I don't think that you can find one, simple, magic equation. There are several components involved, to answering the question about delivery rate:

1. The path loss. Many models exist for this, from the simplest free space propagation model, to complicated ones that consider terrain and antennas heights, such as the Hata model and the much more comprehensive Longley Rice model.

2. Transmit antenna and receive antenna gain. These are added (or subtracted, depending on the sign of gain in dB) to the path loss. Antenna lead losses, expressed in dB, a function of cable type and length, would be subtracted from antenna gain values.

3. Transmitter power. Expressed in dBm, this can be added up front of the dB values calculated in #1 and #2 above, to give you a dBm value which indicates signal strength at the receive antenna connector, to the receiver's tuner.

The path loss model gives loss as a function of distance. So at this point, if you have posited values for each of the variables involved, you will come up with a figure of dBm at the receive end's tuner. And you can determine if there's any chance for reception at all, based on the tuner's sensitivity threshold. (Such as, the signal must be at least -85 dBm, or at least -100 dBm.)

Now you can play games with the question on delivery rate.

At the receive end, you know the signal strength, and you need to measure noise level, or you can just express your delivery rates as a function of possible SNR values at the receiver. In other words, just because you have enough signal strength does not mean you will achieve reception. Someone might be running a noisy vacuum cleaner right next to the receiver, for instance, and forget about reception. So for this, you use Shannon's equation.

Capacity (b/s) = Bandwidth (Hz) * logbase2(1 + S/N)

That S/N is a power ratio, not expressed in dB but expressed as Wsignal/Wnoise. And you have to know the channel bandwidth you are using. (Bandwidth will also factor into the receiver sensitivity figure., so if you have that sensitivity figure, you've already assumed some sort of channel bandwidth.)

Shannon's equation is agnostic to modulation type and any forward error correction. Shannon determines the ideal capacity in b/s. You can trade off modulation type against FEC, Shannon doesn't care. But it will give you the ideal maximum capacity. In practice, you should degrade that ideal Shannon SNR margin by about 3 dB at least, for real world results. So for instance, if your decoder says that 15 dB of SNR is its threshold level, add 3 dB to the marginal SNR Shannon's equation gives you.

So if Shannon says that you can get 20 Mb/s with an SNR of 12 dB, in that channel width, and your decoder spec says it needs at least an SNR of 15 dB for error free decoding of 20 Mb/s in that channel width, and your noise measurements say that you can expect no better than 15 dB of SNR at that location, you just might barely achieve successful reception at 20 Mb/s. If your noise level is comfortably less, so your estimated SNR is, say, 11 dB, then you should notice very solid reception. Or you can change your modulation to a less aggressive one, or improve your FEC, both of which degrade the channel capacity, and achieve more robust reception.

Thanks so much Dr. Albert for the great and helpful answer

As for the link budget you can formulate it in an equation as:

Pr= Prad At Lp Ar , where Pr in watts is the received power at the output of the receive antenna, Prad in watts is the radiated power from the transmit antenna, At the transmit antenna power gain factor , Ar is the receive antenna power gain factor and Lp is the path loss as a power loss factor.

The receiver equivalent input noise N= kTB , where kT is the thermal energy and B is the bandwidth of the receiver.

So, one can get the signal to noise ratio S/N of the link as Pr/ N,

So in order to get the link capacity C which is required by you, then one can use the Shanon expression as proposed by the respected colleagues Albert as

C = B log2 (1+ Pr/N) = Blog2(1+ Prad At Lp Ar/N)

Which is the requested relation in an explicit form.

Best wishes