FLEX-6000 Signature Series – Maestro User Guide
Page 47
Copyright 2019 FlexRadio Systems. All Rights Reserved.
11 NOISE MITIGATION
In order to properly utilize the DSP noise reduction features of the Flex 6000 Series radios, it is
important to understand the type of noise causing interference. There is no single solution for noise
mitigation since different types of noise require different algorithms. There are two primary types
of noise that can be minimized using DSP techniques: white noise and impulse noise.
White noise is defined as random or uncorrelated noise with a uniform frequency spectrum over a
wide range of frequencies. The sound of rain is an example of white noise. Three techniques are
best used to improve signal to noise ratio in the presence of white noise:
• Reduced filter bandwidth
• Optimized AGC threshold (AGC-T) setting
• DSP Noise Reduction (NR)
Reducing filter bandwidth and optimizing the AGC threshold can significantly improve the SNR
without adding distortion or “coloring” the signal so long as the desired signal is not at the antenna
noise floor. However, DSP noise reduction (NR) can provide significant intelligibility improvement on
weak signals which may be near or below the atmospheric noise floor.
Impulse noise is a category of noise that includes almost instantaneous impulse-like sharp sounds
generated by voltage spikes from arcing power lines, automotive ignition systems, electric fences,
etc. Impulse noise can raise the wide band noise floor received at the antenna by tens of dB and thus
completely mask signals that would otherwise be readable.
Traditionally, “noise blankers” have been utilized to mitigate this type of impulse noise. These
techniques detect the noise pulses and literally turn off the receiver during the time of the impulse.
The problem with traditional noise blanking techniques is that they have no way to tell strong signals
on the band from impulse noise and can thus “mix” impulses with the strong signals to cause
unwanted interference. The Flex 6000 Series radios incorporate a Wideband Noise Blanking (WNB)
algorithm that can differentiate between modulated signals and impulse noise, virtually eliminating
the “mixing” problem found in traditional blankers. This WNB algorithm operates in real time over
the entire Spectral Capture Unit (SCU) bandwidth to detect and replace impulses with an estimate of
the desired signal.
To Next Page
Loading...
Need help?
General
