Best SoC and Development Kit to use for small low power RF development (CC2511fx and SimpliciTI).

Richard,

Please see my responses below.

Richard Norris said:

We have been looking at developing a consumer device that uses a Low Power RF SoC solution.

 Basically we need some advice on the best low power RF SoC and Development Kit to purchase and whether the SimpliciTI protocol will meet our needs.  We also need to know what the best antenna DK is for these chips?

Here is a recently released antenna DK that you can take a look at.  The users guide has additional linkes to other antenna design notes.

 http://focus.ti.com/docs/toolsw/folders/print/cc-antenna-dk.html?sp_rid_pod1=MjEzOTYzNTA1MjcS1&sp_mid_pod1=3201316

 Here is some background on what we are setting out to achieve:

 

• We require as low cost a production as achievable given the specification. -- Probably a PCB antenna with a discrete balun if you have the real estate for it.
• We require a small form factor, low power RF capable solution over the 2.4Ghz range as to support the device Globally (I think 2.4Ghz is the best choice?).  -- Globally, 2.4GHz makes the best choice right now,  Smaller form factors can require more expensive discretes (i.e. chip balun, chip antenna, etc.).  Also, keep in mind that smaller antennas will cause the link budget to take a hit as the antenna aperture is reduced.
• The solution requires USB to interface with a personal computer system. (Windows based primarily, MAC and Linux would be useful)  -- probably a CC2511 as first recommendation.  There is also some example USB code for these systems which you can look at here http://focus.ti.com/docs/toolsw/folders/print/cc2511nano_refdes.html and http://www.ti.com/litv/zip/swrc088c.  The nano design includes a meandered F PCB antenna with a delay line balun, probably about the least expensive RF path you'll find.
• We haven’t specified exactly the wireless requirements but basically the solution needs to be able to talk to devices using point to point but using automatic discovery of multiple devices communicating with them in a synchronous fashion (needs to send (Tx) and receive (Rx) alternatively).  Basically allow devices to identify multiple clients and then talk to them individually in a ordered fashion. -- SimpliciTI does provide some support for localizing and joining networks.  Additional mechanisms can always be adapted in user code as well.
• The data packet size will be low, but not sure how much at present.  -- if less then 50 bytes then SimpliciTI can handle without needing the user to partition the packets, otherwise the user code will need to break down the packets into less than 50 byte chunks.
• The range of the device needs to be around 15 meters or more, roughly in building environment (possibly outdoor too) but not necessarily between rooms or walls.  What is the best antenna to use here?  -- Most antennas will handle this distance quite easily especially if there are no walls or floors to penetrate.  Note that if the device will be worn close to the body, some adaptions to the antenna may be necessary as the body absorbs 2.4GHz energy quite well.  It could also shield significants amount of energy if the paired radio is on the other side of the body.  Indoors this may not be much of an issue as reflected energy may reduce these effects.
• We have never developed a wireless protocol before so are hoping SimpliciTI will offer the fastest and most cost effective solution for this.  Is it tailorable enough for use to do multiple device discovery and point to point communication? -- SimpliciTI is fully configurable and is normally distributed as source code so you have full access to tweak what you like at all levels.  It can be a great starting point for developing proprietary protocols or simply use it as is.  If you are new to RF in general and especially packetized protocol stacks, SimpliciTI can ease the learning curve since it is not as full featured as other standards based stacks like Z-Stack (TI's Zigbee protocol stack).

 

We haven't specified this to the point that we have identified all the chip features required but looking at the TI solutions available they all look packed with features.

We have identified the CC2511fx as a possible candidate, as I believe SimpliciTI supports this although the development documents do not mention this.  We did start looking at the new CC2540 low power Bluetooth chip, but I cannot see if there any plans for SimpliciTI to support this chip?

At this point, SimpliciTI does not support the CC2540 device.  SimpliciTI does support the CC2511 part as it is basically the same as the CC2510 part with the USB peripheral included.  I haven't actually developed any code with SimpliciTI and the CC2511 USB stack code together so I cannot say the two pieces of software will work together without issues but it should get you significantly closer to your goal than where you are currently.

 Any advice would be greatly appreciated. TIA.

 Regards

NozFX

Regards,

Jim

Richard Norris said:

It will be interesting to find out whether we can achieve what we need to with SimpliciTI; if we are able to tweak the code though, that may give us the flexibility we may eventually require.  I'm hoping to keep the networking simple, although this may not be the case.  The concept is peer to peer, no repeaters or routers.  Its just the device will need to automatically discover and connect to peers.  Basically the devices need to be discoverable, a peer then tries to connect to a discovered peer.  As soon as connection established the devices no longer talk to any other devices until communication is finished.  Internally we will need to keep track of peers that have already communicated as to ensure reconnection does not happen within a set period of time (kind of cache).  I guess to realistically achieve this we will need to come up with some proprietry protocol on-top of SimpliciTI?

I would say yes, that your application will need to keep track of the associations.

Richard Norris said:

Is the development kit for the CC2510/CC2511 the kit to use for this?  Will there be SimpliciTI examples included with this DK or is this seperate with the SimpliciTI SDK?  Sorry, I guess I should just have a look ;) its just it isn't obvious from looking at the web site.

Yes, the CC2510-CC2511 DK is supported by SimpliciTI in the IAR port version.  There are a number of other kits supported by SimpliciTI as well.  Some are SoC type devices with the 8051 microcontroller integrated with the radio, CC2510-CC2511DK, CC1110-CC1111DK, CC2430DK, CC2430ZDK.

There are other platforms that are supported as well by SimpliciTI that use a 2 chip solution, Microcontroller + radio.  EXP430FG4618 + CC1100, EXP430FG4618 + CC2500, eZ430-RF2500, SmartRF05EB + MSP430F2618 + CC2520.

I saw your other post inquirying the same information regarding possible modules.  I'm not personally aware of a module provider that uses the CC2511, but LS Research does have a module that incorporates a MSP430F5437 + CC2500.  Now this particular MSP430 also does not have USB integrated, but there is an application note with using a TUSB3210 (USB-to-UART) device that could be added to your system.

Richard,

Check out these two links

http://focus.ti.com/docs/toolsw/folders/print/cc2511usb_refdes.html

and

http://focus.ti.com/docs/toolsw/folders/print/cc2511nano_refdes.html

Both are reference designs using the 2511 part.  You can download the users guides for both without purchasing them and also the code files associated with each.

SimpliciTI has sample applications that work directly on the first link.  You will need to incorporate the USB stack files from the software image that comes with it to enable USB operation as the SimpliciTI sample applications only work the radio.  Note that you will want to do your development via a CCDebugger (http://focus.ti.com/docs/toolsw/folders/print/cc-debugger.html)so the USB stack integration can be debugged as well.

As far as FCC qualified, you cannot cross that bridge until you have a unit packaged up as FCC qualification must be done on the final product.  Some RF modules claim FCC qualifications but unless they have a detailed specification of how it must be mounted, what enclosures it is qualified to operate within, what antennas can be used and their orientation/location, etc., they are most likely just pre-qualified which is just saying they have done some looking and there are no FCC related issues with a stand alone module; most modules fall into this catagory.  The above reference designs should work well for you for development and also get you a long ways towards a final product.  The PCB antennas are susceptable to de-tuning from whatever enclosure you wrap them in.  Although this will probably help you pass FCC testing as it will reduce power output in general, you will most likely want to re-tune the RF path and antenna once you are ready for production and before you go to FCC testing.

You should note that you don't have to do FCC testing to sell your product (at least you don't have to pay anyone to do it or you can do it yourself).  If you don't do FCC testing at all you simply cannot state that it "complies" with FCC regulations.  However, nothing can eliminate the fact that your product, if sold in the US, will "be subject to" FCC regulations.  The 2.4GHz band may be an unlicensed band but anything operating in it is still subject to work within the defined regulations.  If you do sell without any FCC testing, the penalties can be higher if your product is found to violate regulations.  As always, its a risk/reward balance that your company must decide how to persue.

Regards,

Jim