Common Oscilloscope Equalization Questions Answered

1. Why do we need balance?

The higher the data rate, the worse the signal quality becomes, meaning there is more loss at higher frequencies. The more loss there is, the worse the eye diagram becomes. High data rates and a lossy channel will cause an eye that is open at the transmitter to be closed at the receiver. The receiver needs to be able to distinguish between a 1 or a 0; however, a closed eye will result in received errors and corrupted data. Equalization uses the voltage level of the other bits of data to correct the voltage level of the current bit, which changes the closed eye to an open eye, significantly reducing errors at the receiver.

2. What is the difference between decision feedback equalization (DFE) and feed forward equalization (FFE)?

Feedforward equalization (FFE) typically only looks at the current and previous bits to open the eye, whereas decision feedback equalization (DFE) can theoretically look at an unlimited number of bits. FFE is implemented in hardware and uses the same algorithm for all devices. DFE is adaptive and can vary from device to device. You can use DFE after using FFE, but you cannot use FFE after using DFE.

3. What role does the 90000A oscilloscope hardware play when equalizing the eye diagram using the Serial Data Equalization software?

The 90000A oscilloscopes have a lower noise floor, which means equalization will open the eye wider. Equalization will amplify any noise that is not caused by ISI, which by definition includes oscilloscope noise. Because the 90000A Series has the lowest noise floor in the industry, very little noise will be amplified. This will result in a more accurate equalization.

4. Why does equalization software need serial data analysis software?

If the real-time eye is fully closed, then the Infiniium clock data recovery (CDR) method unique to the Serial Data Analysis software must be used. The CDR is required for the DFE that closes the real-time eye.

Equalization applications are becoming increasingly important to today's digital designers. Higher data rates, longer backplanes, and interconnect distances result in significant attenuation of the signal at the receiver, which in turn closes the eye diagram at the receiver, making it difficult for designers to distinguish high from low signal levels, even with the most sophisticated decoding algorithms. Infiniium Serial Data Equalization (SDE) software leverages Infiniium's low noise floor and clock recovery algorithms, combined with new equalization software, to accurately characterize DFE, CTLE, and FFE. Now users can model equalization and de-embedding using just an oscilloscope and Infiniium SDE software.