SparkFun Forums 

Where electronics enthusiasts find answers.

All things pertaining to wireless and RF links
By dsandber
#1961
How does RF-24G get along with 802.11x since they both use 2.4GHz? Does the RF-24G make a noticable difference in 802.11x performance of vise-versa?

Thanks!

-Dan
By pittuck
#1963
Good question,

i dont know cause i have not used em yet.

But if sparky dont know, i might be able to set up a PIC @ 3V and experiment!
User avatar
By sparky
#1975
They play together harmlessly - good question!

You gotta love Nordic Semi - they've got all sorts of great docs. I just skimmed this one: http://www.nordicsemi.no/files/Product/ ... RF24XX.pdf
It talks about how to implement a frequency hopping scheme - quite cool.

I have SMiRFs operating all the time roughly 5 feet from the wireless router (dual antenna cheapo from CompUSA). There are many safe guards from interferance:

It was awhile ago when I read how 802.11 and Bluetooth work, but they basically hope channels all the time. So they will be in and out of your channel very quickly. The SMiRF automatically re-transmits if the listening unit does not respond. It's highly unlikely that you'll get jammed for the amount of time to transmit the packet 30 times. 802.11 will be in and gone by the time you get four or five packets into the air.

The SMiRF is setup to listen for very specialized packets that do not look like 802.11 packets. What I mean is the SMiRF must get a correct CRC, a correct Address, and then a correct address in packet location 0. It's highly unlikely the SMiRF will get any of that from a random network packet. The SMiRF uses 16bit CRC - I believe 802.11 is much higher, hence, they really shouldn't have similar looking packets.

BUT - the SMiRF does pickup a random packet every couple minutes. If you watch the Status LED, it will blink every so often (minutes) in a noisy environment. That means it received a packet that passed the CRC and address check. But the packet did not pass the location 0 test so the SMiRF does not do anything with it - just blinks the LED.

What I really want is a guru to figure out how to tap into the 802.11 network with a SMiRF. Wireless device anyone?

-Nathan
By pittuck
#2016
Interestingly enough, the modules and the wifi use different frequencies depending on which continent / country you are in. (Due to some millitary and govement controlled parts of the spectrum)

802.11b/g wireless works from 2412 MHz ( 2.412 GHz ) to 2485 MHz ( 2.485 GHz ).

Bluetooth devices ( phones, PDAs, etc ) work from 2400 MHz ( 2.4 GHz ) to 2480 MHz ( 2.48 GHz ).

Also the RF24-G are in fact able to change frequency, its only a little, but considering the number of frequencies in the wifi range you are unlikely to get any problems.

If you think u are having trouble either find a metal box and test ur rf devices in there (gotta be big enough for u, a pc/laptop and ur project. Perhaps even a light).

Or turn it off / move far away from it.

If u think u might get intermittent problems then you might want to turn crc off and try using manchester encoding. Also increase the preamble a bit.

Manchester can put uses 2x the space of the normal code, but even if some of the message is incorrect it is still able to collect the correct data:

So in binary:
1 = 10
0 = 01

Then to send 11001100 you would actually send 10 10 01 01 10 10 01 01

on the RX u might get something like:

10 10 11 01 10 10 01 11

As interfearence and only turn a 0 into 1 and not vice versa (assuming the signal is not inverted) then we can tell that the real data looks like this:

10 10 xx 01 10 10 01 xx or 11x0110x

Then we can just send many packets of the same data very rapidly and end up with correct result remembering that we only need to find:

mm mm xx mm mm mm mm xx

so the mm's are incosiquential to getting a correct communication.

Although as you are sending data lots of times, a addressing system will need to be used so you can distinquish between packets (you may need 5 of say 15 packets to get the full message, but you may not detect some packets etc. and need a way to know one has stopped transmitting)

Also another idea would be to have a period which the RX can acknowolege [sp] recieving a packet, perhaps by a preamble followed by 8 bytes of 1's... Sparky what you think to this?
By pittuck
#2018
BTW - i got my RF24-G working ;) 16F88 at 3.1V and a max232 working at 5V.

Next level transition ;) oh and programming a interface in asm ;)