Equalizer setting method

Abstract

An equalizer setting method includes, executed by a processor: obtaining an unequalized impulse response; generating a plurality of first impulse responses based on the unequalized impulse response and a plurality of preset buck ratios or based on the unequalized impulse response and a plurality of continuous-time linear equalizer transfer functions; generating a plurality of eye diagrams or a plurality of signal-to-noise ratios based on the plurality of first impulse responses and a plurality of sets of feedforward equalizer tap coefficients; incorporating a target feedforward equalizer tap coefficient, which corresponds to the one with the largest eye diagram parameter in the plurality of eye diagrams or corresponds to the largest value in the plurality of signal-to-noise ratios, into a target equalizer setting combination; and setting the equalizer using the target equalizer setting combination.

Claims

1. An equalizer setting method, comprising executing the following steps by a processor: acquiring an unequalized impulse response; generating a plurality of first impulse responses based on the unequalized impulse response and a plurality of preset buck ratios, or based on the unequalized impulse response and a plurality of continuous-time linear equalizer transfer functions; generating a plurality of eye diagrams or a plurality of signal-to-noise ratios based on the first impulse responses and a plurality of sets of feedforward equalizer tap coefficients; incorporating a target feedforward equalizer tap coefficient, which corresponds to the one with the largest eye diagram parameter in the eye diagrams or the one corresponding to the largest value among the signal-to-noise ratios, into a target equalizer setting combination; and setting an equalizer using the target equalizer setting combination.

2. The equalizer setting method as described in claim 1, wherein generating the first impulse responses includes: The maximum voltage of the non-equivalent pulse response is stepped down using these preset step-down ratios to generate these first pulse responses.

3. The equalizer setting method as described in claim 1, wherein generating the eye diagrams or the signal-to-noise ratios based on the first impulse responses and the feedforward equalizer tap coefficients includes: Each of the first impulse responses is taken as a target impulse response, and the unequalized impulse response is repeatedly compensated by one of the feedforward equalizer tap coefficients to generate the compensated impulse response, and the eye diagram or signal-to-noise ratio corresponding to the compensated impulse response is generated as one of the eye diagrams or one of the signal-to-noise ratios.

4. The equalizer setting method as described in claim 1, wherein generating the first impulse responses includes: The unequalized impulse response is subjected to cyclotron integration with the continuous-time linear equalizer conversion functions to generate the first impulse responses.

5. The equalizer setting method as described in claim 4, further comprising, before performing the cyclotron integral operation to generate the first impulse responses: The inverse Fourier transform is used to convert the transfer functions of these continuous-time linear equalizers, which are predefined in the frequency domain, into the transfer functions of these continuous-time linear equalizers in the time domain.

6. The equalizer setting method as described in claim 1, wherein generating the eye diagrams or the signal-to-noise ratios based on the first impulse responses and the feedforward equalizer tap coefficients includes: Apply these feedforward equalizer tap coefficients to these first impulse responses to generate a plurality of second impulse responses; Multiple sets of decision feedback equalizer tap coefficients are applied to the second impulse responses to generate multiple third impulse responses; and a Lagrange algorithm is performed on the third impulse responses to obtain multiple candidate equalizer setting combinations and generate multiple eye diagrams.

7. The equalizer setting method as described in claim 6, further comprising, before applying the feedforward equalizer tap coefficients to the first impulse responses to generate the second impulse responses: A set of initial feedforward equalizer tap coefficients is determined by the zero-force solution; wherein the set of initial feedforward equalizer tap coefficients is the first set of initial feedforward equalizer tap coefficients applied when applying these feedforward equalizer tap coefficients.

8. The equalizer setting method as described in claim 6, further comprising, before applying the decision feedback equalizer tap coefficients to the second impulse responses to generate the third impulse responses: Determine the sampling time of a clock cycle to obtain the application position of the tap coefficients of these decision feedback equalizers.

9. The equalizer setting method as described in claim 6, wherein generating the eye diagrams includes: The candidate equalizer settings are calculated using peak distortion analysis to generate the eye diagrams.

10. The equalizer setting method as described in claim 1, wherein including the feedforward equalizer tap coefficients corresponding to those with the largest eye diagram parameter in the eye diagrams in the target equalizer setting combination includes: The target equalizer setting combination is incorporated into the feedforward equalizer tap coefficients that correspond to the one with the largest eye diagram area, eye height, or eye width.

Images