Angle diversity in troposcatter communications. Some confirmatory trials

This paper outlines the theoretical background and describes confirmatory trials of an angle-diversity principle which can usefully supplement the frequency-and space-diversity techniques currently used in troposcatter communications. The trials involved modifying the feed of a 25 m steerable antenna to produce twin receiving beams, each about 20 arc minutes wide and splayed vertically by 20 arc minutes at operating frequencies near 2 GHz. Measurements were made of the signal loss due to specified beam deviations from the optimum angle of shoot, together with the crosscorrelation between the signals from the splayed beams. Tests on two troposcatter paths, terminating at the 25 m antenna at Chilbolton, Hants, gave working beam-deviation losses (mean values) of 1.1 and 0.7 dB for the respective paths, compared with 0.8 and 1.2dB predicted theoretically. Measured median-crosscorrelation coefficients between angle-diversity channels were, respectively, 0.22 and 0.16, underthese conditions, against theoretical values of 0.31 and 0.24. The trials have endorsed the supporting theory as a reliable guide to the angle-diversity behaviour of vertically splayed beams, while confirming some design procedures based thereon.