Rec. ITU-R P.833-3 RECOMMENDATION ITU-R P.833-3 Attenuation in egetation (Question ITU-R 0/3) (99-994-999-00) The ITU Radiocommunication Assembly considering a) that attenuation in egetation can be important in seeral practical applications, recommends that the content of Annex be used for ealuating attenuation through egetation between 30 MHz and 60 GHz. ANNEX Introduction Attenuation in egetation can be important in some circumstances, for both terrestrial and Earth-space systems. Howeer, the wide range of conditions and types of foliage makes it difficult to deelop a generalized prediction procedure. There is also a lack of suitably collated experimental data. The models described in the following sections apply to particular frequency ranges and for different types of path geometry. Terrestrial path with one terminal in woodland For a terrestrial radio path where one terminal is located within woodland or similar extensie egetation, the additional loss due to egetation can be characterized on the basis of two parameters: the specific attenuation rate (db/m) due primarily to scattering of energy out of the radio path, as would be measured oer a ery short path; the maximum total additional attenuation due to egetation in a radio path (db) as limited by the effect of other mechanisms including surface-wae propagation oer the top of the egetation medium and forward scatter within it.
Rec. ITU-R P.833-3 In Fig. the transmitter is outside the woodland and the receier is a certain distance, d, within it. The excess attenuation, A e, due to the presence of the egetation is gien by: where: A e = A m [ exp ( d γ / A m ) ] () d : γ : length of path within woodland (m) specific attenuation for ery short egetatie paths (db/m) A m : maximum attenuation for one terminal within a specific type and depth of egetation (db). FIGURE Representatie radio path in woodland Tx Rx d γ Excess loss A e (db) A m Distance in woodland, d 0833-0 It is important to note that excess attenuation, A e, is defined as excess to all other mechanisms, not just free space loss. Thus if the radio path geometry in Fig. were such that full Fresnel clearance from the terrain did not exist, then A e would be the attenuation in excess of both free-space and diffraction loss. Similarly, if the frequency were high enough to make gaseous absorption significant, A e would be in excess of gaseous absorption. It may also be noted that A m is equialent to the clutter loss often quoted for a terminal obstructed by some form of ground coer or clutter. The alue of specific attenuation due to egetation, γ db/m, depends on the species and density of the egetation. Approximate alues are gien in Fig. as a function of frequency.
Rec. ITU-R P.833-3 3 The alue of maximum attenuation, A m db, as limited by scattering from the surface wae, depends on the species and density of the egetation, plus the antenna pattern of the terminal within the egetation and the ertical distance between the antenna and the top of the egetation. Figure shows typical alues for specific attenuation deried from arious measurements oer the frequency range 30 MHz to about 30 GHz in woodland. Below about GHz there is a tendency for ertically polarized signals to experience higher attenuation than horizontally, this being thought due to scattering from tree-trunks. 0 FIGURE Specific attenuation due to woodland Specific attenuation (db/m) 0 0 V H 0 3 0 MHz 00 MHz GHz 0 GHz 00 GHz Frequency V: ertical polarization H: horizontal polarization 0833-0 It is stressed that attenuation due to egetation aries widely due to the irregular nature of the medium and the wide range of species, densities, and water content obtained in practice. The alues shown in Fig. should be iewed as only typical. At frequencies of the order of GHz the specific attenuation through trees in leaf appears to be about 0% greater (db/m) than for leafless trees. There can also be ariations of attenuation due to the moement of foliage, such as due to wind. Measurements in the frequency range 900-800 MHz carried out in a park with tropical trees in Rio de Janeiro (Brazil) showed a frequency dependence of A m : where f is the frequency (MHz). A m = 0.8 f 0.75 ()
4 Rec. ITU-R P.833-3 The mean tree height was 5 m and the receiing antenna height was.4 m. 3 Single egetatie obstruction 3. At or below 3 GHz Equation () does not apply for a radio path obstructed by a single egetatie obstruction where both terminals are outside the egetatie medium, such as a path passing through the canopy of a single tree. At VHF and UHF, where the specific attenuation has relatiely low alues, and particularly where the egetatie part of the radio path is relatiely short, this situation can be modelled on an approximate basis in terms of the specific attenuation and a maximum limit to the total excess loss: where: A et = d γ (3) d : γ : length of path within the tree canopy (m) specific attenuation for ery short egetatie paths (db/m) and A et lowest excess attenuation for other paths (db). The restriction of a maximum alue for A et is necessary since, if the specific attenuation is sufficiently high, a lower-loss path will exist around the egetation. An approximate alue for the minimum attenuation for other paths can be calculated as though the tree canopy were a thin finitewidth diffraction screen using the method of Recommendation ITU-R P.56, 4.. It is stressed that equation (3), with the accompanying maximum limit on A et, is only an approximation. In general it will tend to oerestimate the excess loss due to the egetation. It is thus most useful for an approximate ealuation of additional loss when planning a wanted serice. If used for an unwanted signal it may significantly underestimate the resulting interference. 3. Aboe 5 GHz Attenuation through egetation is important for broadband wireless access systems. These systems are typically based on a star network, with a well positioned hub (or base station) sering many indiidual users with rooftop antennas. In many cases, signals will be obscured by egetation close to the user antenna. For simplicity, the hub antenna will be referred to as the transmitter and the user antenna as the receier. An empirical model of propagation through egetation has been deeloped for frequencies aboe 5 GHz. The model gies the attenuation through egetation as a function of egetation depth, taking into account the dual slope nature of the measured attenuation ersus depth cures.
Rec. ITU-R P.833-3 5 The model predicts the excess loss due to the presence of a olume of egetatie foliage which will be experienced by the signal passing through it. In practical situations the signal beyond such a olume will receie contributions due to propagation both through the egetation and diffracting around it. The diffracted signal can be estimated using the method gien in Recommendation ITU-R P.56, 4.. The dominant form of these two propagation mechanisms will then limit the total egetation loss. The model was deried from a database of measured data oer a range of frequencies 9.6-57.6 GHz, but also takes into account the site geometry in terms of the extent of illumination of the egetation, defined by the minimum illumination area, A min. The attenuation for a egetation depth, d (m), (in addition to free space loss) is gien by: ( R0 R ) A scat = R + d k exp d k (4) Here, the initial slope is: R = af 0 (5) and the final slope is: b R = (6) c f where f is the frequency (GHz) and the turnoer alue of attenuation, at which the scattered component of the receied field becomes of the same order as the attenuated coherent component, ( { }) A k = k A min 0 0 log0 0 exp exp R f f (7) A0 and the parameters a, b, c, k 0, R f and A 0 are gien in Table. TABLE Parameter In leaf Out of leaf a 0. 0.6 b.7.59 c 0.63 0.85 k 0 6.57.6 R f 0.000. A 0 0 0 A min, is the minimum illumination area defined as the product of the minimum width of illuminated egetation, min(w,w,w ) and the minimum height, min(h,h,h ) which corresponds to the smaller of the two antenna spot areas on the front and rear faces of the egetation. These heights and widths are determined by the eleational and azimuthal 3 db beamwidths of the transmit antennas and the
6 Rec. ITU-R P.833-3 physical width, w, and height of the egetation, h, shown in Fig. 3, where the egetation is assumed to be a rectangular block. If the transmit antenna has eleational beamwidth, φ T, and azimuthal beamwidth, θ T, and the receie antenna φ R and θ R then the minimum illumination area is defined as: A A min min = min = min r ( h, h, h ) min( w, w, w ) ϕ tan T, r ϕ tan R, h min r θ tan T, r θ tan R, w (8) FIGURE 3 Geometry to determine the minimum illuminated egetation area, A min (see equation (8)) w Tx r h r Rx h h w w d 0833-03 In practice r >>r and the beamwidth of the receier, B rx, is expected to be only a few degrees. Under these conditions the parts of equation (8) containing r will not normally be required. The diffraction loss for double isolated knife-edges A difw due to diffraction around the sides of the egetation and A difh due to diffraction oer the top of the egetation is calculated as in Recommendation ITU-R P.56, 4. The egetation loss, A, which is then found as the minimum alue of A difw, A difh and A scat. Figure 4 shows an example of the model for two cases of minimum illumination area (0.5 m and m ) and three frequencies 5, 0 and 40 GHz for egetation in and out of leaf. This model for the attenuation due to egetation as a function of depth through the egetation can be incorporated into deterministic models (such as ray-based tools using a 3D database of the local building and tree locations) to gie a more realistic prediction of the extent of coerage for a gien transmitter location. 4 Depolarization Preious measurements at 38 GHz suggest that depolarization through egetation may well be large, i.e. the transmitted cross-polar signal may be of a similar order to the co-polar signal through the egetation. Howeer, for the larger egetation depths required for this to occur, the attenuation would be so high that both the co-polar and cross-polar components would be below the dynamic range of the receier.
Rec. ITU-R P.833-3 7 FIGURE 4 Attenuation for 0.5 m and m illumination area, a) in leaf, b) out of leaf)* 60 0 GHz 50 40 GHz Attenuation (db) 40 30 0 0 5 GHz 0 0 0 0 30 40 50 60 70 80 90 00 Vegetation depth (m) a) 5 GHz 60 0 GHz 50 Attenuation (db) 40 30 0 40 GHz 0 0 0 0 GHz, m 0 0 30 40 50 60 70 80 90 00 Vegetation depth (m) b) 5 GHz, 0.5 m 5 GHz, m 40 GHz, 0.5 m 0 GHz, 0.5 m 40 GHz, m * The cures show the excess loss due to the presence of a olume of foliage which will be experienced by the signal passing through it. In practical situations the signal beyond such a olume will receie contributions due to propagation both through the egetation and diffracting around it. The dominant propagation mechanism will then limit the total egetation loss. 0833-04