Apparatuses and methods for sample rate conversion
Provided, among other things, is an apparatus that converts a signal from one sampling domain to another, and which includes: an input line for accepting an input signal and a processing branch. The processing branch includes a branch input coupled to the input line for inputting data samples that are discrete in time and in value, a quadrature downconverter, a first and second lowpass filter, a first and second polynomial interpolator, and a rotation matrix multiplier that provides a phase rotation. The processing branch generates data samples at a sampling interval that differs from the sampling interval associated with the signal provided to the branch input, e.g., with the difference in the sampling intervals depending on fluctuations in the output period of a local oscillator. Certain embodiments include multiple such processing branches, e.g., operating on different frequency bands of the input signal.
1. An apparatus that converts a signal from one sampling domain to an alternate sampling domain, comprising:
an input line for accepting an input signal;
a processing branch, which includes: (a) a branch input coupled to the input line for inputting input data samples that are discrete in time and in value; (b) a quadrature downconverter, having an input coupled to the branch input, that uses sine and cosine sequences to convert the input data samples into quadrature baseband and in-phase baseband outputs; (c) a first lowpass filter coupled to the quadrature baseband outputs and a second lowpass filter coupled to the in-phase baseband outputs of the quadrature downconverter; (d) a first interpolator coupled to an output of the first lowpass filter and a second interpolator coupled to an output of the second lowpass filter, each also having an input for inputting a variable interpolant value; and (e) a rotation matrix multiplier having a first input for inputting the variable interpolant value and a second input for inputting complex-valued data samples,
wherein the rotation matrix multiplier applies a phase rotation to the complex-valued data samples at its second input,
wherein at least one of: (i) the rotation matrix multiplier is included within the quadrature downconverter or (ii) the second input of the rotation matrix multiplier is coupled to the quadrature baseband and in-phase baseband outputs of the quadrature downconverter, and
wherein input samples and output samples of said processing branch are different discrete-time representations of an underlying signal, with an output sampling interval associated with the output samples differing from an input sampling interval associated with the input samples by an amount that depends on the variable interpolant value.
2. An apparatus according to claim 1 , wherein said rotation matrix multiplier:
(a) is included within the quadrature downconverter and (b) performs rotation matrix multiplication by adjusting at least one of a phase or an amplitude of said sine and cosine sequences.
3. An apparatus according to claim 1 , wherein said rotation matrix multiplier is implemented as a distinct processing element and has its second input coupled to the quadrature and in-phase baseband outputs of said quadrature downconverter.
4. An apparatus according to claim 1 , further comprising a quadrature upconverter coupled to the outputs of the quadrature downconverter, which shifts a center frequency of said outputs of the quadrature downconverter to a different center frequency.
5. An apparatus according to claim 1 , further comprising:
a sampling error estimator having an output coupled to the processing branch,
wherein the output of the sampling error estimator provides a value that is proportional to fluctuations in a nominal oscillation period of a signal provided to an input of the sampling error estimator,
wherein fluctuations in said nominal oscillation period correspond to fluctuations in the nominal sampling period for the data samples that are provided to the input line, and
wherein the output of the sampling error estimator is used by the processing branch in the generation of said variable interpolant value.
6. An apparatus according to claim 1 , further comprising:
a resampling data buffer coupled to an output of said processing branch,
wherein the resampling data buffer receives data samples at a first time interval that is different from a second time interval at which the resampling data buffer delivers data samples.
7. An apparatus according to claim 1 , further comprising at least one additional processing branch that is identical to said processing branch, except that said additional processing branch operates on a frequency band that is different from a frequency band at which said processing branch operates.
8. An apparatus according to claim 7 , wherein at least one of said processing branches also includes an equalizer that is coupled to the outputs of the quadrature downconverter and has at least one complex tap.
9. An apparatus according to claim 8 , wherein the equalizer has plural complex taps.
10. An apparatus according to claim 1 , wherein at least one of the first lowpass filter and the second lowpass filter performs a recursive moving-average operation.
11. An apparatus according to claim 1 , wherein at least one of the sine and cosine sequences is generated using a direct digital synthesis method that employs digital accumulators and phase lookup tables.
12. An apparatus according to claim 1 , wherein at least one of the sine and cosine sequences is generated using recursive operations.
13. An apparatus according to claim 4 , wherein the quadrature upconverter shifts said center frequency using second sine and cosine sequences, and wherein at least one of the sine sequence, the cosine sequence, the second sine sequence or the second cosine sequence is adjustable in at least one of amplitude or phase.
14. An apparatus according to claim 1 , wherein the input signal is continuous in time and value and the branch input is coupled to the input line via at least one sampling/quantization circuit.
15. An apparatus according to claim 1 , wherein at least one of the first and second interpolators performs an interpolation operation based on a second-order function.
16. An apparatus according to claim 1 , wherein at least one of the first and second interpolators performs an interpolation operation based on a first-order function.
17. An apparatus according to claim 1 , wherein the difference between the output sampling interval and the input sampling interval provides compensation for sampling imperfections associated with an analog-to-digital conversion process.
18. An apparatus according to claim 1 , wherein the difference between the output sampling interval and the input sampling interval provides compensation for sampling imperfections associated with a digital-to-analog conversion process.
19. An apparatus according to claim 1 , wherein the variable interpolant value includes an amount that depends on fluctuations in a nominal sampling period associated with said input signal.