# Site Selection Route Selection Path Profile and Field Survey

5.4.1.1 Introduction. Site selection, route selection, path profile, and field survey are carried out in a similar manner as outlined and discussed in Sections 2.2, 2.3, and 2.5. Here, then, we will point out the differences and areas of particular emphasis.

5.4.1.2 Site Selection. Consider these important factors. Whereas on LOS systems several decibels of calculation error may impact hop cost by one or several thousand dollars, impact on over-the-horizon hops may be on the order of hundreds of thousands of dollars or more. Thus special attention must be paid to accuracy in site position, altitudes and horizon angles, and bearings.

Tropospheric scatter/diffraction sites will be larger, often requiring greater site improvement including fresh water, sanitary systems, living quarters, and more prime power and larger backup power plants. Radiated electromagnetic interference (EMI) is of greater concern. Takeoff angle (9et, der) is critical. For each degree reduction of takeoff angle there is a 12-dB reduction (approximately) in median long-term transmission loss.

5.4.1.3 Route Selection. The route should be selected with first choice to those sites with the most negative and last choice to those with the largest positive takeoff angle. The effect of slight variations in path length is negligible for constant takeoff angles. The transmission loss on an over-the-horizon link will vary only slightly for changes in path length of less than about 10 mi (16 km). In a given area, therefore, it is usually best to select the highest feasible site, which also provides adequate shielding from potential interference, even though this may result in a slightly longer path than some location at a lower elevation.

Reference 2 suggests the following formula to estimate takeoff angle, which is valid only for smooth earth (or over water paths):

where 9et, er is in radians, hts is the height in meters of the transmitting antenna above mean sea level (MSL), and hrs is the height in meters of the receiving antenna above MSL. Formula (5.1) is based on an effective earth radius of 4250 km, which is representative of a worst-case condition.

5.4.1.4 Path Profile. A path profile of a proposed tropo/diffraction route is carried out in a similar manner as in Section 2.3. Tropo engineers prefer the use of 3 paper. Takeoff angle is a more important parameter than ^-factor. Thus 3 paper may prove more convenient in the long run. Key obstacles to be plotted are the horizons from each site. The horizon is the first obstacle that the ray beam will graze.

Basic tropospheric scatter path geometry is shown in Figure 5.3. From the path profile we will derive the following:

d = great circle distance between sites (km) dLt = distance from transmitter horizon (km) dLr = distance from receiver to receiver horizon (km) hts = elevation above MSL of the center of the transmitting antenna (km) hrs = elevation above MSL of the center of the receiving antenna (km) hLt = elevation above MSL of the transmitter horizon point (km) hLr = elevation above MSL of the receiver horizon point (km)

5.4.2.1 Introduction. Free-space loss (FSL) is based on theory and on unfaded LOS paths; the calculated receive signal level (RSL) and the measured level will turn out to be within 0.1 and 0.2 dB or less of each other. Transmission loss equations and curves for diffraction and troposcatter paths are empirical, based on hundreds of paths in many parts of the world. The distances are measured in kilometers along a great circle arc.

Figure 5.3. Tropospheric scatter path geometry.

cited references contain empirical methods to calculate tropo transmission loss:

In our opinion, methods following NBS Tech. Note 101 are the most well accepted worldwide. The method to be described is based on NBS 101 as set forth in USAF Technical Order 31Z-10-13 (Ref. 1), which we have simplified. However, the reader is warned that there can be considerable variation between calculated median loss values and measured loss, as much as 6 dB in some cases. It is for this reason that the term service probability must be dealt with, as will be described.

5.4.2.2 Two Definitions. The terms time availability and service probability must be distinguished and understood. We will first calculate the long-term median basic transmission loss Lcr. This is then corrected for a particular climatic region to derive the basic median transmission loss Ln. If the RSL on a link were calculated using this value (i.e., Ln), the RSL would reach or exceed this value only 50% of the time. This is the "time availability'' of the link. Of course, we would wish an improved time availability of a link, usually 99% or better. If we design the link for a time availability of 99%, then 1% of the time the RSL will be less than the objective. In Section 2.7 we called this the link availability or propagation reliability.

The basic median transmission loss is described by the notation Ln(0.5,50), or more generally Ln(Q, q). Here the q refers to the time availability and Q refers to the service probability.

The service probability concept is used to obtain a measure of prediction uncertainty due to our lack of complete knowledge regarding the propagation mechanism, the semiempirical nature of the prediction formulas, and the uncertainties in equipment performance.

A service probability of 0.5 or 50% tells us that only half the links in a large population given identical input conditions will meet the time availability value. Often we engineer links for a service probability of 0.95 or 95%. Then only 5% of the links will fail to meet the time availability. The last step in calculating transmission loss is to extend the transmission loss to take into account prediction uncertainty, which we call here service probability.

5.4.2.3 Propagation Mode. There are two possible modes for over-the-horizon transmission: diffraction or troposcatter. In most cases the path profile will tell us the mode. For paths just over LOS, the diffraction mode will predominate. For long paths, the troposcatter mode will predominate. If the basic median transmission loss of one mode is 15 dB more than the other mode, that with the higher loss can be neglected, and the one with the lower loss predominates.

To aid in determining the mode of propagation, the following criteria (Ref. 1) may be used:

• The distance (in kilometers) at which diffraction and forward scatter losses are approximately equal is 65(100//)1/3, where f is the operating frequency in megahertz. For distances less than this value, diffraction will generally be predominant; for those distances greater than this value, the tropo mode predominates.

• For most paths having an angular distance (see Section 5.4.2.6) of at least 20 mrad, the diffraction effects may be neglected and the path can be considered to be operating in the troposcatter mode.

5.4.2.4 Method of Approach. First, we will discuss the equation for calculation of basic long-term tropospheric scatter transmission loss Lbsr, then the geometry and substeps to carry out the calculation. The next step will be to calculate the basic long-term diffraction transmission loss Lbd for two selected diffraction modes. Then we discuss the mixed-mode case, tropo/diffraction. The next operation is to calculate the reference value of basic transmission loss Lcr and extend to the basic transmission loss value Ln(0.5,50) for a region of interest. This value is then extended for the desired time availability, for the 50% service probability. The last step, if desired, is to again extend the value for an improved service probability, usually 95%.