Analog-to-digital converter with a hybrid structure of segmented linear quantization and noise shaping function

Abstract

The application discloses a segmented linear quantization hybrid structure analog-to-digital converter with noise shaping function, and belongs to the field of analog-to-digital converters.The analog-to-digital converter comprises an input control circuit, a nonlinear quantization circuit, a noise shaping modulation circuit, an output control circuit and a digital decimation filter circuit.The input control circuit controls the sampling of an input signal.The nonlinear quantization circuit determines the nonlinear paragraph where the input signal is located, and provides a reference voltage for the noise shaping modulation circuit.The noise shaping modulation circuit is used for performing noise shaping and fine quantization on the input signal.The output control circuit combines two parts of the quantization output and generates a high-speed low-bit code stream.The digital decimation filter circuit receives the high-speed code stream, and generates a high-precision digital code through decimation filtering.The technology disclosed by the application can provide appropriate nonlinear quantization characteristics according to different signal characteristics, and can realize higher quantization precision and a wider dynamic range.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of analog-to-digital converters, in particular to a segmented linear quantization hybrid structure analog-to-digital converter with noise shaping function. BACKGROUND

[0002] Analog-to-digital converters (ADCs) convert analog signals with continuous time and continuous values into discrete digital signals. The structure of an analog-to-digital converter is mainly divided into full parallel, successive approximation, pipeline, and Sigma-Delta types, which are suitable for different application scenarios and accuracy requirements. Successive approximation analog-to-digital converters use the "bisection method" principle to achieve quantization, and have a simple structure, but it is difficult to achieve high accuracy. Sigma-Delta analog-to-digital converters are composed of a Sigma-Delta modulator and a digital decimation filter, and use oversampling and noise shaping techniques to achieve high accuracy. Zoom and noise shaping successive approximation hybrid structure analog-to-digital converters combine successive approximation analog-to-digital converters with Sigma-Delta analog-to-digital converters, and have the characteristics of low power consumption and high accuracy.

[0003] Traditional analog-to-digital converters mostly use uniform quantization, and the quantization intervals are uniformly distributed within the full-scale range. The closer the signal is to the full-scale value, the higher the signal-to-noise ratio, and the smaller the signal-to-noise ratio of the small signal. For application scenarios such as audio and image signals with large dynamic range, non-linear quantization analog-to-digital converters such as segmented linear quantization can use fine quantization intervals for small signals of interest to improve resolution and signal-to-noise ratio, and use coarse quantization intervals for large signals to accommodate higher amplitudes. Thus, under the premise of constant quantization bits, both weak signals are accurately captured and large signals are prevented from overflowing, ultimately achieving a higher dynamic range than linear analog-to-digital converters.

[0004] Successive approximation analog-to-digital converters have a simple structure, but their accuracy is limited and their dynamic range is low. Although oversampling analog-to-digital converters can achieve high accuracy, they have a complex structure and it is difficult to directly implement non-linear quantization. The dynamic range is limited by the uniform quantization characteristic. Hybrid structure analog-to-digital converters that combine different structures are expected to achieve higher dynamic range through non-linear quantization, but current research on hybrid structure analog-to-digital converters rarely involves non-linear quantization. Existing analog-to-digital converters lack a technology that can achieve higher dynamic range. SUMMARY

[0005] The present application aims at overcoming the deficiencies of the prior art, and provides a segmented linear quantization hybrid structure analog-to-digital converter with noise shaping function.

[0006] To achieve the above-mentioned purpose, the present application provides a segmented linear quantization hybrid structure analog-to-digital converter with noise shaping function, comprising: an input control circuit, a nonlinear quantization circuit, a noise shaping modulation circuit, an output control circuit and a digital decimation filter circuit.

[0007] The input control circuit is connected with the nonlinear quantization circuit and the noise shaping modulation circuit, and is used for receiving an original input signal, performing amplitude conversion, and controlling the nonlinear quantization circuit and the noise shaping modulation circuit to receive the input signal; the nonlinear quantization circuit is connected with the noise shaping modulation circuit, receives the input signal, and determines the nonlinear paragraph where the input signal is located, so as to provide a reference signal for the noise shaping modulation circuit; the noise shaping modulation circuit performs oversampling on the input signal, and performs noise shaping and uniform fine quantization on the signal residual error after nonlinear quantization according to the reference signal from the nonlinear quantization circuit, modulates the noise in the signal band to the outside of the signal band, and outputs an intra-segment code; the output control circuit is connected with the nonlinear quantization circuit and the noise shaping modulation circuit, receives the outputs of the nonlinear quantization circuit and the noise shaping modulation circuit, stores the digital code in the quantization process, and performs buffering, processing and outputting; and the digital decimation filter circuit is connected with the output control circuit, receives the high-speed digital code output by the output control circuit, filters out the high-frequency noise outside the signal band generated by noise shaping, and converts to generate a high-precision digital code.

[0008] As a further improvement of the present application, the nonlinear quantization circuit uses a successive approximation structure for coarse quantization, comprising:

[0009] The nonlinear digital-to-analog converter uses a nonlinear weight capacitor array to realize the nonlinear transfer characteristic of the circuit, controls the nonlinear capacitor array by inputting a thermometer code to realize charge scaling, and generates a reference voltage, comprising two digital-to-analog conversion units, respectively used for generating a start reference signal of the interval of the nonlinear paragraph where the signal is located And an end reference signal ;

[0010] The comparator unit is used for connecting the system input signal sampled by the input control circuit and the signal generated by the digital-to-analog converter in the nonlinear quantization circuit, and comparing the two signals;

[0011] The successive approximation register uses a logic circuit to realize a successive approximation algorithm, receives the output result of the comparator, and controls the quantization circuit to complete the successive approximation type quantization.

[0012] a decoder, connected to the successive approximation register and the non-linear quantization circuit, for converting the output binary code of the non-linear quantization circuit into a thermometer code.

[0013] As a further improvement of the present application, the noise shaping modulation circuit comprises:

[0014] a filter, for converting the quantization residual of the non-linear quantization circuit into an integral signal, and inputting the integral signal into a quantizer for quantization;

[0015] a quantizer, for quantizing the integral signal generated by the filter by one or more bits, and outputting a high-speed digital code as a partial quantization result;

[0016] a logic circuit, connected to the output control circuit and the digital-to-analog converter, for controlling the digital-to-analog converter to output a feedback signal;

[0017] a digital-to-analog converter, connected to the non-linear quantization circuit, for receiving a reference signal generated by the non-linear digital-to-analog converter and generating a feedback signal according to the reference signal under the control of the logic circuit.

[0018] As a further improvement of the present application, the noise shaping modulation circuit can use a Sigma-Delta structure or a noise shaping successive approximation structure; when the noise shaping modulation circuit is implemented in the noise shaping successive approximation structure, the quantizer is a one-bit quantizer, i.e. a comparator, the logic circuit is implemented as a successive approximation register, and the quantizer and the logic circuit can be shared with the non-linear quantization circuit; at this time, the first M bits of the output of the successive approximation register are a segment code, which is used to control the non-linear quantization circuit, and the last N bits are an intra-segment code, which is used to control the noise shaping modulation circuit.

[0019] As a further improvement of the present application, the output control circuit comprises:

[0020] a segment code buffer unit, for buffering and outputting the quantization result of the non-linear quantization circuit in the quantization process;

[0021] an intra-segment code buffer unit, for buffering and outputting the quantization result of the noise shaping modulation circuit in the quantization process;

[0022] an operation logic unit, for integrating the quantization results of the non-linear quantization circuit and the noise shaping modulation circuit to generate a complete quantization result;

[0023] an inverse non-linear transformation unit, for restoring the quantization result output by the operation logic unit into a linear proportional binary digital code corresponding to the original value of the signal according to a lookup table.

[0024] Compared with the prior art, the application has the following advantages:

[0025] The application provides a segmented linear quantization hybrid structure analog-to-digital converter with a noise shaping function.

[0026] Compared with other technologies, the application combines a successive approximation structure and a Sigma-Delta structure, can realize a high-precision nonlinear analog-to-digital converter, and realizes expansion of a dynamic range in the same precision analog-to-digital converter. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 is a block diagram of the segmented linear quantization hybrid structure analog-to-digital converter with the noise shaping function of the application;

[0028] Figure 2 is a block diagram of the segmented linear quantization hybrid structure analog-to-digital converter with the noise shaping function in the specific embodiment of the application;

[0029] Figure 3 is a structure principle diagram of the segmented linear quantization hybrid structure analog-to-digital converter with the noise shaping function in the embodiment 1 of the application;

[0030] Figure 4 is a structure principle diagram of the segmented linear quantization hybrid structure analog-to-digital converter with the noise shaping function in the embodiment 2 of the application;

[0031] Figure 5 is a comparison diagram of signal-to-noise ratio characteristics of the embodiment 1 and the comparative example 1 of the application. DETAILED DESCRIPTION

[0032] The application will be further described in the specific embodiments in combination with the drawings, but the scope of the application is not limited in any way.

[0033] Please refer to Figure 1 and Figure 2 , wherein Figure 1 is a block diagram of the segmented linear quantization hybrid structure analog-to-digital converter with the noise shaping function of the application, Figure 2 is a block diagram of the segmented linear quantization hybrid structure analog-to-digital converter with the noise shaping function in the specific embodiment of the application.

[0034] The core of the application is to propose a segmented linear quantization hybrid structure analog-to-digital converter with a noise shaping function, please refer to Figure 1The mixed structure analog-to-digital converter with noise shaping function comprises an input control circuit 100, a nonlinear quantization circuit 200, a noise shaping modulation circuit 300, an output control circuit 400 and a digital decimation filter circuit 500, wherein the input control circuit 100 is connected to the nonlinear quantization circuit 200 and the noise shaping modulation circuit 300, the nonlinear quantization circuit 200 and the noise shaping modulation circuit 300 are connected, the output control circuit 400 is connected to the nonlinear quantization circuit 200 and the noise shaping modulation circuit 300, and the digital decimation filter circuit 500 is connected to the output control circuit 400.

[0035] The input control circuit 100 is configured to receive an original input signal, perform amplitude conversion, and control the nonlinear quantization circuit 200 and the noise shaping modulation circuit 300 to receive the input signal.

[0036] The nonlinear quantization circuit 200 is configured to receive the input signal, determine the nonlinear segment where the input signal is located, provide a reference signal for the noise shaping modulation circuit 300, and output a segment code of the segment where the input signal is located.

[0037] The noise shaping modulation circuit 300 is configured to oversample the input signal, perform noise shaping and uniform fine quantization on a signal residual error according to the reference signal from the nonlinear quantization circuit 200, modulate noise in a signal band to outside the signal band, and output an intra-segment code.

[0038] The output control circuit 400 is configured to receive output results of the nonlinear quantization circuit 200 and the noise shaping modulation circuit 300, buffer digital codes in a quantization process, integrate a segment code output by the nonlinear quantization circuit 200 and an intra-segment code output by the noise shaping modulation circuit 300, and convert a complete output result into a digital code corresponding to a linear interval.

[0039] The digital decimation filter circuit 500 is configured to receive the digital code output by the output control circuit, filter high-frequency noise generated by noise shaping outside a signal band, and convert to generate a high-precision digital code.

[0040] In this embodiment, the input control circuit 100 receives the input signal, performs amplitude conversion, and generates a clock signal to control the nonlinear quantization circuit 200 and the noise shaping and modulation circuit 300 to sample the input signal; the nonlinear quantization circuit 200 consists of a nonlinear digital-to-analog converter 201, a comparator unit 202, a successive approximation register 203, and a decoder 204; the noise shaping and modulation circuit 300 consists of a filter 301, a quantizer 302, a logic circuit 303, and a digital-to-analog converter 304; the output control circuit 400 includes a segment code buffer unit 401, an intra-segment code buffer unit 402, an arithmetic logic unit 403, and an inverse nonlinear transformation unit 404.

[0041] The nonlinear quantization circuit 200 performs nonlinear successive approximation quantization on the input signal under the control of the successive approximation register 203. The nonlinear transfer characteristics of the nonlinear quantization circuit 200 are determined by the nonlinear weighted capacitor array in the nonlinear digital-to-analog converter 201. The nonlinear digital-to-analog converter 201 includes two digital-to-analog conversion units, which are used to generate the start-point reference signal and the end-point reference signal of the nonlinear segment interval, respectively. In the first conversion cycle, the nonlinear digital-to-analog converter 201 generates an estimated start-point reference signal under the control of the successive approximation register 203. The comparator unit 202 determines the range of the input signal by comparing the magnitude of the start-point reference signal generated by the nonlinear digital-to-analog converter 201 with the magnitude of the input signal, and feeds back the comparison result to the successive approximation register. The binary code output by the successive approximation register 203 is converted into thermometer code by the decoder 204, which then controls the nonlinear digital-to-analog converter 201 to further approximate the input signal. When the segment code determined by the successive approximation register 203 has M bits, the nonlinear quantization circuit 200 needs M+1 cycles to determine the nonlinear segment interval where the input signal is located. The successive approximation register 203 outputs the determined segment code to the segment code buffer unit 401. The nonlinear digital-to-analog converter 201 generates the start reference signal and the end reference signal of the nonlinear segment interval where the signal is located, and connects them to the digital-to-analog converter 304. The digital-to-analog converter 304 samples and holds the start reference signal and the end reference signal under clock control, thereby performing fine quantization.

[0042] Further, the noise shaping modulation circuit 300 oversamples the input signal under the control of a clock and obtains a signal residual according to a reference signal generated by the nonlinear digital-to-analog converter 201; the quantization bit number of the noise shaping modulation circuit 300 is fixed, the quantization intervals are uniformly distributed within the signal located nonlinear paragraph interval, and the size of the quantization interval is determined by the start reference signal and the end reference signal generated by the nonlinear digital-to-analog converter 201; the signal residual is converted into an integral signal after input to the filter 301, the integral signal is input to the quantizer 302 for one-bit or multi-bit quantization, the digital code obtained by quantization of the quantizer 302 is output to the paragraph code buffer unit 402 and the logic circuit 303, and the logic circuit 303 controls the digital-to-analog converter 304 to output a feedback signal according to the quantization result of the quantizer 302, so as to further quantize; the operation logic unit 403 integrates the quantization results of the nonlinear quantization circuit 200 and the noise shaping modulation circuit 300 to generate a complete quantization result, and inputs the complete quantization result to the inverse nonlinear transformation unit 404, the inverse nonlinear transformation unit 404 restores the complete quantization result to a linear proportional binary digital code corresponding to the original value of the signal according to a lookup table, and inputs the linear proportional binary digital code to the digital decimation filter circuit 500, so as to filter out the high-frequency noise generated by the signal band noise shaping and convert to generate a high-precision digital code.

[0043] In some specific embodiments, the noise shaping modulation circuit 300 uses a noise shaping successive approximation structure; the quantizer 302 is a one-bit quantizer, i.e., a comparator, the logic circuit 303 is implemented as a successive approximation register, and the quantizer and the logic circuit can be shared with the nonlinear quantization circuit, at this time, the first M bits output by the successive approximation register are paragraph codes, which are also used to control the nonlinear quantization circuit, and the last N bits are paragraph codes, which are used to control the noise shaping modulation circuit; at this time, the circuit is divided into three quantization stages, which are:

[0044] The first stage is nonlinear successive approximation, which is implemented by the nonlinear quantization circuit 200;

[0045] The second stage is linear successive approximation, which is implemented by the noise shaping modulation circuit 300;

[0046] The third stage is noise shaping, which is implemented by the noise shaping modulation circuit 300.

[0047] Embodiment 1

[0048] Please refer to Figure 3In the embodiment, a segmented linear quantization hybrid structure analog-digital converter with noise shaping function and A-law 13-segment line transfer characteristic is provided. The noise shaping modulation circuit 300 uses Sigma-Delta structure, which is composed of a filter, a quantizer, a logic circuit (scaling logic) and a digital-analog converter (DAC). The nonlinear quantization circuit 200 is composed of a nonlinear digital-analog converter (NLDAC), a comparator, a successive approximation register and a decoder. The nonlinear quantization circuit 200 and the noise shaping modulation circuit 300 work in parallel, and the signal bandwidth is . The oversampling rate of the circuit is OSR, and the sampling frequency of the nonlinear quantization circuit 200 and the noise shaping modulation circuit 300 is . The nonlinear quantization circuit 200 divides the input signal into different nonlinear segment intervals, generates the start reference signal and the end reference signal of the nonlinear segment interval where the signal is located. The noise shaping modulation circuit 300 linearly quantizes the signal in the segment according to the nonlinear reference voltage provided by the nonlinear quantization circuit 200. In order to prevent the input signal from changing beyond the range of the nonlinear quantization circuit 200, the expansion coefficient M is set, and the scaling logic controls the DAC to output the feedback signal according to the output result LSB of the quantizer:

[0049]

[0050] Therefore, the signal at the input end of the filter is the difference between the input signal and the feedback signal, and the actual modulation object is the quantization residual of the nonlinear quantization circuit.

[0051] Embodiment 2

[0052] Please refer to Figure 4The embodiment provides a noise shaping modulation circuit 300 using a noise shaping successive approximation structure with a segmented linear quantization hybrid structure analog-digital converter with a noise shaping function, the noise shaping modulation circuit 300 and the nonlinear quantization circuit 200 share a quantizer (comparator) and a logic circuit (successive approximation register); the nonlinear digital-analog converter 201 in the nonlinear quantization circuit 200 corresponds to the NLDAC module in the figure, the comparator unit 202 corresponds to the comparator module in the figure, the successive approximation register 203 corresponds to the successive approximation register module in the figure, and the decoder 204 corresponds to the decoder module in the figure; the filter 301 in the noise shaping modulation circuit 300 corresponds to the filter module in the figure, the quantizer 302 corresponds to the comparator module in the figure, the logic circuit 303 corresponds to the successive approximation register module in the figure, and the digital-analog converter 304 corresponds to the DAC module in the figure; at this time, the first M bits of the successive approximation register output are paragraph codes, which are used to control the nonlinear quantization circuit, and the last N bits are intra-paragraph codes, which are used to control the noise shaping modulation circuit; at the same time, since the successive approximation register can automatically combine the outputs of the noise shaping modulation circuit 300 and the nonlinear quantization circuit 200, the function of the output control circuit 400 is replaced by the successive approximation register; at this time, the circuit is divided into three quantization stages, which are:

[0053] The first stage is nonlinear successive approximation, at this time, the switch is closed, the switch is opened, the noise shaping modulation circuit does not work, the first M bits of the output of the successive approximation register are paragraph codes, the decoder converts the paragraph codes into thermometer codes and controls the NLDAC to complete the output, the comparator compares the output of the NLDAC with the input signal and inputs the result into the successive approximation register, and the nonlinear successive approximation is completed after M+1 cycles;

[0054] The second stage is linear successive approximation, at this time, the switch , is opened, at this time, the nonlinear quantization circuit 200 does not work, the output result of the NLDAC is kept and provides a reference signal for the DAC; the last N bits of the output of the successive approximation register are intra-paragraph codes, and the DAC is controlled to complete the output, the comparator compares the output of the DAC with the input signal and inputs the result into the successive approximation register, and the nonlinear successive approximation is completed after N cycles;

[0055] The third stage is noise shaping, at this time, the switch is closed, the switch When disconnected, the nonlinear quantization circuit 200 does not operate, the output of the NLDAC is maintained, and a reference signal is provided for the DAC; the successive approximation register controls the DAC to generate a residual signal, and through... The residual signal is input to the filter, processed by the filter, added to the input signal, and then fed back to the comparator to determine the last bit of digital output, thereby achieving noise shaping.

[0056] Comparative Example 1

[0057] This comparative example replaces the nonlinear quantization circuit 200 in the piecewise linear quantization hybrid analog-to-digital converter with noise shaping function that realizes the A-law 13-fold line transfer characteristic described in Example 1 with a linear successive final analog-to-digital converter with the same number of bits, and evaluates the performance of the two analog-to-digital converters.

[0058] To further illustrate the effect of the piecewise linear quantization hybrid analog-to-digital converter with noise shaping function provided by the present invention, Figure 5 This is a comparison chart of the signal-to-noise ratio characteristic curves of Embodiment 1 and Comparative Example 1 of the present invention; as shown Figure 5 As shown in the figure, the horizontal axis represents the normalized amplitude of the input signal in dB, and the vertical axis represents the signal-to-noise ratio (SNR) at the corresponding amplitude in dB. The red line in the figure corresponds to a piecewise linear quantization hybrid analog-to-digital converter with noise shaping function as described in Example 1, and the black line in the figure corresponds to the linear quantization analog-to-digital converter with the same number of bits used for comparison in Comparative Example 1. It can be seen that, under the premise of the same number of quantization bits, the piecewise linear quantization hybrid analog-to-digital converter with noise shaping function provided by this invention, compared with the linear quantization analog-to-digital converter, expands the dynamic range with an SNR greater than 100 dB from 25 dB to 45 dB, and the SNR improvement value for small signals is 20 dB.

[0059] This invention can be applied to sensor fields such as audio sensors, image sensors, humidity and temperature sensors, where the signals of interest are not uniformly distributed and high precision and high dynamic range are required.

[0060] The above-described embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that implementing all or part of the processes of the above embodiments and making equivalent changes in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A piecewise linear quantization hybrid analog-to-digital converter with noise shaping function, characterized in that: It includes an input control circuit, a nonlinear quantization circuit, a noise shaping and modulation circuit, an output control circuit, and a digital decimation and filtering circuit. The input control circuit is connected to the nonlinear quantization circuit and the noise shaping and modulation circuit. The nonlinear quantization circuit and the noise shaping and modulation circuit are connected. The output control circuit is connected to the nonlinear quantization circuit and the noise shaping and modulation circuit. The digital decimation and filtering circuit is connected to the output control circuit. The input control circuit is used to receive the original input signal, perform amplitude conversion, and control the nonlinear quantization circuit and noise shaping modulation circuit to receive the input signal. The nonlinear quantization circuit uses a successive approximation structure for coarse quantization and consists of a nonlinear digital-to-analog converter, a comparator unit, a successive approximation register, and a decoder. It receives the input signal and determines the nonlinear segment in which it is located, thereby providing a reference signal for the noise shaping and modulation circuit. The nonlinear digital-to-analog converter uses a nonlinear weighted capacitor array to realize the nonlinear transfer characteristics of the circuit, and controls the nonlinear capacitor array to achieve charge scaling by inputting thermometer codes to generate a reference voltage; The comparator unit is used to connect the system input signal sampled by the input control circuit and the signal generated by the digital-to-analog converter in the nonlinear quantization circuit, and to compare the two. The successive approximation register uses logic circuits to implement the successive approximation algorithm, receives the output of the comparator, and controls the quantization circuit to complete the successive approximation quantization. The decoder is used to connect the successive approximation register and the nonlinear quantization circuit, and converts the output binary code of the successive approximation register used to control the nonlinear quantization circuit into thermometer code. The nonlinear quantization circuit performs nonlinear successive approximation quantization on the input signal under the control of the successive approximation register. The nonlinear transfer characteristics of the nonlinear quantization circuit are determined by the nonlinear weighted capacitor array in the nonlinear digital-to-analog converter. The nonlinear digital-to-analog converter includes two digital-to-analog conversion units, which are used to generate the start-point reference signal and the end-point reference signal of the nonlinear segment interval, respectively. During the first conversion cycle, the nonlinear digital-to-analog converter generates a guessed starting reference signal under the control of the successive approximation register. The comparator unit determines the range of the input signal by comparing the magnitude of the starting reference signal generated by the nonlinear digital-to-analog converter with the magnitude of the input signal, and feeds back the comparison result to the successive approximation register. The binary code output by the successive approximation register is converted into thermometer code by the decoder, thereby controlling the nonlinear digital-to-analog converter to further approximate the input signal. The noise shaping and modulation circuit oversamples the input signal and performs noise shaping and further quantization on the signal residual after nonlinear quantization based on the reference signal from the nonlinear quantization circuit, modulating the noise in the signal band to outside the signal band and outputting the segment code. The output control circuit receives the output results of the nonlinear quantization circuit and the noise shaping and modulation circuit, stores the digital code in the quantization process, and performs buffering, processing and output. The digital decimation filter circuit receives the high-speed digital code output by the output control circuit, filters out high-frequency noise outside the signal frequency band generated by noise shaping, and converts it into a high-precision digital code.

2. The piecewise linear quantization hybrid analog-to-digital converter with noise shaping function as described in claim 1, characterized in that: The noise shaping and modulation circuit uses a Sigma-Delta structure to achieve noise shaping. The Sigma-Delta structure includes first-order, multi-order, and multi-stage cascaded structures, and consists of filters, quantizers, logic circuits, and digital-to-analog converters. The filter is used to convert the quantization residual of the nonlinear quantization circuit into an integral signal, and input the integral signal to the quantizer for quantization. The quantizer performs one or more bit quantization on the integral signal generated by the filter and outputs a high-speed digital code as a partial quantization result. The logic circuit is connected to the output control circuit and the digital-to-analog converter, and controls the digital-to-analog converter to output a corresponding feedback signal based on the quantization result input to the output control circuit.

3. The piecewise linear quantization hybrid analog-to-digital converter with noise shaping function as described in claim 2, characterized in that: The digital-to-analog converter is connected to the nonlinear digital-to-analog converter, receives the reference signal generated by the nonlinear digital-to-analog converter, and generates a feedback signal based on the reference signal under the control of the logic circuit.

4. The piecewise linear quantization hybrid analog-to-digital converter with noise shaping function as described in claim 3, characterized in that: The noise shaping and modulation circuit and the nonlinear quantization circuit operate in parallel.

5. The piecewise linear quantization hybrid analog-to-digital converter with noise shaping function as described in claim 1, characterized in that: The noise shaping and modulation circuit uses a modulator with a noise shaping successive approximation structure, which includes a comparator unit, a successive approximation register, a filter, and a digital-to-analog converter; The comparator unit and the successive approximation register are shared with the nonlinear quantization circuit. The first M bits of the output of the successive approximation register are segment codes, which are used to control the nonlinear quantization circuit, and the last N bits are intra-segment codes, which are used to control the noise shaping and modulation circuit. The digital-to-analog converter is connected to the nonlinear digital-to-analog converter, receives the reference signal generated by the nonlinear digital-to-analog converter, and generates a feedback signal based on the reference signal under the control of the successive approximation register.

6. The piecewise linear quantization hybrid analog-to-digital converter with noise shaping function as described in claim 5, characterized in that: If the noise shaping modulation circuit uses a noise shaping successive approximation structure, the piecewise linear quantization hybrid analog-to-digital converter with noise shaping function is divided into three quantization stages: In the first stage, nonlinear successive approximation is achieved through the nonlinear quantization circuit. The second stage, linear successive approximation, is achieved through the noise shaping modulation circuit of the noise shaping successive approximation structure. The third stage, noise shaping, is achieved through the noise shaping modulation circuit of the noise shaping successive approximation structure.

7. The piecewise linear quantization hybrid analog-to-digital converter with noise shaping function as described in claim 1, characterized in that: The output control circuit includes a segment code buffer unit, an intra-segment code buffer unit, an arithmetic logic unit, and an inverse nonlinear transformation unit. The segment code buffer unit is used to buffer and output the quantization result of the nonlinear quantization circuit during the quantization process; The intra-segment code buffer unit is used to buffer and output the quantization result of the noise shaping modulation circuit during the quantization process; The arithmetic logic unit is used to integrate the quantization results of the nonlinear quantization circuit and the noise shaping and modulation circuit to generate a complete quantization result; The inverse nonlinear transformation unit restores the quantization result output by the arithmetic logic unit to the linear proportional binary code corresponding to the original signal value according to the lookup table.

Images