Forum

  • Trying to understand link budget.  And I have been studying few articles and sites about it

    https://www.pasternack.com/t-calculator-link-budget.aspx

    I get it that if I want to find out the value of received power I will take all the gains and subtract the losses from the Transmitter to the Receiver and that will give me the end result basically left to right.

    What formula I can use to find out the Transmitted power if I have all the values on the right?

    Thank you.

  • There is no exact formula due to the variant behavior of RF between the transmitter (Tx) and receiver (Rx) and the antenna gain at the Tx.

    That is, if I know the received power and the gain of the Rx antenna and the distance (even if it's free space) between the Tx and Rx, I still cannot know the Tx power as I am missing information about the Tx antenna and the factors that may impact free space propagation.

    Now, given assumptions about freespace, I can get pretty close to the Tx EIRP (that which comes out of the Tx antenna), but I still cannot be certain because the assumptions would be just that, assumptions.

    Hope this helps, if I correctly understood your question.

    Tom

  • Thank you Tom for the reply that makes sense.  However what if I do have the value of the Tx antenna gain and cable loss, can this formula that we normally use to find out the Received Power be adjusted to find the Transmitted power?

    Received Power = Tx Power + Gains - Losses

    Thank you

  • The question you are asking sounds strange because we use link budget to determine what power level we need in order to have a receive signal high enough to meet our objective. For example, let's say that we have a requirement to deliver 100Mbps to a station AP. In order to accomplish this we have chosen 2 - 802.11n devices. We look in the data charts for our AP's and we see what our required MCS rate and associated received signal needs to be in order to achieve our goal of 100Mbps. That's when we start determining what our transmit power and antenna gain need to be in order to overcome the path loss we will incur by traversing a certain distance. The antenna gain and transmit power are variables that we can control in order to overcome the loss we will incur with FSPL. If your Tx power doesn't get you to your desired goal, you can either increase it or change antennas to increase your gain.

    Let me know if that helps. 

  • Thank you Eric so along the lines of what you said. 

    "That's when we start determining what our transmit power and antenna gain need to be in order to overcome the path loss we will incur by traversing a certain distance. The antenna gain and transmit power are variables that we can control in order to overcome the loss we will incur with FSPL."

    So based on that if I know what data rate I need to achieve and what Rx signal needs to be.  But I know the antenna gain and cable loss on the Tx side, how best I can determine at what Power in (dbm) Tx needs to transmit?

    Hope that makes more sense, please let me know if I can clarify more.

  • I would set the problem up like this:

    Required Receive Signal => -50 (I just made that up)

    FSPL = -100

    Now I know that in order to overcome the FSPL and achieve a -50 receive signal, I'm going to have to create a combined power level of -50.

    If you have been given a Tx and Rx antenna of 16dbi and both cable connections have -1db of loss then you will need Tx power of 20 to create the link.

    So solve for x with x=Transmit Power Control

    x+(16-1)-100 + (16-1)=>-50

    Combine antenna gain and cable loss

    x+15-100+15=>-50

    Combine the gain and subtract antenna gain from FSPL

    x+30-100=>-50

    x-70=>-50

    Solve for x

    x=>20

  • Awesome thank you that makes so much more sense now.  Just for understanding sake if I reverse the cable loss and antenna gain it should look something like this?  NOTE:  I'm not focusing on the received power just swapped values to understand the math behind it.  Hope that makes sense.  Thanks so much again.

    x+(1-16)-100 + (1-16)
    x+ -15 -100 + -15
    x+ -30 -100
    x+ -130 ....

  • Most of these calculations assume that the end-point devices actually meet the sensitivity requirements in the 802.11 standard.     For most new radios this will probably be true.   In fact, many new radios substantially perform better than the requirements by as much as 6 dB, especially at the lower rates.

    The problems start with older devices, or poorer quality controlled ones, that may meet the requirements on average, but have substantial variation between individual radios.    In that case you will need a higher margin.

    It really helps if you can test to the actual PHY specs for your radios, but very few people have this $$ equipment.    The best approach is to use high quality radio manufacturers, whose designs stress consistent power output, and sensitivity.   In addition, poor EVM measurements can definitely affect the (discoverable) range for mobile devices.

    Generally speaking, manufacturers list their "best" data, by channel and/or rate.   Rarely do they specify the variation from channel to channel, rate to rate, or unit to unit.    The best examples I have seen come from Cisco.  Qualcomm radios also tend to be very consistent in performance. 

    Always remember that you get what you pay for.  

    And as a side note, I have usually found that (measured) sensitivity is a much better predictor of range performance than is output power.

  • No, no, no. Look what you've done. You got to -130 by changing the gain in antennas to loss. Your antennas have positive gain. They should not be represented by -15. Antenna gain will always be a positive number. If it was negative it wouldn't be an antenna.

  • Ah I see what you mean I was going by the mathematical rule i.e; 1-16 = -15 but since we are talking about the antenna gain here, that does not apply and it will be a positive value.

    So as a rule antenna gain will always be positive.  Transmitter can be a negative value or positive.  Receiver can be a negative or positive.  Cables/FSPL will be a negative.

Page 1 of 2