Hi there,
There are some kinds of RF module in seeedstudio, including RFID, 315/433MHz RF, RFBee(868~915MHz), Bluetooth,xBee, etc…
We’d like to know what kinds of RF you need most, and where are they often used in. We’d be also glad to know your special needs of RF.
Thank you all!
-Icing
Hi,
802.15.4 (e.g. based on TI 2500 series) would be great.
My dream solution would consist of the following 3 components:
This would service very simple sensors and actuators, all intelligence for this node would be handled else in the network. Battery life is key here, preferably a few years or solar powerable. Sort of smartdust 
This node can do local calculations and take local decisions. It can also command & control the light version. The atmega has an arduino bootloader for easy development, programming via FTDI cable or another Arduino. Low power is nice here as well, but not as important as on the light version
The latter could also be achieved by having a radio+USB connected to a laptop or embedded linux board. Low power is not crucial here as its probably fed by the powergrid.
In all cases cost is IMHO a driver on large deployments. E.g. if we can cut costs by half (e.g. by mass production) this could be a revolutionary change in the internet of things 
If you need help with the software side of things then I am happy to volunteer
Cheers,
Hans
Hi Hans,
Thank you so much for your advice.The purpose we release RFBee using cc1101 working in 868/915MHz is to bring much easier control of RF module, and much cheaper price than xBee. However the security of transmitting data is not guaranteed, which may need to be done by user as demands.
As for the 802.15.4, we will consider your valuable suggestions seriously, and we are going to be very glad to your help of software side things!
All the best!
-Icing
Icing,
Glad to be of help
With regards to encryption of radio transmission, I can imagine an AT command for the RFbee which sets an encryption key. Once a key has been set, the RFbee firmware can automatically start encrypting transmission. Another AT command could reset/flush the key when needed. Only receivers having the same key and using the same encryption algorithm would be able to decipher the payload. Another trick would be message authentication where the message would be plain text but the receiver would be able to tell that a trusted sender was the originator of the message. Again: if you need help with the algorithm or the coding, just let me know
The 802.15.4 standard solves more than “just” the encryption problem, it also controls frame validation, guarantees time slots and handles node associations. Yes, you can build such functionality in your own code, but using a standard increases interoperability, avoids reinventing the wheel over and over and leaves more flash for fun stuff
Cheers,
Hans
ps. the forum sends notifications from you@yourdomain.com, don’t know if thats a missing config item
Hi Hans,
We are so appreciate for your advice. That’s also what we want to know.
The hardware and software of RFBee is open source project, so that we expect to see more customers to improve them or design firmware for themself. Seeeduino will surely reward to those volunteers who make the project much better.
Yes, 802.15.4 standard will make the RF more compatible and robust, so we may put more energy on it.
Thank you all,
-Icing
Well, the RFbee is sold out now, maybe I pick up a pair once they are available again just for the fun of playing with it
By the way, if you want see the future of “Smart Dust”:
ansem.be/ans1801.html
A ISM radio running purely on scavenged energy
Icing,
if you want to use to the Atmel line instead of the TI line of devices the ATmega128RFA1
( atmel.com/products/zigbee/zi … ily_id=676 ) which consists of a ATmega1281 and AT86RF230 combination.
The ATmega128RFA1 is priced approx USD10 per piece at Mouser, but at a 100 the price goes down to USD 6 per piece
Cheers,
Hans
I will talk of what I want in terms of functionality. In doing projects I often want a reliable 2way data stream between 2 nodes within 200m of each other, That is the thing I want most, but that is not easy to accomplish with alot of the boards/chip sets out there. The amount of config is way too high, and the zigbee is really not designed to do that. Yes you can configure it if you spend the time learning it, but it took way too long to figure it out. The next case for me is being able to poll a set of sensors, and sometimes it would be nice to allow them raize an alarm condition.
Encryption is often needed for commercial products, but gets in the way of getting started. I rarely would be interested in hiding the content, but knowing it came from the proper node is important.
Been trying to get a couple of RFBees working, looks fun when the software gets a little better, I may find time to help out there. Open source sensor networks and RC functions are what interest me.
Hi guys,
Your advice really helps a lot, we are so appericiate for that.
As for zigbee, ATmega128RFA1 and TI 2500 series are both good choice, however we will consider the development tools, technical support as well as the cost price.
rich, RFBee is designed for the wireless commucation with very simple configuration. And it support point to point and point to multipoint, so they can be intergrated into your own sensor networks. We want them can be easily used without the complicated configurations. At present, it can only reach about 120m, and we may add the power amplifer in the feature.
BTW, we are preparing for a new batch of RFBee, they’ll be back in about two weeks.
All the best,
-Icing
Icing,
just stumbled upon a site about Atmel Z-link projects:
uracoli.nongnu.org/index.html
Especially the user projects page ( uracoli.nongnu.org/projects.html ) is inspiring
If you want to go high end (capacity and complexity, not cost ) the STM32W (ARM cortex M3 + 802.15.4) seems to be a nice runner up. ( st.com/mcu/inchtml-pages-stm32w.html ).
Another example of what the end result could look like is the MicaZ mote ( xbow.com/Products/productdet … px?sid=164 ) , currently the price is on request only but they used to be pretty expensive.
On the software side it would be cool to port Contiki Os ( sics.se/contiki/) to our platform which would make it useful for the professional use as well. (thus larger audience, thus larger series possible, thus lower price and thus more deployments etc… ) Contiki requires 2K Ram and about 40K Rom and there is already a port for the AVR Raven platform.
Cheers,
Hans
Hi Hans,
Thank you so much!
I beleive you are more professional and have more experience on wireless project than us . The infomation you supply looks great, and we will check them out and do some evaluation.
All the best,
-Icing
Icing,
thanks for the compliment !
Embedded IT is currently more of a hobby to me, but I do have a degree in it (long time ago though )
I’ll keep an eye open for more info
Feel free to ask any questions you have !
Cheers,
Hans
I’d like an RF module that can indicate what signals it’s receiving and how strong each frequency is. I’d like to be able to display that information on my PC, although that’s a supporting software detail.
I’d also like a module (could be the same one) that let’s me monitor various frequencies (possibly even assign them to groups indicating that they belong together for some purpose) and let me assign (an optional) protocol to individual frequencies to facilitate communication with various devices. That way I could use one unit but monitor several RF enabled devices in the house.
Mostly, what I described above is most importantly needed in the receiver function (grouping for instance), but transmission capability would vastly improve its usefulness (but harder to implement - even the antenna becomes more complicated).
Also, many functions could be implemented on the PC side, but implementation in the module (possibly excluding display and advanced options) would also greatly improve it’s usability with other modules, and after all, that’s what makes this so much fun!
One that can be programmed with reasonable support. That is with an embedded processor with more memory than a 168. Tired of not being able to do the right thing because I run out of memory.