Deaf Filtering; General - Safran CORTEX Series User Manual

3.3.2.3

DEAF Filtering

3.3.2.3.1

General

The equivalent channel of a satellite communication system is constituted of:
1. the emitter filter: for frequency division multiplexing (FDM), the pass-band emitter filter is
generally a cavity filter. Base-band filtering depends on the application.
2. the channel: the channel could be considered as an amplitude fading channel with additive
White Gaussian Noise. Amplitude fading is resolved by automatic gain control. One can add the
possibility of multi-paths created by the structure of the satellite that can be modeled by a Finite
Impulse Response filter.
3. the receiver filter: they are many possibilities, commonly used filters are Integral & Dump and
Root Raised Cosine (RRC) filters.
In order to get perfect recovery of transmitted pulses or symbols, the equivalent channel must verify the
well-known Nyquist criterion (for a symbol rate R = 2*fc where fc is the physical channel bandwidth, the
impulse response of the equivalent channel should be equal to 0 for each multiple of 1/R). This
property is verified for a chain made of a RRC emitter filter and a RRC receiver filter, as RRC are well-
known Nyquist filters. However, emitter filters at RF frequencies are very often cavity filter design to
meet the regulatory rejection specifications, they are able to work with the signal coming from the high-
power amplifier and they have a non-linear phase response. Moreover, changes of emitter filter
characteristics (magnitude and phase) could not be avoided when there are large variations of
temperature like those which take place in space. So, the equivalent channel seldom verify the Nyquist
criterion and this induces inter-symbol interferences (ISI) that are harmful for performances.
A second aspect that is harmful for performances is noise. Two types of noises are present: the thermal
noise and the interference noise, coming from close channels.
In order to reduce the ISI and the influence of noise, two types of receiver could be used:
The Matched Filter
-
The Wiener Filter
-
As they are, these filters are not useful because the equivalent channel is changing. So, they are
implemented through conventional equalizer which are adaptive filters, but these filters need a training
sequence which causes a payload loss.
A better solution for equalizing the incoming signal is to use an automatic adaptive equalizer.
© Safran Data Systems – IMP000074 e14r1
HIGH DATA RATE RECEIVER
HDR-4G+ USER'S MANUAL
This document is the property of
It cannot be duplicated or distributed without expressed written consent.
Ref. DTU 100782
Is.Rev 3.5
Date: June 1, 2021
Safran Data Systems
.
Page 80