Transmit but not Receive

Hi,

First off let me state that I'm a newbie in RF hardware so my apologies if this is a dumb question. For an experiment I'm conducting I want to be able to transmit two data streams from two WiFi(b) routers at the same time. I realize that they have carrier sense avoidance so that if they are right next to each other they will not transmit at the same time. Is there a piece of hardware (filter, diode? or something?) that i can put between the output of the routers and their respective antennas such that this device will pass the transmit signal to the antenna but not pass any radiation received from the antenna to the routers? The idea then would be that the carrier sense would not come into play as no power would be received from the antennas. Any ideas or suggestions would be greatly appreciated. Thanks,

Kev

Reply to
kevin.pavin
Loading thread data ...

A circulator would do the trick.

formatting link
formatting link
Rather expensive, sometimes to be found on Ebay.

w.

Reply to
Helmut Wabnig

Thanks for the suggestion, a circulator/isolator is what i need. If anyone has any suggestions on a bargain vendor I'd appreciate it.

Also as a follow up does anyone have a sense as to what the default carrier sense power threshold is on WiFi routers in dBm? I've seen a few isolators so far that range from 20dB to 40dB isolation, my router puts out 20dBm so it would be good to know if one isolator will do the trick. Thanks again.

Kev

Reply to
kevin.pavin

Why would there be a significant problem in the first place.... these aren't CW devices, they are Packetized Spread Spectrum and two devices next to each other, even on the same channel, would not run into each other, due to different spreading code roots anyway..... Hmmmmm, non problem, from where I sit.....

Reply to
You

Ummm... realize that a circulator/isolator hooked between the antenna, and the antenna jack of a typical 802.11b router, is going to dump essentially _all_ of the incoming power from the antenna into its dummy load.

This will include any power generated by the other AP's transmitter and reflected back from the antenna.

It'll also include any power picked up by the antenna.

In short, the AP's receiver will "go deaf". It won't be overwhelmed by transmitted signal from the other radio, but neither will it "hear" the signals it's supposed to respond to.

That's fine if all you're doing is broadcasting/multicasting packets, and don't need to accept packets or acknowledgements in return. It won't work, however, with a typical 802.11b AP-and-client system, in which the client device has to transmit to the AP in order to associate with it.

I really don't know if an 802.11 AP is going to try to broadcast or multicast packets if it "thinks" that there aren't any client devices associated with it.

Reply to
Dave Platt

I'm running 802.11b @ 2.4Ghz. As I understand it at 11Mbps (which I'm using) the signal is "spread" using the CCK code, and each packet/user uses the same spreading function. So a transmission only goes forward if the channel is "empty", i.e. the received power at the transmitter is below some carrier sense threshold (dBm)......

Reply to
kevin.pavin

Yah let me clarify that, I have a hacked up router for running experiments, all I'm using it for is to transmit broadcast packets, so no receive information is necessary for my app....I just want to get rid of carrier sense so that I can have more than one co-located tx running at once...and as i see it the easiest way is a hardware fix.....

Reply to
kevin.pavin

Hi All,

Spread Spectrum is merely the 0 level of the protocol stack and has nothing to do with "going forward." That is a software issue at a much higher level that deals with packet collisions. The hardware may augment collision detection through carrier sense (which would only occur if both share the same code). In fact, thresholds may be well below the noise floor (dBm) and still work with notably lower throughput, but this is an even more remote issue, except you may be fighting poor isolation from even the best of isolators.

73's Richard Clark, KB7QHC
Reply to
Richard Clark

snipped-for-privacy@gmail.com hath wroth:

Through the same antenna or seperate antennas? If one antenna, use a Wilkinson combiner to isolate the xmitters so there's no intermod.

Sorta. 802.11 is CSMA/CA or collision avoidance, not collision detection. It will do it's best to prevent simultaneous transmissions, but it still can happen. The receiver will listen for data, but will only wait until the end of the random contention time period. Allegedly, it will wait 3 contention periods before transmitting. Unfortunately, I've found that it will transmit anyway after these 3 periods. This is to prevent broadband noise from creating the ultimate denial of service attack and shut down all wireless transmissions by transmitting continuous noise.

formatting link

Others have suggested a circulator or isolator. That will work. So will a hybrid ring.

It might be easier to simply disable the receiver section of your unspecified wireless device. Many Wi-Fi router use seperate RF power amplifiers in the TX path. That means there's an exposed T/R swich somewhere on the board. Find it, chop the RX input line, ground or terminate the RF input pin, and you're done. If you can identify the radio chip, you should be able to easily find the RX pin.

Reply to
Jeff Liebermann

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.