Back to results
US 9,197,262 B2Granted
Low-power and low-cost adaptive self-linearization system with fast convergence
AssigneeBroadcom Corporation
InventorRoy G. Batruni
Life of this patent
Estimating the term…
Abstract
A signal processing method includes inputting a digital signal, providing a plurality of coefficients; and determining an output. The output is approximately equal to an aggregate of a plurality of linear reference components, and each of the linear reference components is approximately equal to an aggregate of a corresponding set of digital signal samples that is scaled by the plurality of coefficients.
Claims20
A system, comprising:
- a radio frequency receiver configured to receive an undistorted signal;
- a separation block configured to:
- generate, using a plurality of adaptable coefficient multipliers, a reference signal component of a signal and an error signal, wherein the signal is a distorted version of the undistorted signal,
- update the plurality of adaptable coefficient multipliers based on the error signal, and
- generate, based on the reference signal component, a target signal component of the signal;
- a digital signal processor (DSP) configured to:
- filter the reference signal component according to a plurality of filter coefficients to provide an output signal that is substantially similar to the target signal component, and
- adjust the plurality of filter coefficients to drive a feedback signal to a predetermined level;
- and
- a combiner configured to subtract the output signal and the target signal component to provide the feedback signal.
7 more elements
A system, comprising:
- a radio frequency receiver configured to receive an undistorted signal;
- a separation block configured to:
- generate, using a plurality of adaptable coefficient multipliers, a reference signal component of a signal and an error signal, wherein the signal is a distorted version of the undistorted signal,
- update the plurality of adaptable coefficient multipliers based on the error signal, and
- generate, based on the reference signal component, a target signal component of the signal, wherein the reference signal component and the target signal component are related by a first transfer function;
- and
- a digital signal processor (DSP) configured to:
- filter the reference signal component according to a second transfer function to provide a filtered version of the reference signal component, and
- adjust the second transfer function such that the target signal component and the filtered version of the reference signal component are substantially equal.
6 more elements
A method, comprising:
- receiving, using a wireless receiver, an undistorted signal;
- generating, using a plurality of adaptable coefficient multipliers of the wireless receiver, a reference signal component of a signal and a first error signal wherein the signal is a distorted version of the undistorted signal,;
- updating the plurality of adaptable coefficient multipliers based on the first error signal;
- generating, based on the reference signal component, a target signal component of the signal;
- filtering the reference signal component according to a plurality of filter coefficients to provide an output signal;
- subtracting the output signal from the target signal component to provide a second error signal;
- and
- adjusting the plurality of filter coefficients to drive the second error signal to a predetermined value.
5 more elements
Dependencies are inferred from the claim text.