs to be configured at their end, too) The remaining original antenna broadcasts to my home wireless clients.
Easier, and probably just as cheap, is to use a client-mode device (many routers can be configured as clients) attached to the external antenna, hardwired to an AP which will broadcast to the inside of the boat (on a non-overlapping channel).
Perhaps I misunderstood your posts then, but I didn't see a requirement to simply connect to your marina's wifi from a boat, but to be able to connect to _any_ available signal. Connecting to a fixed signal for which you're an expected user is pretty trivial unless you have distance issues.
I don't think you are a pariah, Skip, it's just that you were thrashing so hard that I gave up trying to follow your constantly morphing problem description, proposed solutions, and optimizations thereof.
Nobody's told him what? That there are a bunch of different options? That there are a half-dozen scattered threads on your subject?
On the contrary, you are our local expert on this subject, please jump in and tell the OP what things you have tried and how well they worked. Point him at the other threads. Pass along the helpful advice you've gotten in this newsgroup.
There are going to be serious multipath issues, so I'm voting for a pair of WAP54G devices, one in client mode, one in AP mode on a different non-overlapping channel.
You're not a pariah or whatever. My problem is that I don't think I can solve your problems with your Senao 2611 radios. Your scheme should have worked. If you have other questions (that don't involve the Senao 2611 radios), I'll me more than happy to take a swing at them (time permitting).
No, your project was to get a single Senao 2611 to act as a proper client adapter to your marina shared wi-fi. This is different.
Full Disclosure: I hate store and forward repeaters, mesh networks, and range extenders. I consider them RF polluters because they generate many duplicate packets in order to move just one packet.
Assuming that moving around inside the vessel hull is of paramount importance[1], this is a tolerable application for a modified "range extender". The problem is how to get adequate RF coverage both below and above decks. The obvious way is to use two radios. A client radio (WRT54G with DD-WRT v23 SP1 in client mode) on the outside. CAT5 and power through the hull. Any AP (access point) on the inside. This is the way I used to play repeater before the "range extender" type repeaters became available. One nice thing about it is that the two radios can run on different channels thus avoiding mutual interference and avoiding the inherent 50% slowdown.
However, 2 radios seems a bit like overkill. It should be possible with one repeater[2] and two antennas. I have such a system at a friends house. On the roof is an 8dBi omni antenna that is the main site for a neighborhood LAN. 25ft of LMR-400 coax goes down into the house where it connects to a 2 port splitter (power divider) as in:
formatting link
made my own, but these look nicer. The other port of the splitter goes to an 8dBi panel antenna for in house coverage. The radio is a Netgear something access point. The problem that this solved was that with just the rooftop antenna, wireless coverage inside the house was terrible. My friend didn't want to deal with two access points, or package a radio for outdoor use. So, half the power goes to the outside antenna, and half to the inside antenna.
Ignoring the coax cable losses, the transmit loss for each port is
3.5dB which will yield a 33% loss of range. Not good, but possibly useable if the marina signal is strong. However, that's only in transmit. In receive, the loss per port is only 0.5dB for a 6% loss range. This is one place where an overly powerful xmitter (100mw or more) is proper.
Note that this splitter system only works if the two antennas cannot see each other. Were the antennas too close to each other, the antenna patterns will interact resulting in odd nulls and peaks, multipath effects, frequency selective nulls, and general weirdness. Because the metal hull effectively isolates the two antennas, this should be an ideal application.
The selection of the repeater is going to be a problem. Because repeaters seem to be highly proprietary, manufactories test them usually only with their own products. Compatibility issues are a major problems. Even if they are able to function, timing problems seem to cause problems with thruput and connection reliability. In addition, some of the new and wonderful enhancements (Super-G, Turbo-G, MIMO, etc) don't work with repeaters. All I can suggest is not to buy anything that can't be tested and returned.
Please note that I haven't tried using a splitter and repeater in combination on either land or water. Treat this as somewhat of an R&D project. My suggestion would be to install a wireless client radio of some sorts inside the hull. Run the coax through the hull and install an outside antenna. Run CAT5 from the client radio to where the laptop normally sits below decks. Nail the laptop to the table to keep it from moving. Keep it simple.
The secret to successful engineering is to know when to give up.
[1] When I tried to type while underway, I got sea sick. Also, moving around below decks carrying a laptop seems like a great way of insuring that it will be dropped.
[2] For purposes of this discussion, the terms "repeater", "range extender", and "store and forward repeater" are the same. I'm just sloppy and tend to use the terms interchangeably.
The problem is the algorithm used by the diversity switch. It's not like a VHF/UHF police scanner, that constantly scans for a signal. It's a messy algorithm, that does it's best NOT to switch antennas unless the signal is gone. The diversity switch simply will not switch fast enough (or often enough) to play repeater, especially when it has to switch on literally every packet.
If you wanna watch it fail with just an ordinary access point, try setting up a test, where your access point or wireless router has two highly directional antennas, preferably in separate rooms. Use AP mode, not repeater mode. Place a wireless laptop in each room and try to transfer data between the laptops. If there's some leakage between rooms, reduce the AP xmit power, use attenuators, or longer coax cables. The idea is to not have the laptops or antennas hear each other.
I've done this test. It will sorta function but the thruput will be awful and very bursty. The expected throughput should be 50% of the connection rate if it were working as expected. I think I got about
2% to 5% of connection speed.
I recently found a nifty article on the different wireless diversity systems. Unfortunately, I can't find it. Maybe later.
OK Seems like the easiest way is going to be the USB stick on a cable, with some sort of waterproofing on it. I have tried this one at home with my laptop and indeed it does work. True I don't get the wireless part but I do get to sit on a seat rather tan outside or contorted at a window.
The problem I have with this though is getting my Lappy to ignore the Intel thing and using the USB Stick. The only way I have seen to do this is to disable the mini PCI in the bios. Not very convenient.
Is there another way of turning the internal one off in favor of the USB? I have tried the Device manager. It switches of the internal one ok, but I cant then make the USB kick in.
Again any suggestions the Lappy is a Dell 6000 Inspiron.
Yes, that's the easiest, but not necessarily the best. I'm assuming that the primary use will be in a crowded marina, with lots of other wireless clients and plenty of highly reflective masts and rigging. For that kind of application, a directional antenna does wonders for eliminating reflections and reducing interference from other users. There are directional antennas with integral USB radios but I prefer more gain than typically offered (about 8dBi typically).
Kicking the USB radio will void the warranty. Not recommended. Got any symptoms for the USB is not working? Duz the USB radio work when the miniPCI wireless card is disabled in the BIOS?
Sure. However, I don't like to guess. What make and model USB radio? What manner of minPCI card is Dell shipping in the Inspiron 6000 this week?
The diversity switch chip switches between the two antennas to a common receiver. In order for it to work, the access point would need to build a table that says "to get to this MAC address, use #1 antenna" and "to get to that MAC address, use #2 antenna". If you have a radio that actually does this, I would be interested in the model number.
What I did was attach a wire to the diversity (PIN diode) switch driver and connect it to my oscilloscope. I think I did it on a WRT54G v1.1 running some early version of DD-WRT v23. I tried to figure out the algorithm employed. There was no scanning. As near as I can guess, it preferred the left antenna (as looking at the back of the WRT54G) and would only switch to the right if the left antenna received garbage. I have a very difficult time getting the diversity switch to even try the other antenna.
However, I did a very sloppy job testing it on my desk. It really should be done in an RF shielded box, with external cables to the antennas. This is the kind of stuff that I find interesting, so I'll try it again shortly.
Also, there was an access point that I tried that did work well as a repeater through the two antennas. The problem is that I've played with so many of them, I don't recall which one. I'll dig through my notes (if possible).
I don't know what you're using for testing so I can't tell for sure if this resembles your setup. However, it most certainly does resemble the OP's setup. He has a shielded hull seperating the two antennas. He has two wireless connections, one inside and one outside. He is using the radio as a repeater with two antennas which can be setup using the WRT54G with DD-WRT in repeater mode. With the noteable exception of the marina wi-fi access point being replaced by another wireless computer, the setups are identical.
Note: I still hate repeaters, if for no other reason that they generate discussions like this.
The USB will work if the miniPCI is disabled. In the Lappy is Intel PROset/wirless. Can't see where it is though took a few covers off it must be deeper inside. I was looking to see if I could attach an external aerial as on post suggested.
cheapo 802.11 hardware I've seen seems to use "switched diversity". Note that there are several methods of measuring S/N ratio or signal quality necessary to decide which antenna to use.
which says: Diversity antennas are not designed to extend the coverage range of a radio cell, but to enhance the coverage of a cell. The enhanced coverage is an effort to overcome issues that arise from multipath distortion and signal nulls. Attempts to use the two antennas on an access point to cover two different radio cells can result in connectivity issues. and: Note: The two antenna ports on the access point are designed for spatial diversity, and the radio only checks the other antenna when it encounters errors.
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.