Measurements of tropospheric scintillations on a 14 degree path at 12, 20, and 30 GHz

Abstract

This paper presents results of scintillation measurements on the OLYMPUS downlinks using fixed terminals at each frequency and a 20130 GHz movable diversity terminal. All terminals shared a common local oscillator system and were frequency coherent. Observed scintillation spectra were consistent with Tatarski's theory, showing high frequency slopes between -813 and -1 113, with most of the power below 1 Hz. Scintillation intensity at 20 and 30 GHz followed a scaling law, but comparisons of 12 GHz intensities with the higher frequencies showed more variability and an unexplained diurnal variation. Measured probability densities and cumulative distributions of scintillation intensity agreed with the Mousley-Vilar model, particularly at 12 GHz. Calculations based on the CClR scintillation model compared well with measurements for time percentages between 0.1 and 10% There has been increasing interest in the use of Very Small Aperture Terminals (VSATs) in satellite communication links operating in Ku and Ka bands. VSAT links can be strongly impaired by rain and by tropospheric turbulence. Turbulence produces scintillations, rapid fluctuations of signal caused by multiple scattering from the turbulent small-scale refractive index (n) inhomogeneities in the lower atmosphere. The theory of radio wave propagation in turbulent media1t2 has been widely studied and a number of confirmina experimental studies have been performed at optical wavelengths because inhomogeneities in n cause the twinkling of stars and image jitter in optical telescopes. Studies on satellite links at microwave frequencies are less common. The VA Tech OLYMPUS Eqmiuxxu OLYMPUS was an ESA geostationary satellite launched in July 1989. It carried three frequencycoherent propagation beacons at 12.5, 19.77 and 29.66 GHz. (These wil l be referred to as 12, 20, and 30 GHz in the rest of the paper). The satellite was seen from VA Tech at an elevation angle of 14. From August 1990 through mid 1993, the VA Tech Satellite Communications Group measured amplitude and phase of the OLYMPUS beacon signals using an experimental setup composed of fixed-location receivers at 12, 20, and 30 GHz and a movable, dualfrequency 20130 GHz diversity terminal. Each receiver was equipped with a total power radiometer, used to set the zero level of the beacon signal. Environmental data (rain rate, temperature, relative humidity, wind speed and direction and barometric pressure) were collected simultaneously. All the local oscillator signals were locked to the 12.5 GHz beacon via a frequency lock loop (FLL). Detection was performed in a 3 Hz bandwidth via a digital IQ circuit. The beacon signals were sampled at 10 Hz. The power budgets of the beacon receivers are summarized in Table 1. The measured dynamic range was about 20 dB for the 12 GHz system and 30 dB for th5 20 and 30 GHz receivers. See reference 3 for more details. Scintillations are short-term fluctuations of the received signal. They are conveniently characterized by the scintillation intensity, ox, the standard deviation of the signal attenuation (in dB) with respect to clear air. The square of a, is the mean square of the log-amplitude signal fluctuations. Scintillation intensity can be calculated theoretically, subject to assumptions about a number of important factors. Not least among these is antenna size, an in some cases it is useful to refer to values for hypothetical point receiving antennas. Differences observed for larger antennas are referred to by the termsnaperture smoothing and aperture averaging. In this experiment, the scintillation intensity for each one-minute measurement interval was computed in two steps. First a 30 second average attenuation with respect to clear air ACAn was computed for each sample (0.1s) for the 30 s interval centered on the sample time. This average was then subtracted from the samples giving: ACA = ACA, ACA, (1) The purpose of this averaging (which is equivalent to a high pass filtering) is to separate the fluctuations due to scintillations from those due to rain. Next the scintillation intensity was obtained by computing the standard deviation of the filtered data, in dB, as: opyright c 1994 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved Table 1: Power budget of OLYMPUS receivers. Terminal 12 GHz 20120 GHz 30 GHz Frequency, GHz 12.502 19.77 29.66 Polarization Y Switched X,Y Y ElRP (maximum, d B W 13.1 31.7 27.7 Pointing loss to Blacksburg, dB 4.0 10.0 10.0 Clear Sky Attenuation, dB .25 1 .O 0.8 Antenna Diameter, m 4.0 1.5 1.2 Pointing Loss, dB 0.3 0.1 0.2 Power available at antenna, dBW -147.0 -149.0 -148.8 Switching loss 0.0 6.0 0.0 Losses, dB 1.4 1 .O 2.1 LNA gain, dB 33.1 32.7 38.0 LNA noise figure, dB 2.45 2.92 4.10 Filter loss, dB 0.8 0.4 1 .O Mixerlpreamp gain, dB 25.0 28.0 27.6 Coax loss, dB 9.9 12.4 7.8 Signal at IF, dBW -101 .O -103.0 -94.1 System noise temperature, K 448.3 692.