Go Down

Topic: Digi XBee versus Nordic RF (Read 3582 times) previous topic - next topic

KirAsh4

Does anyone have experience with both of these hardware/protocols?  Let's put the cost aside, I'm looking specifically at performance.  I have a project proposal on the table for which I have one year to design a final product.  The short of it is being able to control LED lights on stage, wireless with a computer that's probably going to be used by someone sitting in the audience.  I'm thinking simply commands here, not streams of data.

My main concern is the venue, and the audience itself, specifically the possibility of interference and distance.  I would need things to be reliable, in other words, very little chance of losing data in transmission.  I know there will always be a chance, but the more I can eliminate that, the better.  I suppose I can forgo wireless and use an Ethernet connection to talk to everything on stage, but let's try this first.

So, anyone have a comparison made?  Range primarily, followed by reliability.

Personally I would love to stick to the Nordic modules because I already know how they work, with XBees I'm starting from scratch.  But, I will entertain the notion at least.

gerg

#1
Feb 17, 2012, 12:29 am Last Edit: Feb 17, 2012, 12:31 am by gerg Reason: 1
I don't have experience with xbee but from what I've read of the technical details its claim to fame to two part. One, it allows for best effort, multi-hop delivery. As far as I know, NRF24L01(p) does not have this stack. Two, xbee is available on a variety of frequencies. NRF24L01(p) is limited to 2.4GHz spectrum. I believe (not 100% sure) that the NRF24L01(p) based modules do have an advantage, which is higher potential throughput. IIRC, the xbee is comparable to the NRF24L01P operating at 250kbps mode; whereby the NRF24L01P supports up to 2Mbps.

If the limitations of the the NRF24L01P is acceptable for your project, I'd simply go with it. There are a variety of modules available which are comparable to the xbee in range (given the same frequency - 2.4GHz).

There are some non-tangable advantages with the xbee which is its well known, widely available, well supported, and lots of people have experience with these modules. The RF24 is fairly new and young. Having said that, you'll find people have pretty good experience with them around here with both the Mirf and RF24 libraries. And if you need something more complex, the RF24Network (albeit with some limitations) is certainly good enough for many smaller networks.

Given you seem to imply you only require point to point, unless you require something other than 2.4GHz, the NRF24L01P would be my preference. Is sharing spectrum with WIFI and bluetooth and various other radios an issue for you? If so, the 2.4GHz radios might be right out, which forces your hand to use xbee on a different base frequency.

I will disclosed, while I don't intend to be biased, it is possible I am slightly biased. I honestly don't think I am. But in the spirit of full disclosure, I'm a significant contributor to Maniac Bug's RF24 library.
http://maniacalbits.blogspot.com

KirAsh4

Well, there in lies the problem.  The stage IS surrounded by WiFi antennas.  And it is extremely possible that people will have bluetooth ear pieces and/or their WiFi turned on on their phones while in the auditorium.  And I have to also deal with wireless mics being used.  The whole sound system uses wireless mics.  So it is definitely a concern.  I know I can get XBees in different frequencies, like the 900MHz range.  But, I'm mainly concerned about distance.  Can I have the computer sitting at the top of the auditorium sending signals to the stage, or should that person be sitting close to the stage. 

gerg

#3
Feb 17, 2012, 03:32 pm Last Edit: Feb 17, 2012, 06:28 pm by gerg Reason: 1
Its impossible to say what kind of interference issues you'll find in the 2.4 spectrum. There are nRF24 modules supposedly capable of 1km or so. They roughly have the same form factor as the xbees with the antenna; perhaps a little smaller. Many people start off with something like this but find the range is rather limited. Localized WIFI can definitely overpower their TX. For long range, and something which more like an xbee, many people do something like this plus one of these.

One thing about the 2.4GHz spectrum is that WIFI tends to clump to WIFI channels 3, 6, and 9. And while modern bluetooth devices frequency hop all over the spectrum, their TX power is generally in line with the low-end module above. Which all means, as intended, many, many devices can co-exist at the same time in the available spectrum. Its also worth mentioning these devices support international frequencies which are not necessarily used nor legal in the US.

But long story short, if you are overly concerned about using 2.4GHz, it seems to me xbee is your natural candidate.
http://maniacalbits.blogspot.com

KirAsh4

Yeah, I'm thinking I may just have to build up a test unit and go sit in there and test it from various locations within the auditorium.  I mean, if the computer has to live near the stage, then so be it.  I would rather it doesn't, but one can't have *everything* all the time, right? :)

Anyway, thanks for the suggestions!

gerg

Don't forget that the RF24 has a single bit RSSI so its possible to build a so-so scanner in the 2.4Ghz spectrum. With it, you can scan what's in the auditorium. That, of course, doesn't account for what people bring with them. But, as for bluetooth, after pondering it a second or two, the good news is, most bluetooth should hop away from the nRF24's frequency, seeing it as interference. Of course, should and does are two different things. ;)

At any rate, check out the RF24 examples. In there you should find an arduino scanner application. With it, some slight modification, and an LCD, you should have a portable scanning which allows you to walk around and see what's there in the spectrum.
http://maniacalbits.blogspot.com

gerg

Found it.

Check out the above thread. From that thread I'm quoting the following. I don't recall what was going on at the time when I did that capture but you can see what's possible.

Code: [Select]

055 (2455Mhz): 05 - *****
056 (2456Mhz): 05 - *****
057 (2457Mhz): 04 - ****
058 (2458Mhz): 05 - *****
059 (2459Mhz): 06 - ******
060 (2460Mhz): 08 - ********
061 (2461Mhz): 08 - ********
062 (2462Mhz): 07 - *******
063 (2463Mhz): 05 - *****
064 (2464Mhz): 03 - ***
065 (2465Mhz): 01 - *
066 (2466Mhz): 03 - ***
067 (2467Mhz): 04 - ****
068 (2468Mhz): 05 - *****
069 (2469Mhz): 03 - ***
070 (2470Mhz): 03 - ***


Also in that thread I share test results on RF bleed per operating mode. Others chime in on legal frequencies in the US as well as which legal frequencies in the 2.4GHz spectrum are likely to be open.

Hope this helps.
http://maniacalbits.blogspot.com

KirAsh4

The thing is, while I can scan when it's empty, it will only pick up the WiFi signal.  However, on a performance night, it will be filled with everyone and their grandma, each with a cellular device of sorts.  Some which will connect to the local WiFi and others which will be actively seeking a signal, and those people who will bring in a bluetooth headset.  And while we ask people to shut things off, that never happens.

So, what I see on an empty room will be completely pointless compared to a full performance when it's filled and the sound system is turned on.  And I can't walk around and scan at that time. :)

gerg

Well, yes and no. A scan of an empty room still tells you something, but it doesn't paint the whole picture.

By scanning, you'll get an idea of what other signals are in the area. So for example, if there are WIFI AP's which are accessible from the room, and they are in use when you scan, you will know which channels they are on. Likewise, most people are not going to events to setup local wifi spots; though that certainly can happen. The point being, if you scan and see the usable spectrum is already pretty well saturated, communication is going to be difficult. If on the other hand, you go and find, for example, channels 3 and 6 are in use, that still leaves a lot of spectrum for use by you. In 250kbps mode, it takes but a fraction of what a WIFI channel uses.

As such, if after a scan, you only see a couple of WIFIs, which may even be on the same channel, use of the 2.4 spectrum, regardless of which you pick (xbee or RF24), seems very viable.

Modern bluetooth devices actively detect interference and seek interference free channels. In this case, again regardless of which device you use, and modern and properly behaving device should see your signal as interference and hop away to a frequency with less interference.

In a nut shell, a scan will tell you basic viability of the 2.4 spectrum for your project.

Beyond that, RF inside can be very tricky, regardless of the base frequency. Add to that many chairs (reflects) and people (reflects and absorbs) RF, the end result can be surprising. Having known professional roadies, I can tell you even they find surprises at concerts. And so to plan for it, because the show must go on, they bring a variety of equipment and frequencies to account for whatever they find.

So no, I can't tell you with 100% certainty that if you scan today and find things look good it will work come show time. But that's true regardless of whichever unit you use. But to have the best chance for success, unobstructed life of sight between the TX and RX is always best.
http://maniacalbits.blogspot.com

Jack Christensen

Sounds like pretty much a line-of-sight situation? About how far? I may be overly optimistic and I don't know what I don't know, but I'd be surprised if there was a lot of trouble making it work.

Given a year lead time, I might pick one technology or the other for starters and test them during actual performances. For the XBees, all that would be needed on the stage end is a powered up XBee placed in a likely location, and hook the other to a laptop running Digi's X-CTU program; the range test feature should be quite useful, as it tracks signal strength and errors. That would be a very minimal and low-effort - yet very good - real-world test. I'm not familiar with the Nordic technology, but perhaps there is something similar available.
MCP79411/12 RTC ... "One Million Ohms" ATtiny kit ... available at http://www.tindie.com/stores/JChristensen/

If cost is not an issue, Xbee is probably the better solution just because of maturity and support.  There is a full professional stack available, and even books out there written on it.

Nordic units will handle the range, especially if you use one with an amplified PA/LNA and a nice fat antenna.  Also, you can easily use frequency bands that are outside wifi and bluetooth spectrum.  I've had a very hard time getting the amplified units to lose signal from each other within a typical home, which has plenty interference and walls etc.

KirAsh4

Funny you say that maniac, I was able to get my Nordic units up and running within minutes of plugging them in.  My XBees are still sitting there.  I tried twice to get them to work, even bought the book "Building Wireless Sensor Networks" and I STILL could not get them to play nice.  So I figured, I must be missing something very fundamental here.  At the time, I needed a solution, and I needed it fast, so I went to the Nordic which, as I said, worked without any trouble.  One of these days I will go back and make the XBees my be-atch.  They're going to work whether they want to or not. :)

(yes, they do work, I was able to get them to respond under X-CTU, even downloaded the new firmware on them as Digi instructed me to when I foolishly called them thinking something was wrong with the units.  Turns out, something's wrong with the operator.)

PaulS

Quote
One of these days I will go back and make the XBees my be-atch.  They're going to work whether they want to or not.

What kind of XBees are they? Series 1 models are trivial simple to make talk to each other. Fire up X-CTU, change 3 parameters and write the changes, and they work. Series 2 are not that much more difficult.

cyberteque

I'd go with XBee.

With the amount of 2.4G around here I'm amazed at how well they work!
Have a look at my Rover 5 video, both the XBee and the camera are on 2.4G.
Neither minds a bit.
My analogue 2.4G video sender/receiver is totally swamped, I had to shut down all the Phones, cell and cordless, AP's, Laptops, Netbooks to get a picture.
So to use the analogue video link on my rover I bought a pair of 900Mhz XBee modules.


Funny you say that maniac, I was able to get my Nordic units up and running within minutes of plugging them in.


Good to hear it!  I've got a fair bit of resources if they help you at all

http://maniacbug.github.com/RF24
http://maniacbug.github.com/RF24Network

Go Up