This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Multiple CC1120, one receiver - Listen-Before-Talk Throughput?

Other Parts Discussed in Thread: CC1120

Hi,

I have a number of questions here. I'm looking at the CC1120 as a possiblity to use in a product, but I've no expert in RF so hoping to gain some better understanding of the limitations before proceeding with design.

I need up to 30 transmitters, each sending 35byte packets, up to 5 packets per second each to a single receiver. All will be within 300m of each other. I see in the datasheet that CC1120 automates Listen-Before-Send.

1: is there a handshake in the CC1120 packet transmission, so that the transmitter knows when the packet has been received and gets and acknowledgement of some kind? ie - is succesful data transfer thus (eventually) assured?

2: The fixed + random listening time is 5-10ms, so this would limit the possible transmissions to <100 - 200 per second if the transmissions were instantaneous.so thats 3 transmissions from each of my 30 transmitters I guess before things start to get really cramped. I guess in practice there will be collisions and retires with so many transmitters, so throughput would be drastically reduced. This is what makes me think this approach is entirely unsuitable for what I am trying to achieve. can I get a second opinion from someone who knows his onions?

3: If I could manage to convince the customer that one transmission per second was enough, how likely am I to encounter lots of traffic on the same bands from other stuff in the area. Let's assume this product was to be used in a packed olympic stadium.

4: If anyone has any insights into a technology type better suited to my application, I would be very pleased to hear it.

I thank you very much in advance for any input.


 

  • 1) One option is described in 8.2.4 in www.ti.com/.../swru295e.pdf. CC1120 implements the PHY in a protocol. Sending an ACK is typically something that has to be done in software.

    2) You are not saying anything about the datarate you plan to send with. Lest assume 50 kbps, 4 byte preamble, 4 byte sync, your packet will take 1/50000*8*(4+4+35+2)=7.2 ms to send. Even if you managed to send packets back to back a transmission 5 times a second from 30 transmitters would not be possible.

    LBT with random backoff is fairly effective since if lets say 3 transmitters try to send at the same time at time 0, the next time they try to send will be at random times.

    3) This will be very dependent on both the band you are operating on and the area.

    4) To increase the throughput one option is to use more than one radio in the receiver enabling the system to listen to more than one channel at the time.