# US 2016/0241252 A1 · High Dynamic Range Analog-to-Digital Conversion with Selective Regression Based Data Repair

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2016/0241252 A1
- Document ID: US20160241252A1
- Status: Published application
- Kind: A1
- Application number: 15/040,214
- Filed: 2016-02-10
- Published: 2016-08-18
- Earliest claimed: 2015-02-16
- Assignees: Sound Devices LLC
- Applicants: Sound Devices LLC
- Inventors: Steven Popovich; Matt Anderson
- Web page: https://patentgrep.com/patent/US20160241252A1

## CPC

- H03M 1/188: {Multi-path, i.e. having a separate analogue/digital converter for each possible range}
- H03M 1/0675: {using redundancy}
- H03M 1/121: {Interleaved, i.e. using multiple converters or converter parts for one channel}
- H03M: CODING; DECODING; CODE CONVERSION IN GENERAL (using fluidic means F15C4/00; optical analogue/digital converters G02F7/00; coding, decoding or code conversion, specially adapted for particular applications, see the relevant subclasses, e.g. G01D, G01R, G06F, G06T, G09G, G10L, G11B, G11C, H04B, H04L, H04M, H04N; ciphering or deciphering for cryptography or other purposes involving the need for secrecy G09C)

## Abstract

A multi-stage analog-to-digital conversion method and system use window functions and translation to match high gain frames of data to target frames of data. The technique selects window data packets for the output stream based the stage of data having the highest gain satisfying selection criteria, such as requiring a frame of data for the respective stage to satisfy a predetermined accuracy of fit value compared to a target frame of data for a zero gain stage.

## Claims

### Claim 1

A method of converting an analog input signal to a high fidelity digital output signal, comprising the steps of:
transmitting an analog input signal from an analog signal source;
providing the analog input signal in at least two stages having distinct levels of amplification;
converting each resultant amplified analog input signal to a digital input signal for the respective stage;
assigning data samples for each stage to correspond to a frame in dependence on when they are received;
applying a window function overlapping in time to each frame of data to create windowed data packets for each respective stage;
providing the windowed data packets as inputs to a combiner;
selecting a windowed data packet for one of the stages to be used by the combiner to generate the digital output signal.

### Claim 2 (depends on 1)

The method according to claim 1 where each frame is stored in a buffer.

### Claim 3 (depends on 1)

The method according to claim 1, wherein a frame of data in at least one buffer for one stage is translated to match a frame of data from another buffer for another stage prior to the application of the window function.

### Claim 4 (depends on 2)

The method according to claim 2, wherein translation is performed utilizing least squares regression.

### Claim 5 (depends on 3)

The method according to claim 3, wherein performance of least squares regression utilizes auto-correlation and cross-correlation statistics collected between two digital data inputs over a time span equal to or greater than the span of each data buffer.

### Claim 6 (depends on 2)

The method according to claim 2, wherein translation includes adjusting at least one of gain, DC offset, slope or delay.

### Claim 7 (depends on 1)

The method according to claim 1, wherein the window function is constructed such that the addition of subsequent window data packets derived from a digital input signal for the same stage would substantially reproduce the original digital input signal for the stage.

### Claim 8 (depends on 2)

The method according to claim 2, wherein the decision on which windowed data packets to select for a given frame is based at least in part on an accuracy of fit parameter created during the translation of one data buffer to match another.

### Claim 9 (depends on 8)

The method according to claim 8 wherein the translation coefficients are locked to a predetermined value if the accuracy of fit parameter indicates a high degree of fit for an extended period of time between translated data and target data.

### Claim 10 (depends on 7)

The method according to claim 7 wherein the accuracy of fit parameter is defined by the sum of squares for the difference between at least one buffer and at least one translated buffer.

### Claim 11 (depends on 1)

The method according to claim 1, wherein the analog input signal for one stage has an amplification gain of zero.

### Claim 12 (depends on 1)

The method according to claim 1, wherein the analog input signal for one stage has an amplification gain of 30 dB.

### Claim 13

A method of converting an analog input signal to a high fidelity digital output signal, comprising the steps of:
transmitting an analog input signal from an analog signal source;
providing the analog input signal in at least two stages having distinct levels of amplification;
converting each resultant amplified analog input signal to a digital input signal for a first stage and to digital input signal for a second stage;
translating a frame of data for the second stage to match a frame of data for the first stage;
providing the translated frame of data for the second stage and the frame of data for the first stage for use by a combiner to generate the digital output signal.

### Claim 14 (depends on 13)

The method according to claim 13 further comprising the step of storing frames of data samples for each stage to a corresponding data buffer prior to the translating step.

### Claim 15 (depends on 13)

The method according to claim 13, wherein translation is performed utilizing least squares regression.

### Claim 16 (depends on 15)

The method according to claim 15, wherein performance of least squares regression utilizes auto-correlation and cross-correlation statistics collected between two digital data inputs over a time span equal to or greater than the span of each data buffer.

### Claim 17 (depends on 13)

The method according to claim 13, wherein translation includes adjusting at least one of gain, DC offset, slope or delay.

### Claim 18

A multi-stage analog-to-digital conversion system comprising:
a first stage and second stage analog-to-digital converter, each having an analog input transmitted from a common analog input source and each outputting a digital input signal, wherein the amplification gain of the second stage analog-to-digital converter is greater than the amplification gain of the first stage analog-to-digital converter;
a digital processor that receives the digital input signal from the first stage analog-to-digital converter and the digital input signal from the second stage analog-to-digital converter and outputs a constructed digital output signal, said digital signal processor providing:
means for applying a window function that overlaps in time to the frames of digital input data for each stage to generate windowed data packets for each stage;
and
a combiner configured to select a windowed data packet for one of the stages and sequentially combine the selected data packets to construct the digital output signal.

### Claim 19 (depends on 18)

The system of claim 18 further comprising data buffers for storing frames of digital input signal data for each stage.

### Claim 20 (depends on 18)

The system of claim 18, wherein the digital processor further provides translation means for matching data from the data buffer for the second stage to the data buffer for the first stage prior to the application of the window function.

### Claim 21 (depends on 20)

The system of claim 20, wherein the translation means includes the operations of at least one of scaling, adding a DC offset, adjusting a slope or delaying.

### Claim 22 (depends on 21)

The system of claim 21, wherein the translation means operates on a basis of minimizing the sum of squares for the error between a data buffer and translated data buffer.

### Claim 23 (depends on 18)

The system of claim 18, wherein the sum of sequential window functions is unity.

### Claim 24 (depends on 18)

The system of claim 18, wherein the combiner is configured to select windowed data packets based at least in part on an accuracy of fit parameter.

### Claim 25 (depends on 24)

The system of claim 24, wherein the accuracy of fit parameter incorporates a measure of the sum of least squares between at least one data buffer and at least one translated data buffer.

### Claim 26 (depends on 18)

The system of claim 18 further comprising one or more additional stages and analog-to-digital converters, wherein each data buffer other than that derived from the digital input signal created by the analog-to-digital converter having the lowest level of analog input amplification is translated to match the data buffer derived from the digital input signal created by the analog-to-digital converter having the lowest level of analog input amplification.

### Claim 27 (depends on 25)

The system of claim 25, wherein the combiner selects window data packets that correspond to those derived from the digital input signal whose analog-to-digital converter has the highest level of input amplification in deriving its analog input signal and whose translated buffers are within a prescribed tolerance when compared to the data buffer derived from the digital input signal from the analog-to-digital converter having the lowest level of amplification.

### Claim 28 (depends on 23)

The system of claim 23, wherein each data buffer other than that derived from the digital input signal from the analog-to-digital converter having the highest level of quantization noise is translated to match the data buffer derived from the digital input signal from the analog-to-digital converter having the lowest level of quantization.

### Claim 29 (depends on 18)

The system according to claim 18, wherein the multi-stage analog-to-digital conversion system is part of an apparatus including a multi-track audio recorder and each digital output signal is provided as an input to the multi-track audio recorder.

### Claim 30 (depends on 18)

The system according to claim 18, wherein the multi-stage analog-to-digital conversion system is part of an apparatus including a multi-track audio mixer and each digital output signal is provided as an input to a multi-track audio mixer.

### Claim 31 (depends on 18)

The system according to claim 18 further comprising a wireless microphone apparatus, wherein the multi-stage analog-to-digital conversion system is part of said apparatus.
