Generating PAM4 Signaling For Receiver Compliance Testing
- PAM4 Signaling Overview
PAM4 signaling has gained popularity over the past few years with serial data rates approaching 56Gb/s per channel. This more bandwidth efficient 4-level signal modulation technique (compared to the simpler 2-level signal modulation technique (PAM2) that is most commonly referred to as NRZ) avoids the signal degradation caused by the increase in bandwidth. This is because by transmitting two bits per symbol, PAM4, for a given data rate, cuts the bandwidth in half. For example, a 56Gb/s PAM4 signal only runs at 28Gbaud/s compared to a 56Gb/s PAM2 signal that runs at 56Gbaud/s. Here we need to distinguish the symbol rate (referred to as the baud rate) from the data rate to make the comparison.
While a PAM4 signal experiences more ISI than a PAM2 signal at a given baud rate, it experiences much less at a given data rate. This minimization of ISI at a given date rate on bandwidth limited channels like electrical backplanes is the main motivation for using the different modulation technique.
This article focuses on a straight forward method to generate test PAM4 signaling for receiver evaluation. A variety of different signals are needed to subject a receiver to worst case operating conditions for standard (and future) compliance reasons. In this way, a receiver’s weaknesses are understood and its tolerance to various signal impairments is measured.
- Generating a PAM4 Test Signal
A block diagram of the proposed set-up to generate differential PAM4 signaling is shown in Figure 1. It includes two ADSANTEC ASNT6118-MOD boxes plus two Marki Microwave passive combiners. In order to generate a PAM4 test signal up to 32.5Gbaud/s, one of the ASNT6118-MOD’s needs to be designated as the MSB and given an output amplitude that is twice the size of the output from the other ASNT6118-MOD (now designated as the LSB). The resulting signal will meet all the electrical PAM4 signaling and transmitter performance parameters as specified in the standards for different symbol/data rates.
Input data and clock signals to the setup can come from common lab PRBS/pattern generators and clock synthesizers. An example of a suitable data source is the ADSANTEC (ASNT_PRBS45) – http://www.adsantec.com/276-asnt_prbs45.html. Phase matched cables are needed after the passive combiners.
The ASNT6118-MOD (shown in Figure 2.) is a 32.5Gb/s 4-tap pre-emphasis amplifier that contains numerous control settings in order to generate a clean (or stressed) output eye. Besides the pre-emphasis capability, the instrument features an adjustable output signal amplitude, a method to increase/decrease output signal under/overshoot, and output eye crossing and duty cycle adjustment. All functions are controlled by a simple, easy to use GUI. More information on the ASNT6118-MOD can be found here – http://www.adsantec.com/289-asnt6118-mod.html.
The Marki Microwave passive combiner (shown in Figure 3.) with over 40GHz of analog bandwidth has been specifically designed for generating PAM4 signals in this configuration.
- PAM4 Receiver Tests
A list of possible receiver tests that can be run using the setup shown in Figure 1 is presented below.
- Tolerance to Jitter and Noise
With four separate symbol levels to ascertain, PAM4 receivers must tolerate poor (compared to PAM2) signal-to-noise ratios and small symbol-to-symbol voltage swings. An easy method to judge a receiver’s sensitivity is to lower the input signal amplitude until the BER reaches an unacceptable level. This can be done by properly reducing the output amplitudes of both ASNT6118-MOD boxes in parallel.
Eye closure can also be accomplished by using the other already mentioned capabilities of the ASNT6118-MOD boxes including mainly the 4-tap pre(de)-emphasis to introduce ISI type characteristics. The boxes can be configured to model a full passive frequency dependent loss profile thus removing the need of having to have many physical reference channels (e.g. compliance test boards) to test with on hand.
Additionally, sinusoidal jitter (SJ) can be added to one or both of the boxes (via their data/clock input signal) to further decrease eye opening.
The measurement of a receiver’s sensitivity to input signal linearity and level separation mismatch is also needed. All three PAM4 eye diagram heights can be manipulated together or separately as desired in the proposed setup using the output amplitude adjustments on the boxes.
- Stressing of Clock Recovery
An essential test for a receiver’s clock recovery circuit consists of evaluating its tolerance to various degrees of SJ over a certain frequency range. The degree of SJ at each frequency is specified in the jitter tolerance mask given by the standards. In general, PAM4 receivers must tolerate both several UIs of SJ (between 5UI and 0.05UI) at SJ frequencies below the clock recovery bandwidth and limited UI SJ (0.05UI) for SJ frequencies above the clock recovery bandwidth.
As mentioned in the previous section, SJ can be added to the PAM4 test signal by introducing it to the data/clock inputs of the ASNT6118-MOD boxes. An easy way to do this is to use appropriate ADSANTEC active phase shifters and modulate (at a select frequency) the input signals with a sinusoidal control signal. More information on these active delay lines can be found here – http://www.adsantec.com/36-variable-delay-lines.
- Testing of Receiver Side Equalization Capabilities
A receiver’s equalization capability (CTLE and/or DFE) can be tested by subjecting it to the maximum ISI of any realistic transmitter and channel pairing. As stated earlier, ISI type effects can be generated using the 4-tap pre(de)-emphasis of the ASNT6118-MOD boxes.
- Tolerance to Electrical Receiver Interference and Crosstalk
PAM4’s present day and future standards (will) require separate compliance tests for cases of high interference/crosstalk with low ISI/insertion loss and low interference/crosstalk with high ISI/insertion loss. This makes sense since a receiver’s tolerance to interference/crosstalk can vary in different situations and thus separate tests are necessary. Again, the ASNT6118-MOD boxes can use their individual amplitude, 4-tap pre-emphasis, and jitter insertion capabilities to help create the required testing conditions. Further alterations of these elements can be used to gauge the receiver’s performance beyond the mandatory test conditions.
The system setup presented in Figure 1 provides users with an easy to use, complete, and extremely flexible test signal generator to evaluate PAM4 receivers over a broad range of frequencies. It is a simple approach to conduct the necessary tolerance tests to ensure compliance with the current and possible future PAM4 standards.