Fractional sampling and robust equalization
Abstract
The performance of a transversal filter as an equalizer varies widely depending on the sampling rate of the received signal. A conventional linear synchronous equalizer suffers from aliasing effects; its performance is very sensitive to receiver sampler phase. It is pointed out that a fractionally spaced equalizer is superior in performance with respect to both the problems. It can also synthesize the optimum linear receiver structure in one transversal filter. Two other problems common with all high data rate transmission schemes are channel fluctuation and sampler phase jittering (SPJ) at the receiver. It is shown that fractionally spaced (FS) equalizers can be more robust in the system performance with respect to SPJ than conventional synchronous equalizers. It can also be guided to achieve some kind of inverse filtering of the channel which gives robustness with respect to channel fluctuations; two different types of anti-Nyquist distortions are considered in this context in order to demonstrate such robustness.
- Publication:
-
Journal of the Franklin Institute
- Pub Date:
- March 1984
- Bibcode:
- 1984FrInJ.317..183S
- Keywords:
-
- Data Sampling;
- Digital Filters;
- Equalizers (Circuits);
- Performance Prediction;
- Robustness (Mathematics);
- Channel Capacity;
- Network Synthesis;
- Phase Error;
- Signal Distortion;
- Signal Reception;
- Communications and Radar