Optically derived elevation angle dependence of fading for satellite PCS

Images of urban Japan taken vertically through a 180° fisheye lens were analyzed to derive, as a function of elevation the fraction of sky that is clear, shadowed by trees, or blocked by buildings. At 32° elevation, results match those derived from satellite measurements fit to a 3-state fade model. Using the same model, for the first time the elevation angle dependence of mobile satellite fading is predicted. INTRODUCTION Fading for mobile satellite communications can be modeled by assuming that distinct signal level statistics pertain to three major propagation states, i.e., when the lineof-sight is clear (C), shadowed by trees (S), or blocked by buildings (B). In [1], Karasawa et al. derived percentages of (C, S, B) by fitting L-Band satellite fade data obtained in urban Japan at 32° elevation to a cumulative probability distribution consisting of a weighted linear combination of Rice (C), Loo (S) [2], and Rayleigh (B) fading, as in f(v) = C*f , . (v ) + S*f, (v) + B* L ,. .(v), (I) J v\ / ^ J Rice ^ J J Loo V / J Rayleigh V / ' V / where f(v) denotes the density function for the signal envelope [3] and the individual fade distributions are the Ricean density function, /«« 0 = 7 exp-~(v 2 +a 2 ) 2a V cr (2) Loo's density function, f , . 8.686v 71 [ (201og(z)-/n) 2 ( v 2 + z 2 ) 2s 2PS and the Rayleigh density (3) v //fc^() = TP -v (4) 117 H'l _.„•,-. DUQ^ The parameters assumed for the three fade distributions, with a, the direct signal's voltage set to V2 for a direct power of Pj= 1, are given in Table 1. Table 1 Parameters for fade densities state C