Back to results
US 2018/0006857 A1Published application
Drift Robust Convolutional Induced Reception Scheme with Noise Prediction
AssigneeApple Inc.
InventorYuval Avner
Life of this patent
Estimating the term…
Abstract
This document describes techniques for decoding a convolutionally coded signal using a trellis decoder in a drift robust manner. A convolutionally coded and differentially modulated signal may be received. The signal may be decoded using a trellis. A noise prediction loop may be used to reduce noise characteristics of the signal. A frequency offset estimation loop may be used to reduce a frequency offset drift of the signal. The noise prediction loop and the frequency offset estimation loop may be applied at each branch of the trellis.
Claims20
A wireless device, comprising:
- an antenna;
- a radio operatively coupled to the antenna;
- and
- a processing element operatively coupled to the radio;
- wherein the wireless device is configured to:
- wirelessly receive a convolutionally coded and differentially modulated signal comprising a plurality of channel samples;
- and
- decode the differentially modulated signal using a trellis,
- wherein the wireless device comprises a noise prediction loop configured to reduce noise characteristics of the differentially modulated signal,
- wherein the wireless device further comprises a frequency offset estimation loop configured to reduce a frequency offset drift of the differentially modulated signal,
- wherein the noise prediction loop and the frequency offset estimation loop are applied at each edge of the trellis.
8 more elements
A method, comprising:
- at a wireless device:
- receiving a convolutionally coded and differentially modulated signal;
- and
- decoding the signal using a trellis, wherein decoding the signal using the trellis comprises:
- computing error signals for each of a plurality of channel samples associated with a first edge connecting a first previous state of the trellis to a given state in the trellis, wherein each error signal is computed using a noise prediction loop and a frequency offset estimation loop;
- computing a first distance metric candidate for the first edge, wherein the first distance metric candidate is computed based at least in part on a distance metric value for the first previous state of the trellis and on a summation of absolute values of the error signals for each of the plurality of channel samples associated with the first edge;
- computing error signals for each of a plurality of channel samples associated with a second edge connecting a second previous state of the trellis to the given state in the trellis, wherein each error signal is computed using the noise prediction loop and the frequency offset estimation loop;
- computing a second distance metric candidate for the second edge, wherein the second distance metric candidate is computed based at least in part on a distance metric value for the second previous state of the trellis and on a summation of absolute values of the error signals for each of the plurality of channel samples associated with the second edge;
- and
- selecting a minimum of the first distance metric candidate and the second distance metric candidate to be a distance metric value of the given state.
7 more elements
An integrated circuit, comprising:
- a trellis decoder circuit portion configured to determine an information bit sequence from a convolutionally coded and differentially modulated signal;
- an error estimation circuit portion configured to calculate estimated error signals for channel samples at each state of the trellis decoder;
- a noise prediction circuit portion configured to perform noise prediction using the estimated error signals for channel samples at each state of the trellis decoder;
- and
- a frequency offset estimation circuit portion configured to estimate frequency offset using the estimated error signals for channel samples at each state of the trellis decoder.
2 more elements
Dependencies are inferred from the claim text.