Mismatch error correction for time interleaved analog-to-digital converter over a wide frequency range

Area of Science:

• Electrical Engineering
• Signal Processing
• Instrumentation

Background:

• High-speed, high-resolution analog-to-digital conversion is critical for waveform digitization in physics experiments and other fields.
• Channel mismatch errors in Time Interleaved Analog-to-Digital Converter (TIADC) systems degrade performance, particularly for wide-band signals.

Purpose of the Study:

• To present a novel, fully digital correction algorithm for TIADC channel mismatch errors.
• To enable correction for wide-band input signals using a single set of filter coefficients.
• To investigate the algorithm's applicability to under-sampling situations and higher Nyquist zones.

Main Methods:

• Development of a fully digital correction algorithm for TIADC channel mismatch.
• Focus on wide-band signal correction, allowing frequency-independent filter coefficients.
• Simulation and experimental validation using two TIADC systems (1.6-Gsps 14-bit and 10-Gsps 8-bit).

Main Results:

• Significant performance improvement in both tested TIADC systems after applying the digital correction.
• Effective number of bits improved to >9.5 bits for the 1.6-Gsps 14-bit TIADC and >5.5 bits for the 10-Gsps 8-bit TIADC within their bandwidth (-3 dB).
• Successful validation of the correction algorithm for input signals in the second Nyquist zone (under-sampling).

Conclusions:

• The proposed fully digital correction method effectively mitigates TIADC channel mismatch errors.
• The technique enhances high-speed, high-resolution waveform digitization across a broad range of frequencies, including under-sampling scenarios.
• This approach offers a robust solution for improving TIADC performance in demanding applications like physics experiments.