About Adaptive Equalization (Digital Demod)

Adaptive equalization removes linear errors from modulated signals by dynamically creating and applying a FIR Finite Impulse Response (feed-forward) compensating filter. Linear errors can come from filters in a transmitter or receiver's IF or from the presence of multiple paths in the transmission path, such as reflections in a cable system. These types of problems appear as group-delay distortion, frequency-response errors (tilt, ripple), and reflections or multipath distortion.

Equalization allows measurement of some impaired channels and can be used to isolate linear from non-llinear error mechanisms. Equalization does not require symbol lock or prior knowledge of the signal (such as a training sequence) and is compatible with recorded data.

Filter Coefficients

By default, the equalization filter has a unit impulse response which yields a flat frequency response (only one tap in the filter has a non-zero value and data simply passes through the filter). The position of the unit impulse is a function of the filter length and is positioned to provide the most optimum efficiency for most situations. The position cannot be adjusted.

The equalization filter has a unit impulse response when:

the application is first run
the equalizer filter is reset
Points / Symbol is changed
the measured or reference filter is changed
the symbol rate is changed
the clock delay adjustment is changed
the equalizer filter length is changed
the application is preset

Aside from the above conditions, the application uses the last computed coefficients when equalization is enabled. For example, if equalization is used in a previous measurement, the application uses the coefficients from the previous measurement unless the equalization filter is reset or Points / Symbol is changed. Therefore, it is good practice to reset the Equalization Filter to initialize the filter coefficients before a measurement is started.

Important Concepts

The following paragraphs highlight important concepts about Digital Demod Adaptive equalization:

Digital Demod adaptive equalization uses the entire error vector (all sample points) for adaptive improvements to the equalization filter, while Custom IQ Modulation Analysis provides more robust equalization controls (including Minimize for EVM Error vector magnitude (EVM): A quality metric in digital communication systems. See the EVM metric in the Error Summary Table topic in each demodulator for more information on how EVM is calculated for that modulation format.).
Equalization is available most Digital Demod Formats with the following exceptions:
- Not available for FSK signals.
- Not available for EDGE signals when the Measurement Filter is set to EDGE (Windowed RC). Use the Equalization Filter when the Measurement Filter is set to None.
The primary application of the equalizer's impulse-response display is for evaluating multi-path environments. Multi-path environments usually require longer filter lengths.
Another application of the equalizer's frequency-response display is for evaluating the transmitter or receiver signal-path for errors such as passband ripple and group-delay distortion. Short filter lengths usually work well for these types of measurements.
The VSA estimates new filter coefficients with each measurement and then uses the new coefficients to adapt the filter for the next measurement.
Generally, there is no advantage to using more than 2 points/symbol when using equalization. Use more than 2 points/symbol for better resolution of such displays as eye diagrams but the tradeoff is slower measurement speed.
To see the channel frequency-response over the entire bandwidth of the signal, use 2 points/symbol or greater. The channel frequency-response cannot be seen over the entire bandwidth of the signal if 1 point/symbol is used.