Olympus Satellite Research Articles

Engineering a hard real‐time system: From theory to practice

AbstractMore and more programmers find their software being used in performance critical applications. Unfortunately, they have limited techniques at their disposal to help guarantee this particular aspect of their programs. There has been considerable activity in recent years on developing analysis techniques for hard real‐time systems. Inevitably these techniques make simplifying assumptions so as to reduce the complexity of the problem to be solved. For example hard real‐time schedulability analysis techniques often assume that the timing properties of the underlying kernel can be accounted for by incorporating extra execution time into the application tasks. Furthermore, they assume that the application task structure is very simple and uniform. This paper considers the implications of using these techniques in the analysis of a typical single processor application, the attitude and orbital control system (AOCS) for the Olympus satellite. The paper outlines a common approach for estimating the response times for tasks, and then extends the scheduling equations so that they can be used in the engineering of realistic real‐time systems.

Phase calibration of Olympus 20 GHz switched polarisationsatellite beacon

By analysing slant path propagation events observed using the 20 GHz switched polarisation beacon of the Olympus satellite, it is shown that the phase offset on transmission between the two polarisations was approximately 49°, with the near-vertical polarisation lagging. Knowledge of this offset makes it possible to determine the medium transmission matrix with all terms properly calibrated in their relative phase.

Keynote paper: A performance analysis of a hard real-time system

This paper analyses the performance of an Attitude and Orbital Control System for the Olympus Satellite. The system was designed using HRT-HOOD and implemented in Ada 83 augmented with some of the facilities of Ada 9X. The program consisted of multiple tasks communicating via protected shared data areas. A worst-case execution time analysis tool was used to estimate the execution time of each task, and Deadline Monotonic schedulability analysis was used to predict the system performance, taking into account all kernel overheads. An instrumented version was also produced to gather statistics on the actual run-time performance.

Results of 20/12.5 and 30/20-GHz XPD-frequency scaling measurements with the Olympus satellite

Reports on three years of crosspolar discrimination (XPD) frequency-scaling measurements in Germany performed by the Research Center of Deutsche Telekom. Concurrent measurements were carried out at 12.5, 20 and 30 GHz using the OLYMPUS satellite. Two different types of 20/12.5 and 30/20 GHz statistics were analyzed: (1) equiprobability and (2) instantaneous XPD-frequency scaling statistics. The results of equiprobability crosspolar scaling studies for total and ice depolarization are in reasonable agreement with the scaling laws proposed by ITU-R with some deviations, however, depending on the depolarization mechanism. In addition, a variability analysis of instantaneous XPD scaling factors was carried out from their joint conditional distributions, showing the spread which can be expected in long-term scaled SPD statistics. Although total depolarization is strongly affected by ice, the variation of the scaled data is relatively small since the scaling behavior of rain and ice depolarization is very similar in the considered frequency range.

Frequency scaling of rain attenuation for satellite communication links

One year of copolarized signal data from the OLYMPUS satellite's 12, 20, and 30 GHz beacons were examined for frequency scaling of attenuation. The statistics of the ratios of attenuation in dB for the frequency pairs 30/20, 20/12, and 30/12 GHz computed at each 0.1 s-sample instant were found to be nearly independent of fade depth. It was found that attenuation in dB scales with frequency to the 1.9 power. Also, attenuation ratios computed from the separate statistics of attenuation at each frequency for the same level of occurrence are very close to those found from instantaneous attenuation ratios.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Olympus end of life anomaly — a perseid meteoroid impact event?

On August 11, 1993 at 23:32 Zulu (UT), the OLYMPUS satellite lost earth pointing and began spinning, leading to a chain of events which culminated with the satellite's demise. The spacecraft automatic control system attempted to despin and reorient the spacecraft but was unsuccessful. These manoeuvers caused significant fuel depletion. During reacquisition under manual control, it was determined that insufficient fuel remained to return to station. The decision was made to take OLYMPUS out of service, removing the spacecraft as far as possible from the geosynchronous orbit.The morning of August 12 was predicted to be the peak of the Perseid's meteoroid shower. There was the possibility that the meteoroid stream might reach storm conditions with relatively fresh material released during the last appearance of Comet Swift-Tuttle returning near the Earth. Assessments of the situation had indicated that it was unlikely to be a problem, but an on-board micro-accelerometer package was operated to record impacts, and operators and support staff were put on special alert in case of operational difficulties.This paper will describe the spacecraft, the sequence of events, tests and analysis, and make operational and design recommendations. While an impact by a meteoroid could not be proven, it is a possible scenario. The impact by a small meteoroid may have generated a plasma triggering a discharge of charged surfaces entering the grounded spacecraft via the umbilical and an external sensor. Such a scenario is particularly interesting for other spacecraft since the Perseid shower is likely to be worse for the next few years.

Empirical models for rain fade time on Ku- and Ka-band satellite links

Rain attenuation data from the OLYMPUS satellite beacon measurements are used to investigate fade time in the Ku and Ka bands. Using statistical procedures, an empirical model is developed which predicts fade time as a function of attenuation level, frequency of operation, and fade duration interval. Total annual fade times are predicted over a frequency span of 12-30 GHz for attenuation levels in the 3-18 dB range and for fade duration intervals of 30-60 s, 60-120 s, 2-5 min, and 5-20 min. The predicted fade times are in good agreement with the measured values. An alternate model, described by two simple relationships in two different ranges of attenuation level, is also presented by simplifying the original single-equation model. The simplified model accounts for fades associated with stratiform rain and thunderstorms separately.

Results from the Virginia Tech propagation experiment using the Olympus satellite 12, 20 and 30 GHz beacons

A comprehensive set of propagation experiments was performed using the Olympus satellite 12, 20, and 30 GHz beacons. This set of experiments is unique in North America because of simultaneous reception of signals spanning the Ku- and Ka-bands from the same orbital slot, which permits direct inference of the frequency behavior of signal variations. The elevation angle from the receiving site in Blacksburg, VA, to the satellite was 14 degrees. Beacon, radiometric, and weather data for one year were analyzed. The statistical results for rain rate, beacon attenuation, attenuation ratios, radiometrically derived attenuation, fade duration and fade slope are presented. They are important to the design of Ku- and Ka-band satellite communication systems. The beacon attenuation results include cumulative statistics for attenuation with respect to free space and with respect to clear air. Attenuation ratio data are presented using attenuation with respect to clear air to focus on rain effects. Instantaneous attenuation ratios computed from instantaneous beacon levels were found to be nearly identical to statistical attenuation ratios obtained from cumulative attenuation statistics at each frequency.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Prediction of scintillation effects on satellite communications above 10 GHz

The paper describes a model which predicts the effects of tropospheric turbulence on communication systems. Turbulent eddies are represented by cells crossing the Fresnel&apos;s zones of the earth-space link, with parameters determined from the physics of clouds and ground meteorological measurements. From this representation, the transfer function of the turbulent atmosphere is calculated as a function of the parameters of the liaison, as well as the corresponding scintillation. An inversion process then synthesises a turbulent layer statistically compatible with the actually measured scintillation, and from which the effects on communication systems can be evaluated. The inversion process is validated by data measured on the Olympus satellite of the European Space Agency. Hence, from scintillation measured on a satellite beacon at one frequency, a turbulent layer is synthesised, which yields the prediction of scintillation at other frequencies with other antennas, as well as the effect on any communication link, analogue or digital.

Fraction of ice depolarisation onsatellite links in Ka band

The contribution of ice depolarisation and hence implicitly that of rain depolarisation is estimated from two conditional cumulative XPD statistics, derived from three Earth-satellite links at 20 and/or 30 GHz, using ESA&apos;s OLYMPUS satellite.

Physical-mathematical model of dynamics of rainattenuationwith application to power spectrum

The author presents a physical-mathematical model of the dynamics of rain attenuation and applies it to predict the power spectrum of attenuation. Compared to experimental data at 19.77 GHz (beacon from satellite Olympus) the prediction is very good.

The performance of coherent BPSK in the presence of oscillator phase noise for ka-band pico-terminal inbound links

In microwave systems, the phase noise is well known to be a significant source of degradation especially for low data rate transmission, giving rise to a significant irreducible error rate. In an earlier study [1,2] based on the European Space Agency's Olympus Satellite 30/20 GHz payload parameters, it was found that several thousands of simultaneous transmissions of 50 to 200 bits/sec each could be supported from hand-held pico-terminal earth stations. It is important to use the most appropriate modulation scheme in the presence of oscillator phase noise for such low data rate systems. It is generally assumed that noncoherent or differentially coherent modulation schemes will perform best in applications where the coherence time of the oscillators is short compared to the bit period. However, our analysis of these schemes has shown that they are subject to irreducible error rates and minimum useable data rates. This has led us to reconsider coherent BPSK (CPSK). It is shown that the choice of the optimum noise bandwidth of the phase locked loop (PLL) in the carrier recovery circuit is very important when considering the coherent schemes. The PLL bandwidth must be wide enough to track the oscillator phase noise at its input, but, on the other hand, it must be as narrow as possible to exclude the thermal noise. Results show that coherent BPSK in the phase noise contaminated channel is also subject to an irreducible error rate, which increases as the noise bandwidth is reduced. For the optimum loop bandwidth, however, this appears to be lower than the corresponding figure for DPSK [3]. There is a minimum useable data rate for any given bit error rate (BER) requirement. Results of a complete analysis are presented in the form of BER versusE b/N o plots.

Use of radiometers in atmospheric attenuation measurements

Measurements from radiometers operating at 20 and 30 GHz were used to predict attenuation along the 14 degrees elevation angle from Blacksburg, Virginia, toward the Olympus satellite. Both NOAA Dicke switched radiometers and Virginia Tech total power radiometers collected data in Blacksburg for two months. The radiometrically predicted attenuation for the total power radiometers agreed well with that of the Dicke switched radiometers, indicating that the total power radiometer approach can be used for accurate attenuation prediction.

Correction of satellite beacon propagation data using radiometer measurements

Virginia Polytechnic Institute and State University has performed a comprehensive set of propagation measurements using the Olympus satellite beacons at 12.5, 20 and 30 GHz. Total-power radiometers were included in each terminal. Radiometer data were used both to set the absolute level of the beacon data and to predict path attenuation. The paper shows how radiometer data are used to correct beacon data for spacecraft-induced diurnal fluctuations and to extract clear-air attenuation.

Frequency dependence of slant-path amplitude scintillations

The Letter reports scintillation measurements using the beacons of the Olympus satellite at 12.5, 20 and 30GHz. Based on a period of 30 months, the frequency dependence of amplitude scintillations has been investigated. In the Letter, complete monthly distributions of short-term signal variances, which could be characterised by a single parameter, were taken into account. Results have been compared with those of an earlier investigation based on a one-year subset of the data base. The latter considered only variances which were exceeded for low time percentages. Furthermore, results have been compared with the existing ITU-R scheme and with experimental results found in the literature.

Results from 12- to 30-GHz German propagation experiments carried out with radiometers and the OLYMPUS satellite

The rationale for Ka-band propagation studies is discussed, and the relevant measurement program carried out in Germany by the Research Centre of the Deutsche Bundespost Telekom is reviewed. Current problems are briefly presented. Radiometer measurements from a 30 GHz site diversity experiment are discussed. The results indicate that availabilities of 99.9% can be achieved with diversity spacings of about 15 km. If, however, much lower values of around 97-99% are acceptable, margins of 3 dB are sufficient in single-site operation with gaseous and cloud absorption being of similar statistical importance as rain. Measurements carried out with the OLYMPUS satellite at 12.5, 20, and 30 GHz are also discussed. A comparison with the predictions based on the current CCIR procedures shows that rain attenuation is overestimated for time percentages above 0.01%, whereas depolarization is underestimated significantly. Instantaneous frequency scaling of attenuation and scintillations is discussed, and an improved scaling procedure for the latter is presented. >

Propagation research in Europe using the OLYMPUS satellite

The launch of the European Space Agency's (ESA) large telecommunications satellite OLYMPUS in the summer of 1989 offered a unique opportunity to study the effects of the atmosphere on satellite communication links at Ku-band (14/12 GHz) and Ka-band (30/20 GHz). The community of OLYMPUS Propagation Experimenters (OPEX) operates experiments at more than 50 locations covering all of Western Europe with its different climatic regions. The authors introduce the European propagation research scenario, summarize the major features of the current OLYMPUS campaign, discuss the significance of the first results obtained, and give an outlook for new activities.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Microscale diversity in satellite communications

Results from an investigation into a microscale diversity scheme for the purpose of mitigating attenuation due to tropospheric scintillations are presented. Data from an experiment using the 20-GHz Olympus satellite beacon were analyzed to evaluate the performance of short-baseline site-diversity systems. Results indicate that with a baseline separation of about 50 m (at an elevation angle of 14 degrees in this experiment) the signals received by the main and diversity terminals become decorrelated. Examination of several high-scintillation events shows that such a diversity technique could substantially increase the uptime of low-power-margin VSAT systems. The performance of microscale diversity for several baseline separations is evaluated in terms of a parameter called diversity recovery. Applications and conditions under which microscale diversity performs best are addressed. >

Frequency and attenuation dependent fade slope statistics

Results obtained from slant path propagation experiments carried out with the OLYMPUS satellite are reported. Concurrent attenuation measurements at 12.5, 20 and 30 GHz have been analysed with regard to the rate of change of rain attenuation (fade slope). The results indicate that pronounced fade slopes occur mostly in the high attenuation range.

Quadruple-site diversity experiment in austria using 12 GHz radiometers

The paper reports on a quadruple-site diversity experiment using 12 GHz radiometers in Austria. Separations between sites were 10.9 to 26.2 km, providing orthogonal path separations (vertical or horizontal) of between 9.3 and 25.9 km. The radiometers were directed along the future look-angle of the ESA satellite OLYMPUS (azimuth 223.16° from true North, elevation 25.43°). The measurements were conducted from May 1980 to April 1983, with dual-site joint uptimes exceeding 94.9% for all combinations. The experimental results indicate that, for this location and geometry, dual-site diversity performance is insensitive to separations above 10 km, all pairs achieving similar diversity gain statistics. The results were close to those predicted by the Hodge model, with the prevailing weather direction having a significant influence on diversity performance. In general triple- and quadruple-site groupings provided better improvement in diversity than the dual-site combinations but, while the statistical improvement was slight, the number of joint events was significantly reduced from dualsite combinations.