Fig. 4. Frequency for which the transmission efficiency remains equal to or greater than 809? as a function of persistence time for the best and worst conditions encountered. model 1, potentially large errors in the model can only be due to errors in the probabilities of a cflos and not to details of the Tj dependence. The statistical sample size of Lund and Shanklin's C} data is large and they have reportedly checked the applicability of their model to other mid-latitude sites, so we reassert the belief that model 1 places a lower bound on the expected performance of space-to-Earth transmission using lasers. For a few climatic conditions, the sky-cover occurrencefrequency curve is J-shaped as shown by the dashed histogram in Fig. 5. If the intermediate values of Kj are statistically significant, i.e., K2_6 £: 0.1, then observable differences between the predictions of models 1 and 2 will develop under these conditions. Most of the sky-cover data, however, are U-shaped and large differences between the first two models never occur. Laser power transmission to a single, dedicated site cannot achieve the power availability and persistence time necessary for commercial viability of the laser-SPS concept. Multiple receptor sites must be available for each transmitted beam, and rapid-switching capability is essential. The development of a power-availability model for multiple sites requires an estimate of the joint probability of a cflos for at least 1 of N available sites (P) which, in turn, requires a climatic record of sky-cover observations taken simultaneously from each of the N sites (11). Climatic summaries of sky-cover occurrence frequencies, as used previously, will not suffice. Such simultaneous data are scarce, and we can only project estimates of the number of sites N and centroid radius r necessary to achieve viable performance levels. If we neglect the contribution of partial transmission through certain cloud types, then our calculated power availability (P) and Lund's joint cflos probability are identical for N=l. Obtaining commercially acceptable transmission frequencies,
RkJQdWJsaXNoZXIy MTU5NjU0Mg==