Advantech INTRAC-305 5, 5. TECHNICAL DESCRIPTION

INTRAC-305 TECHNICAL DESCRIPTION

5. TECHNICAL DESCRIPTION

This section looks at and explains the Operational Modes and

Functions of the INTRAC-305 and at the tracking algorithm

that makes the INTRAC-305 one of the most accurate

tracking antenna controllers available.

The INTRAC-305 achieves its very high accuracy satellite

tracking by building a model of the satellite’s orbit and then by

using that orbit model to direct the antenna.

The algorithm used to build the orbital model has been

continually developed and enhanced by Advantech AMT

Limited since 1983.

The tracking accuracy is typically similar to that achieved by a

monopulse system and can, under some conditions, be better

than that achieved by a monopulse system.

THE TRACKING ALGORITHM During initial acquisition the INTRAC algorithm tracks the

satellite using a third order (for each axis) unbiased tracking

filter. This algorithm dynamically adjusts the period between

the step cycles to match the perceived orbit inclination and

received beacon signal level fluctuations and noise level.

During this initial period the tracking accuracy is only very

slightly lower than the full long term INTRAC tracking

accuracy.

The most significant difference during the learning period (first

24hrs) is not the accuracy of tracking but the time for which

the system can predict in the event of loss of the beacon

signal. This can be overcome by using Intelsat IESS-412 or

NORAD data to establish an initial INTRAC model so that the

full prediction ability is available from the start. This model is

then modified and optimised by the INTRAC algorithm in the

same manner as it would continuously update a model it had

“learnt”.

The key to deriving a reliable and accurate orbital model is the

ability to derive accurate estimates of the many parameters

involved in the model. Much specialised noise processing

expertise and experience has been applied in the design of the

INTRAC algorithm to ensure that INTRAC can build an

accurate model and can maintain it even when the beacon

signal is subject to severe fluctuations.

The INTRAC algorithm uses a robust pointing error estimator to obtain the raw satellite

position estimate, normally at 10 minute intervals. The raw

satellite position estimate is filtered with a narrow

noise-bandwidth tracking filter to produce the basic, multi

parameter, orbital model. To correct short-term errors in the

basic model resulting from modelling error, windage and

satellite station keeping manoeuvres, the difference between