
Overview
In 1967, several very smart people released a paper called Transmission Loss Predictions for Tropospheric Communication Circuits. It was a major attempt to estimate radio path losses based on real-world path geometries, ground conditions, atmospheric refractivity and other location-specific environmental characteristics.
This was very useful in predicting the actual coverage of TV and FM broadcast stations and for the then expanding two-way radio business.
Most people should already be familiar with the standard free-space path loss equation (in Perl):
# Free-Space Loss (dB) Frequency (MHz) Distance (kilometers) $free_space_loss = 32.447782 + (20 * (log10 $freq)) + (20 * (log10 $distance));
This equation predicts radio path loss under ideal, isotropic conditions (i.e. nowhere on Earth). Example: two antennas operating at 2450 MHz over a distance of 5 kilometers will have an estimated path loss of 114 dB. That is, any radio signal traveling between those two antennas will be reduced by a factor of 251,188,643,151 to 1. But in a real-world situation, there may be significantly more path loss because of atmospheric refraction, poor ground conductivity, knife-edge diffraction, trees, cows, etc.
The GBPPR staff has adapted the SPLAT! (Signal Propagation, Loss, And Terrain) software program, written by John A. Magliacane, KD2BD, to provide a web based, easy-to-use interface for calculating Longley-Rice path losses for a given transmitter location just about anywhere in the United States. It will produce a high quality PNG image of the estimated Longley-Rice path loss in reference to the local ground elevation for the area.
This tool, combined with the other GBPPR microwave and link analysis tools, should be very useful in real-world plotting of wireless access point coverage, repeater coverage, radio line-of-sight, etc.
Getting Started
Point your browser to the following URL: http://gbppr.dyndns.org:8080/longley.main.cgi. Yes, I know, it's slower than a $2600 writer - but just live with it. You should see a nice form with all kinds of strange questions to answer or select. The first five questions are fairly self-explantory. For the first question, enter the transmitter site's latitude in the degrees/minutes/seconds format. You are in the Northern Hemisphere, so don't enter a minus (-). The second question is for the transmitter site's longitude, also in the degrees/minutes/seconds format. You are in the Western Hemisphere, so your longitude is negative, but don't enter the minus (-) sign, as it's automatically taken care of in the program.
In the third question, you should enter the antenna height at the transmitter site. This is measured Above Ground Level (AGL). This means that if your antenna is on a 20 foot pole in your backyard, you'd enter 20 feet. The program will automatically calculate the ground elevation Above Mean Sea Level (AMSL) (i.e. zero elevation). You should ideally measure the antenna height to the center of the antenna's main radiating element (center-of-radiation). This height is measured from the ground to the middle of an omni-directional antenna or to the feed element on a parabolic dish.
The fourth question is for the highest transmitted frequency in megahertz (MHz). Don't mix up megahertz (MHz) and gigahertz (GHz)! For 802.11b setups, the frequency range is between 2402 and 2483 MHz.
The fifth question is for the transmiter and receiver's antenna polarization. You can only select horizontal or vertical polariation at the moment. If your antenna's main radiating element is parallel to the ground, it is horizontally polarized. If it's perpendicular to the ground, it is vertically polarized.
The sixth question is for the antenna height (AGL) at the receiver site. The same rules apply as at the transmitter site's antenna location.
The seventh question is for the ground dielectric constant (also called relative permittivity) of the general location. This should be a comprimise between both the transmitter and receiver location's ground type. A chart is provided below which shows the approximate dielectric constants for various locations around the United States. For the vast majority of urban, small city locations use a dielectric constant of 5.
| Surface Type | Typical Area(s) | Dielectric Constant (Relative Permittivity) |
|---|---|---|
| Salt Water | Oceans | 81 |
| Fresh Water | Lakes, Ponds | 80 |
| Pastoral, Low Hills, Rich Soil | Dallas, TX to Lincoln, NE | 25 |
| Pastoral, Medium Hills, Rich Soil | OH and IL | 14 |
| Flat Country, Marshy, Densely Wooded | LA near Mississippi River | 12 |
| Pastoral, Medium Hills & Forestation | MD, PA, NY, Exclusive of Mountains & Coastline | 13 |
| Pastoral, Medium Hills & Forestation, Heavy Clay Soil | Central VA | 15 |
| Rocky Soil, Steep Hills | Mountainous | 2-4 |
| Sandy, Dry, Flat, Coastal | Desert | 10 |
| Cities, Industrial Areas | Small Cities, Urban | 5 |
| Cities, Heavy Industrial Areas, High Buildings | Large Cities | 3 |
The eighth question is for ground conductivity measured in Siemens-per-meter (S/m). A map is provided below which shows the soil conductivity around the United States. It should be noted that the map shows the conductivity in milliSiemens-per-meter (mS/m). You'll need to manually adjust this to the correct format of Siemens-per-meter. Example: 5 mS/m is 0.005 S/m. Use 0.005 S/m (average ground) if you're too fucking stupid to read a map.

The nineth question is for the effective Earth radius, also called the K Factor. This is a "factor" which is used to correct for atmospheric refractivity of electromagnetic (radio) waves. Basically, radio waves don't actually travel in straight lines (like you read on Slashdot), they "bend" slightly to follow the Earth's curve. It's a pretty envolved process to calculated this bend factor exactly, so just use the 4/3 K Factor for now.
The tenth selection may be the most confusing, the selection for the local climate type. This is a rough indication of the local meteorological conditions for the general area. We're going to have to cheat a little bit on this one. Here is a Vigants-Barnett Climate Factor map for the United States:

Now, this map is for a completely different set of equations, but the overall climate types are fairly similar. Locate your location on the map and then use the following key:
If you are unsure about your location, use the climate type: Continental Temperate. The various maritime and subtropical climates depend on the amount of salt water in the air or surface. You may want to do your own research on that for your specific region.
The eleventh and twelfth selections are for the confidence and time variabilities. The default is set for a 50% confidence that reception will be received 90% of the time. This is the standard for mobile and two-way radio setups. Systems using digital transmissions may wish to select a 90% confidence and 90% time variability.
The three final miscellaneous questions are for the general specifications of the produced coverage plot image. The state selection allows city names and county boundries to be draw on the plot. The coverage radius selection allows you to reduce the coverage plot, which increases the plotting speed significantly. The final selection chooses the quality of the final output coverage plot. Select Low / Fast to produce a crappy JPEG image or High / Slow to produce a high quality PNG image. Obviously, the higher quality image will take longer to produce and download. Select the lower quality image to get a general idea of the coverage area, then switch to the higher quality setting when everything looks O.K.
Results
After double checking your data, it's time to press the Submit button. After what will seem like an eternity, you should see a picture that looks like something shown at the beginning of this article. It will take awhile. The program is very CPU intensive and my computer is very slow. Don't hit Stop or anything while it's computing or I'll hunt you down and beat you like a Democrat.
The script will output what parameters you entered and when finished will show the Height Above Average Terrain (HAAT) for the transmitter site and the standard FCC Part 73.313(d) azimuths.
Below that is the actual Longley-Rice Path Loss Analysis coverage plot image. It may be quite large, so adjust your browser accordingly. The rainbow of colors represent path losses in 10 dB increments. The color key is shown below:
| 80 dB | 90 dB | 100 dB | 110 dB | 120 dB | 130 dB | 140 dB | 150 dB |
| 160 dB | 170 dB | 180 dB | 190 dB | 200 dB | 210 dB | 220 dB | 230 dB |
When looking at the coverage plot output, the GREEN color represents approximately 80 dB loss, the PINK is 90 dB loss, the CYAN is 100 dB loss, etc.
Is it Good or is it Wack?
Now that you have the estimated path losses for a 360° coverage area from your transmitter site, you can (somewhat) accurately predict the received power level (in dBm) at any receiver location.
Say your transmitter is putting out +36 dBm (4 Watts) Effective Isotropic Radiated Power (EIRP) and your receiver has a threshold (sensitivity) of -90 dBm (7 microvolts). In any location in the 80, 90, 100, 110, 120 and 130 dB Longely-Rice path loss zones, you should be able to establish a wireless link (provided the receiver is using a 0 dB gain antenna and no feedline losses). That was determined by taking the EIRP (in dBm) and subtracting the path loss, then adding the receiver antenna gain (minus feedline loss) and comparing it to the receiver threshold level.
Example:
Now, this is under ideal conditions. The program doesn't take into account man-made objects or foilage blocking the path - which will really kill a wireless signal in the microwave bands.
An interactive CGI script to determine EIRP (and K factor) is available at: http://gbppr.dyndns.org:8080/wireless.super.main.cgi. This is the GBPPR Wireless Network Link Analysis - Super Edition script which will calculate a whole bunch of variables. It may be hard to figure out at first, so play around with it and if you're still having trouble you can email me. Determining the receiver's threshold may be difficult, so ask the manufacture if you have any questions on that.
Notes
You may download the original Technical Note 101 - Transmission Loss Predictions for Tropospheric Communication Circuits, at the following website:
http://its.bldrdoc.gov/pub/ntia-rpt/tn101
A point-to-point microwave radio path analysis utility is available at:
http://gbppr.dyndns.org:8080/path.main.cgi
Additional interactive wireless / RF design utilities may be accessed at the main GBPPR website: