Look-ahead delta sigma modulators with quantizer input approximations
First Claim
1. A signal processing system comprising:
- a look-ahead delta sigma modulator comprising;
a noise shaping filter to process a signal and generate N state variables, wherein N is an integer greater than or equal to two;
an approximation generator coupled to the noise shaping filter to generate approximated state variables determined from at least a subset of the N state variables; and
a quantizer coupled to the approximation generator to quantize quantizer input data determined from the approximated state variables and determine a quantization output value with an objective of minimizing quantization error at a current and at least one future time step.
1 Assignment
0 Petitions
Accused Products
Abstract
A signal processing system includes a look-ahead delta sigma modulator having an approximation generator to approximate quantizer input signals. A look-ahead delta-sigma modulator can be implemented as a noise shaping filter and a quantizer. The quantizer can be implemented as a function generator. State variables of the noise shaping filter provide the input data from which the function generator determines a quantizer output signal. Latter state variables are more dominant in determining the quantizer output signal. Accordingly, earlier state variables can be approximated to a greater degree than earlier state variables. The approximations can result in slightly lower output signal accuracy but can significantly decrease implementation cost. Additionally, latter state variables can completely dominate (i.e., be deterministic) the quantizer output signal. This situation can result in a further, slight increase in the non-linearity of one or more quantization region boundaries.
55 Citations
33 Claims
-
1. A signal processing system comprising:
a look-ahead delta sigma modulator comprising; a noise shaping filter to process a signal and generate N state variables, wherein N is an integer greater than or equal to two; an approximation generator coupled to the noise shaping filter to generate approximated state variables determined from at least a subset of the N state variables; and a quantizer coupled to the approximation generator to quantize quantizer input data determined from the approximated state variables and determine a quantization output value with an objective of minimizing quantization error at a current and at least one future time step. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19)
-
20. A method of processing a signal using a look-ahead delta-sigma modulator that includes a noise shaping filter having N state variables, wherein N is an integer greater than or equal to two, the method comprising:
-
filtering an input signal to the look-ahead delta-sigma modulator using the noise shaping filter to generate the N state variables; generating approximated state variables determined from at least a subset of the N state variables; and quantizing the quantizer input data determined from the approximated state variables to determine a quantization output value with an objective of minimizing quantization error at a current and at least one future time step. - View Dependent Claims (21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32)
-
-
33. An apparatus comprising:
-
means for filtering an input signal to a look-ahead delta-sigma modulator to generate N state variables, wherein N is an integer greater than or equal to two; means for generating approximated state variables determined from at least a subset of the N state variables; and means for quantizing the quantizer input data determined from the approximated state variables to determine a quantization output value with an objective of minimizing quantization error at a current and at least one future time step.
-
Specification