• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 3 answers
  • Subscribers 54 subscribers
  • Views 802 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Does sub-1 GHz work for this?

Brian Peavey
Intellectual 390 points
Other Parts Discussed in Thread: CC1310, CC2543

I have an application that requires connecting 25 peripherals to a Windows host computer wirelessly.  The peripherals are all located within 10 meters of the host computer.  The peripherals need to send data at a rate of 300 kbps (~37 kBps) back to the windows host.  The peripherals will be battery operated and rechargeable, so low power draw is a consideration.  The system will be used for half a day, and then re-charged for half a day.  I've ruled out Bluetooth and the various 802.15.4 protocols (ZigBee, 6LowPAN, etc) as the max data rate for these is too close (or below) my target data rate.  Wi-fi seems like it would work - I can have lots of connections and the data rate supports my requirement.  I'm wondering if any of the sub-GHz protocols would work here.  I've never used any of these, and not sure how they would compare to a wi-fi based solution.  Any thoughts or advice?  Do any of the sub-GHz offerings make sense for this application?

Thanks!

  • 0
    TI__Mastermind 36375 points
    Brian,

    Do you need 300kbps from each node or is the 300kbps the sum of all 25 nodes?

    Regards,
    /TA
  • 0 in reply to TA12012
    Intellectual 390 points
    I need 300 kbps from each node.
  • 0 in reply to Brian Peavey
    TI__Mastermind 36375 points
    Then Sub1GHz is likely not the right answer, the fastest PHY we have is about 1.5mbps at 902-928MHz. So to create something like what you are asking for, I would recommend creating a system of 4 nodes to 1 collector and then scaling up by adding more collectors on the same PCB on the collector side. So to get 25 nodes, I would recommend going with 7-8 CC1310 on the collector side, where each is collecting a stream of data from 4 nodes.

    We do not have examples to show this, but I think it is theoretically possible.

    /TA
  • 0 in reply to Brian Peavey
    Intellectual 390 points
    Sorry - that's incorrect. It's 300 kbps for the system. Each node should be capable of transmitting a minimum of ~12 kbps.
  • 0 in reply to Brian Peavey
    TI__Mastermind 36375 points
    Ahh,

    Then Sub1GHz would work. Are you talking about deployments in the US only? I ask because the regulations are different outside of the US.

    I would recommend 1mbps or 1.5mbps mode of our radio and then simply setup a scheme where each radio gets a short frame of time to transmit its data and then the next radio and so on. This is called TDMA time division multiple access. I wrote an appnote on this topic some time ago and some software on some older devices.

    www.ti.com/.../swra433.pdf

    Actually this protocol could be used as is, as these devices are still in production and have USB built-in (which makes it easy to connect to a Windows host).

    Hope I did not confuse you with all the options.

    Regards,
    /TA
  • 0 in reply to TA12012
    Intellectual 390 points

    US only should be sufficient. A couple of follow-up questions -
    - How does the effort to develop this system with sub-1 GHz technology compare to using Wi-fi (802.11)? One advantage to wi-fi is that a Windows host already has wi-fi capabilities, whereas I believe some type of gateway would be needed for the sub-1 GHz solution, correct?
    - I'm guessing there will be a trade-off between development cost (time) and system cost. Do you have a relative ballpark guess on either of these factors when compared to using wi-fi? Will a wi-fi solution be quicker, but more expensive? Or are there other factors to consider?
    - Is one technology better than the other when considering power usage, and therefore battery size?
    - Nodes are actually sensors that will be placed on a person's body. Given this, does one technology provide better connectivity? For instance, at any point in time a person's body could be located between the sensor and the receiving station (PC).
    - Not sure how the USB connection will help here. All communications will need to be done wirelessly.

    - I was reading one of the TI Wireless Connectivity articles and see that 802.15.4g supports data rates up to 500 kbps.  How does this solution compare to the TDMA solution described in your App Note.  Which transceivers would support this solution?

    - The App Note references CC2543/4/5.  Aren't these all 2.4 GHz transceivers?  Does the technique described apply to other transceivers in the sub-1 GHz range?  If so, which would you recommend?

    Thanks...

    Brian

  • 0 in reply to Brian Peavey
    TI__Mastermind 36375 points

    US only should be sufficient. A couple of follow-up questions -

    - How does the effort to develop this system with sub-1 GHz technology compare to using Wi-fi (802.11)? One advantage to wi-fi is that a Windows host already has wi-fi capabilities, whereas I believe some type of gateway would be needed for the sub-1 GHz solution, correct?

    - Wifi is a standard that is used by almost all Windows based systems today and therefore you would not need a gateway. For Sub1GHz, this is not included in computers today and you will need a gateway. A gateway does not have to a big complicated this, think about your keyboard and mouse dongle that you plug into your computers USB port.

    - I'm guessing there will be a trade-off between development cost (time) and system cost. Do you have a relative ballpark guess on either of these factors when compared to using wi-fi? Will a wi-fi solution be quicker, but more expensive? Or are there other factors to consider?

    - I think range and power consumption are the two most important features. A second advantage is that you not operating your system inside the crowded 2.4GHz band and therefore might get a more reliable system and better user experience in deployments.

    - Is one technology better than the other when considering power usage, and therefore battery size?

    - Sub1GHz is widely used and recognized for best in class power consumption for a given range and data rate.

    - Nodes are actually sensors that will be placed on a person's body. Given this, does one technology provide better connectivity? For instance, at any point in time a person's body could be located between the sensor and the receiving station (PC).

    - if you are only transmitting from the body to a device that is also on the body then 2.4GHz is fully adequate. This is what Bluetooth is designed to do, its called a Personal Area Network (PAN). However, if you are transmitting from the body to a host that is further away (like a sports event where you are collecting data from a number of players on the field, then Sub1GHz is a better solution). In short there will be effects from proximity to a human body, but Sub1GHz is less than 2.4GHz.

    - Not sure how the USB connection will help here. All communications will need to be done wirelessly.

    - easy connection to a PC for creating a dongle (like described above).

    - I was reading one of the TI Wireless Connectivity articles and see that 802.15.4g supports data rates up to 500 kbps.  How does this solution compare to the TDMA solution described in your App Note.  Which transceivers would support this solution?

    actually the TIstack does not support 500kbps today, the max rate is 200kbps. The TI stack is a great start and a potential candidate. The stack has more overhead than a TDMA type stack because it is more flexible. So as I was trying to get you to a solution that could use close to 100% of the avialable data rate then my thoughts fell on TDMA type networks.

    - The App Note references CC2543/4/5.  Aren't these all 2.4 GHz transceivers?  Does the technique described apply to other transceivers in the sub-1 GHz range?  If so, which would you recommend?

    - this techniques is portable to Sub1GHz, but not something I have planned.

    Thanks...

    Brian

  • 0 in reply to TA12012
    Intellectual 390 points
    Thanks for the detailed responses! My main concern at this point is in developing my own upper layer stack/protocol. This is beyond the scope of what my client is willing to pay for - hence my interest in the TI Stack. Since this doesn't have the required throughput, I'm looking for other alternatives. Given the requirements of throughput, # of connections, and availability of a standard stack, is Wi-Fi my only choice? This is a "heavier" solution than I would like, but I'm not finding a suitable alternative that meets these requirements.
  • 0 in reply to Brian Peavey
    TI__Mastermind 36375 points
    Brian,

    Are you willing to wait for me to have a team member port the solution described in SWRA433 to the CC1310? I have been looking for a reason to have someone do this work for some time and this is a good opportunity for me to assign someone to do it.

    It will take 6-8 weeks.

    Regards,
    /TA
  • 0 in reply to TA12012
    Intellectual 390 points
    Sorry for the delay getting back to you TA. We met with our client, and they have decided to go with a Wi-Fi based solution. There was a desire to have as close to a "standard" stack to run on as possible. Thanks again for all the info - I think this will likely fit in well for some of our future work down the road with other clients.

    Brian