LTE Link Budget & MAPL Calculator

MAPL is the maximum allowable path loss: the most loss the path between the base station and a terminal can have while the cell edge still gets its service at the stated location probability. It is the EIRP less the receiver sensitivity, plus the receiving antenna gain, less the losses and margins. The uplink and the downlink each have one, and the smaller sets the cell. A propagation model turns it into a range: this calculator uses Okumura-Hata or COST 231 Hata only inside 150 to 2000 MHz, base station heights of 30 to 200 m and terminal heights of 1 to 10 m, and flags a range outside 1 to 20 km, or outside 1 to 100 km up to 1500 MHz, where ITU-R P.529-3 extends it. The LTE Cell Range tool takes the same MAPL to the 3GPP TR 38.901 models as well.

The base station receiver integrates noise over the resource blocks a terminal is scheduled, not the whole carrier. Using the channel bandwidth overstates the uplink noise floor by 10 log₁₀ of the ratio, which is about 14 dB for four resource blocks on a 20 MHz carrier. The error is pessimistic, so it can lead to more sites than a network needs.

No. At a wide uplink allocation the uplink usually limits, because the terminal has much less power than the base station. As the allocation narrows the uplink noise falls with it, and at some point the downlink becomes the limiting direction. With a typed SINR the calculator reports the allocation where that happens; with a cell edge data rate it reports the allocation that gives the uplink the most. A narrower downlink allocation does not help the downlink in the same way: the user receives only its share of the cell power, so the signal falls with the noise.

The calculator takes the data rate over the allocated resource blocks, allows for the frame share and overhead of the resource grid, and inverts the attenuated Shannon bound from 3GPP TR 36.942 Annex A.1. That curve is a generic 3GPP link level model rather than a particular vendor's equipment, so if you have a measured vendor figure you can enter the SINR directly instead.

Shadow fading is treated as log normal around the median path loss, so the margin is the standard deviation times the normal quantile for the chosen probability. For example, 90 per cent of cell edge locations needs about 1.28 standard deviations. The standard deviation comes from ITU-R P.1546-6 Annex 5 section 12 for the environment, or you can enter your own.

No. The range comes from a clutter based propagation model that does not see terrain, which suits a first estimate of cell size. For terrain aware prediction, use the Coverage Predictor.