A background channel mismatch calibration method and system

Through the background channel mismatch calibration method, the time interleaving analog-to-digital converter and digital signal processing circuit are used to adjust the mismatch error in the SerDes system in real time, solving the performance degradation caused by the mismatch effect in high-speed communication, and achieving the correct Real-time tracking and calibration of residual mismatch errors improves system stability and adaptability.

CN118890048BActive Publication Date: 2024-12-06CHENGDU NOVACORE MICROELECTRONICS TECHNOLOGY CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411368259.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-29
Publication Date
2024-12-06
Estimated Expiration
2044-09-29

AI Technical Summary

Technical Problem

In high-speed communication, the existing SerDes system has a performance degradation due to the mismatch effect caused by the manufacturing process and sampling time deviation of the time interleaved analog-to-digital converter, and the front-end calibration scheme cannot track and calibrate the residual mismatch error changes in real time, affecting the system stability. sex.

Method used

Using the backend channel mismatch calibration method, the analog-to-digital converter, digital signal processing circuit and filter are used to interleave the analog-to-digital converter, and iteratively adjust the operating parameters of the sub-analog-to-digital converter through time interleaving analog-to-digital converter, and real-time estimation and compensation. Gain mismatch and timing mismatch error ensure the registration of digital signals.

Benefits of technology

Without interrupting the normal operation of the SerDes system, real-time tracking and calibration of residual mismatch errors and their changes is achieved, system performance and stability are improved, and adaptability to process, voltage and temperature changes are enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118890048B_ABST
    Figure CN118890048B_ABST
Patent Text Reader

Abstract

The present invention provides a background channel mismatch calibration method and system, and relates to the field of mismatch calibration technology. First, a sub-signal processing module is pre-calibrated, and then the calibrated sub-signal processing module is used to determine the decision parameters at different sampling times under the corresponding data transmission channel. Subsequently, a filter is used to determine the residual mismatch error estimation function according to the decision parameters at adjacent sampling times, and finally the residual mismatch error estimation function is used to iteratively adjust the corresponding sub-analog-to-digital converter's own operating parameters until the mismatch compensation meets the preset value, and the aligned digital signal is obtained. The present invention uses the calibrated sub-signal processing module to determine the decision parameters of the data transmission channel at different sampling times, and then adjusts the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal, which can track and calibrate the residual mismatch error and the residual mismatch error change in real time without interrupting the normal operation of the SerDes system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of mismatch calibration, and in particular to a background channel mismatch calibration method and system. Background Art

[0002] The Serializer and Deserializer (SerDes) system is a common high-speed communication system in electronic devices, which is widely used in computer networks, communication systems, storage devices, graphic displays and other fields. Since most traditional SerDes systems use analog architectures, which often depend on process, voltage and temperature, it is difficult to expand as the communication rate requirements increase. Based on this, the digital architecture of the digital signal processor (Digital Signal Processing, DSP) based on the analog-to-digital converter (Analog-to-Digital Converter, ADC) is widely used in the prior art to increase the communication rate.

[0003] In order to achieve high-speed analog-to-digital conversion under the premise of low power consumption, a time-interleaved analog-to-digital converter (TI-ADC) is often used as the ADC implementation structure in a high-speed SerDes system. Among them, the time-interleaved analog-to-digital converter includes multiple sub-analog-to-digital converters (sub ADCs), each of which corresponds to a data transmission channel. Each data transmission channel constitutes a time-interleaved structure, and each data transmission channel is sampled alternately and in parallel according to a multi-phase timing sequence, thereby realizing high-speed sampling of the SerDes system under controllable power consumption. However, due to interference such as the manufacturing process of the time-interleaved analog-to-digital converter and the sampling time deviation between each data transmission channel, each data transmission channel will produce a series of mismatch effects, resulting in a decrease in the performance of the entire SerDes system. The above-mentioned mismatch effects include gain mismatch errors, timing mismatch errors, etc.

[0004] To avoid the above mismatch effect, a calibration algorithm is used to adjust each data transmission channel before the time-interleaved analog-to-digital converter actually works. The above calibration algorithm includes a foreground calibration scheme and a background calibration scheme. Based on the consideration of the robustness of the calibration algorithm, the industry mostly adopts the foreground calibration scheme. Among them, the foreground calibration scheme needs to inject a known reference signal in advance, and then calibrate the mismatch error between different data transmission channels based on the known reference signal. The above calibration process will interrupt the normal operation of the SerDes system. For example, when the SerDes system executes the foreground calibration scheme, it will first suspend the conversion of the analog signal.

[0005] When the SerDes system is operating normally, the mismatch error may change due to factors such as process, voltage, and temperature, but the foreground calibration solution cannot perform residual calibration in real time after the residual mismatch error changes. Moreover, when the residual mismatch error changes, it is necessary to re-inject a known reference signal to perform secondary calibration on the mismatch error between different data transmission channels.

[0006] In summary, a calibration solution is urgently needed, which can track and calibrate the residual mismatch error and the residual mismatch error change in real time without interrupting the normal operation of the SerDes system. Summary of the invention

[0007] The present invention provides a background channel mismatch calibration method and system, which can track and calibrate the residual mismatch error and the residual mismatch error change in real time without interrupting the normal operation of the SerDes system.

[0008] The embodiments of the present invention can be implemented as follows:

[0009] In a first aspect, the present invention provides a background channel mismatch calibration method, which is applied to a background channel mismatch calibration system. The background channel mismatch calibration system includes a time-interleaved analog-to-digital converter, a digital signal processing circuit and a filter. The digital signal processing circuit is respectively connected to the time-interleaved analog-to-digital converter and the filter, and the filter is also connected to the time-interleaved analog-to-digital converter; the time-interleaved analog-to-digital converter includes a plurality of sub-analog-to-digital converters; the digital signal processing circuit includes a plurality of sub-signal processing modules; each sub-analog-to-digital converter is connected to a sub-signal processing module to form a plurality of data transmission channels. The method includes:

[0010] For each data transmission channel, a sub-analog-to-digital converter is used to convert the input analog signal into a digital signal;

[0011] Sending a preset mismatch calibration signal to the sub-signal processing module, wherein the sub-signal processing module performs calibration according to the preset mismatch calibration signal;

[0012] Using the calibrated sub-signal processing modules to obtain the decision parameters of each data transmission channel at different sampling times based on the digital signal;

[0013] The residual mismatch error estimation function of each data transmission channel is determined by using a filter according to the decision parameter at the current sampling time and the decision parameter at the previous sampling time; the residual mismatch error estimation function includes a gain mismatch error estimation function and a timing mismatch error estimation function;

[0014] The timing mismatch error estimation function and the gain mismatch error estimation function are used to iteratively adjust the operating parameters of the corresponding sub-ADCs in sequence until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal meet the preset values, thereby obtaining the aligned digital signal.

[0015] In one possible implementation, the digital signal after registration is expressed as:

[0016] ;

[0017] ;

[0018] ;

[0019] in, For the k Data transmission channels in The digital signal after registration at the moment; For the k Gain mismatch error of each data transmission channel; For the k Data transmission channels in Gain mismatch error estimation function at time ; is the channel impulse response function; is the sampling period; For the k The timing mismatch error of each data transmission channel; For the k Data transmission channels in The timing mismatch error estimation function at time instant; For the k Data transmission channels in Additive noise at the moment; For the k Data transmission channels in The time corresponds to modulation symbols under symbol numbers; is the total number of data transmission channels.

[0020] In a second aspect, the present invention further provides a background channel mismatch calibration system, which includes a time-interleaved analog-to-digital converter, a digital signal processing circuit and a filter. The digital signal processing circuit is connected to the time-interleaved analog-to-digital converter and the filter respectively, and the filter is also connected to the time-interleaved analog-to-digital converter; the time-interleaved analog-to-digital converter includes a plurality of sub-analog-to-digital converters; the digital signal processing circuit includes a plurality of sub-signal processing modules; each sub-analog-to-digital converter is connected to a sub-signal processing module to form a plurality of data transmission channels, wherein:

[0021] For each data transmission channel, a sub-analog-to-digital converter is used to convert the input analog signal into a digital signal;

[0022] A sub-signal processing module, used for performing calibration using a preset mismatch calibration signal, and obtaining decision parameters of each data transmission channel at different sampling times based on the digital signal;

[0023] A filter, used to determine a residual mismatch error estimation function of each data transmission channel according to a decision parameter at a current sampling time and a decision parameter at a previous sampling time; the residual mismatch error estimation function includes a gain mismatch error estimation function and a timing mismatch error estimation function;

[0024] The sub-ADC is also used to iteratively adjust its own operating parameters according to the timing mismatch error and the gain mismatch error in sequence until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal meet the preset values ​​to obtain the aligned digital signal.

[0025] The beneficial effects of the present invention include, for example:

[0026] The present invention provides a background channel mismatch calibration method, which is applied to a background channel mismatch calibration system. The background channel mismatch calibration system includes a time-interleaved analog-to-digital converter, a digital signal processing circuit and a filter. Each sub-analog-to-digital converter under the time-interleaved analog-to-digital converter corresponds to and is connected with each sub-signal processing module under the digital signal processing circuit to form a plurality of data transmission channels. For each data transmission channel, the input analog signal is converted into a digital signal by using the sub-analog-to-digital converter. Then, a preset mismatch calibration signal is sent to the sub-signal processing module, and the calibrated sub-signal processing modules are obtained and used to determine the decision parameters of each data transmission channel at different sampling times according to the digital signal. Subsequently, the filter is used to determine the residual mismatch error estimation function of each data transmission channel according to the decision parameters at the current sampling time and the decision parameters at the previous sampling time; wherein the residual mismatch error estimation function includes a gain mismatch error estimation function and a timing mismatch error estimation function. Each sub-analog-to-digital converter is used to iteratively adjust its own operating parameters according to the timing mismatch error and the gain mismatch error, until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal meet the preset value, and the aligned digital signal is obtained.

[0027] The present invention first calibrates the sub-signal processing module using a preset mismatch calibration signal, and then uses each calibrated sub-signal processing module to determine the decision parameters of each data transmission channel at different sampling times according to the digital signal, thereby determining the residual mismatch error (i.e., decision error) and the decision result under the corresponding data transmission channel. At the same time, the residual mismatch error change caused by the influence of process, voltage, temperature, etc. can be determined based on the decision result and the residual mismatch error at different sampling times, and then the residual mismatch error and the residual mismatch error change can be tracked and calibrated in real time without interrupting the normal operation of the SerDes system. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0029] Figure 1 One of the structural schematic diagrams of a background channel mismatch calibration system provided by the present invention;

[0030] Figure 2 The second structural diagram of a background channel mismatch calibration system provided by the present invention;

[0031] Figure 3 A third structural diagram of a background channel mismatch calibration system provided by the present invention;

[0032] Figure 4 A schematic diagram of the steps of a background channel mismatch calibration method provided by the present invention;

[0033] Figure 5 It is a schematic diagram of step 304 in the present invention;

[0034] Figure 6 It is a schematic diagram of step 3041 in the present invention;

[0035] Figure 7 A structural schematic diagram of a method for implementing background channel mismatch calibration provided by the present invention;

[0036] Figure 8 is a schematic diagram of step 3042 in the present invention;

[0037] Fig. 9 It is a schematic diagram of step 305 in the present invention.

[0038] Icons: 100-backstage channel mismatch calibration system; 101-time interleaved analog-to-digital converter; 102-digital signal processing circuit; 103-filter; 201-sub-analog-to-digital converter; 202-sub-signal processing module; 2011-phase adjustment component; 2012-gain adjustment component; 2013-processor. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0040] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0042] In the description of the present invention, it should be noted that if the terms "upper", "lower", "inside", "outside", etc. appear to indicate an orientation or position relationship, they are based on the orientation or position relationship shown in the accompanying drawings, or are the orientation or position relationship in which the product of the invention is usually placed when used. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0043] In addition, the terms “first”, “second”, etc., if used, are merely used to distinguish between the descriptions and should not be understood as indicating or implying relative importance.

[0044] It should be noted that, in the absence of conflict, the features in the embodiments of the present invention may be combined with each other.

[0045] As described in the background technology, in order to avoid the above-mentioned mismatch effect, a calibration algorithm is used to adjust each data transmission channel before the time-interleaved analog-to-digital converter actually works. The above-mentioned calibration algorithm includes a foreground calibration scheme and a background calibration scheme. The existing background calibration scheme usually needs to estimate the residual mismatch error between the data transmission channels corresponding to each sub-analog-to-digital converter based on the expected statistical characteristics of the data. In the background calibration scheme, the SerDes data will be subject to high-frequency loss after passing through the channel, resulting in poor robustness of the background calibration scheme based on expected statistical characteristics.

[0046] Considering the robustness of the calibration algorithm, a foreground calibration scheme is usually adopted. The foreground calibration scheme needs to inject a known reference signal in advance, and then calibrate the mismatch error between different data transmission channels based on the known reference signal.

[0047] However, when the SerDes system is operating normally, the foreground calibration solution cannot calibrate in real time the mismatch error changes caused by process, voltage, temperature, etc. And when the residual mismatch error changes, it is necessary to re-inject the known reference signal. The above adjustment method will interrupt the normal operation of the SerDes system.

[0048] Based on this, the present invention provides a background channel mismatch calibration system, which can track and calibrate the residual mismatch error and the residual mismatch error change in real time without interrupting the normal operation of the SerDes system.

[0049] The embodiments of the present invention can be implemented as follows:

[0050] Please refer to Figure 1 The present invention provides a background channel mismatch calibration system 100 (equivalent to a SerDes system), comprising a time-interleaved analog-to-digital converter 101, a digital signal processing circuit 102, and a filter 103. The digital signal processing circuit 102 is connected to the time-interleaved analog-to-digital converter 101 and the filter 103, respectively, and the time-interleaved analog-to-digital converter 101 is also connected to the filter 103.

[0051] Please refer to Figure 2 In this embodiment, the time-interleaved analog-to-digital converter 101 includes a plurality of sub-analog-to-digital converters 201; the digital signal processing circuit 102 includes a plurality of sub-signal processing modules 202; each sub-analog-to-digital converter 201 corresponds to and is connected to each sub-signal processing module 202, and forms a plurality of data transmission channels. The filter 103 is also connected to each sub-analog-to-digital converter 201. The data transmission channel S1 is represented by the reference numeral i =1 and the corresponding analog-to-digital converter j = 1 constitutes a data transmission channel. Similarly, the data transmission channel S MRepresented by the label i =M sub-ADC and corresponding label j =M sub-signal processing modules constitute a data transmission channel, where M is a positive integer and satisfies M≥1. In this embodiment, the total number of sub-analog-to-digital converters is the same as the total number of sub-signal processing modules.

[0052] In one possible implementation, each sub-signal processing module under the digital signal processing circuit includes a feed forward equalizer (FFE) and a decision feedback equalizer (DFE) connected in sequence. After the sub-signal processing module completes the processing of the signal, the processed signal can be sent to the deserializer, and finally output through the deserializer. In this embodiment, each sub-signal processing module can directly determine the decision parameters of the corresponding digital signal based on the above-mentioned feed forward equalizer and decision feedback equalizer, wherein the decision parameters include a decision error and a decision result. The present application does not limit the specific method of determining the decision parameters.

[0053] In this embodiment, for any data transmission channel, the sub-analog-to-digital converter converts the input analog signal into a digital signal.

[0054] The sub-signal processing module receives the digital signal of the corresponding data transmission channel and needs to calibrate the sub-signal processing module using a preset mismatch calibration signal before processing the digital signal. The calibrated sub-signal processing module is then used to determine the decision parameters at different sampling times based on the digital signal.

[0055] In one possible implementation, a mismatch calibration circuit may be added to calibrate the sub-signal processing module using a preset mismatch calibration signal. Specifically, a known reference signal is injected using the mismatch calibration circuit, and the known reference signal may include a known gain mismatch calibration signal and a known timing mismatch calibration signal, etc. Then, the mismatch values ​​between different data transmission channels are calibrated using the above-mentioned known reference signals, so that the gain mismatch value and timing mismatch value of each channel tend to 0, thereby achieving the purpose of improving the performance of the time-interleaved analog-to-digital converter. At this point, the SerDes system can output a relatively stable decision parameter, which can further improve the accuracy of the background channel mismatch calibration system in the present application.

[0056] Subsequently, the filter uses the decision parameters at different sampling times, especially the decision parameters at the current sampling time and the decision parameters at the previous sampling time to determine the residual mismatch error estimation function of each data transmission channel; the residual mismatch error estimation function includes a gain mismatch error estimation function and a timing mismatch error estimation function.

[0057] The sub-ADC iteratively adjusts its own operating parameters according to the timing mismatch error and the gain mismatch error until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal meet the preset values, thereby obtaining the aligned digital signal.

[0058] Please refer to Figure 3 The present application also provides a structural schematic diagram of a background channel mismatch calibration system 100, wherein the sub-analog-to-digital converter 201 includes a phase adjustment component 2011 and a gain adjustment component 2012, and the background channel mismatch calibration system 100 also includes a processor 2013, wherein the filter 103 is connected to the phase adjustment component 2011 and the gain adjustment component 2012; the phase adjustment component 2011 is connected to the gain adjustment component 2012 and the processor 2013; and the gain adjustment component 2012 is also connected to the processor 2013.

[0059] The phase adjustment component 2011 is used to iteratively adjust its own operating parameters according to the timing mismatch error estimation function.

[0060] The gain adjustment component 2012 is used to iteratively adjust its own operating parameters according to the gain mismatch error estimation function.

[0061] Processor 2013 is used to determine whether the timing mismatch compensation value corresponding to the digital signal at the current sampling moment is equal to the first preset value; it is also used to determine whether the gain mismatch compensation value corresponding to the digital signal at the current sampling moment is equal to the second preset value; it is also used to use the digital signal in the current state as the aligned digital signal when the timing mismatch compensation value corresponding to the digital signal at the current sampling moment is equal to the first preset value and the corresponding gain mismatch compensation value is equal to the second preset value.

[0062] In this embodiment, the phase adjustment component can be an analog circuit. For example, an analog phase shifter (Tapped Delay Line-TDL) can be added to the ADC sampling clock in the sub-analog-to-digital converter to calibrate the timing mismatch error by phase shifting, so that the timing mismatch compensation value corresponding to the digital signal at the current sampling moment is equal to the first preset value. The gain adjustment component can be a multiplier. For example, when the gain error under any data transmission channel is estimated to be a , the parameters of the multiplier can be adjusted so that the corresponding gain mismatch compensation value is equal to the second preset value. For example, the output of the current data transmission channel is multiplied by (1+ a ). On this basis, the processor can determine whether the timing mismatch compensation value and / or gain mismatch compensation value corresponding to the digital signal meets the preset value requirements after the analog phase shifter and / or multiplier adjusts its own operating parameters, and then obtain the calibrated digital parameters through the above iterative processing.

[0063] As an implementation method, based on the above-mentioned background channel mismatch calibration system, the present application provides a background channel mismatch calibration method, please refer to Figure 4 , the method includes steps 301 to 305.

[0064] Step 301: For each data transmission channel, use a sub-analog-to-digital converter to convert an input analog signal into a digital signal.

[0065] Step 302: Send a preset mismatch calibration signal to the sub-signal processing module.

[0066] The sub-signal processing module performs calibration according to a preset mismatch calibration signal.

[0067] Step 303: Utilize the calibrated sub-signal processing modules to obtain decision parameters of each data transmission channel at different sampling times based on the digital signal.

[0068] Step 304: using a filter to determine a residual mismatch error estimation function of each data transmission channel according to the decision parameter at the current sampling time and the decision parameter at the previous sampling time; the residual mismatch error estimation function includes a gain mismatch error estimation function and a timing mismatch error estimation function.

[0069] Step 305, iteratively adjust the operating parameters of the corresponding sub-ADCs according to the timing mismatch error estimation function and the gain mismatch error estimation function, until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal both meet the preset values, thereby obtaining an aligned digital signal.

[0070] In this embodiment, the calibrated sub-signal processing module is used to obtain the decision parameters under different data transmission channels, and then the decision parameters are filtered by a filter to obtain the corresponding timing mismatch error estimation function and gain mismatch error. Subsequently, the corresponding sub-analog-to-digital converter operating parameters are iteratively adjusted in sequence according to the timing mismatch error estimation function and the gain mismatch error estimation function until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal meet the preset value, and the aligned digital signal is obtained.

[0071] The operating parameters can be understood as the parameters of the corresponding components under the sub-ADC, for example, the parameters of the structures such as the multiplier and the analog phase shift circuit under the sub-ADC. Based on this, the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal can meet the preset values ​​by adjusting the parameters of the structures such as the multiplier and the analog phase shift circuit.

[0072] In one possible implementation, in order to meet the requirements of ever-increasing communication rates, a four-level pulse amplitude modulation (Pulse Amplitude Modulation, PAM-4) may be used to replace a conventional non-return to zero (NRZ) modulation scheme to reduce signal loss at the Nyquist frequency.

[0073] In this embodiment, the decision parameters include the decision result and the decision error. Figure 4 Based on Figure 5 The steps of using the filter to determine the residual mismatch error estimation function of each data transmission channel according to the decision parameter at the current sampling moment and the decision parameter at the previous sampling moment include step 3041 and step 3042.

[0074] Step 3041: For each data transmission channel, determine the gain mismatch error estimation function of the current data transmission channel according to the decision result at the current sampling time and the decision error at the current sampling time.

[0075] Step 3042: Determine a timing mismatch error estimation function of the current data transmission channel according to the decision result at the current sampling time and the decision error at the previous sampling time.

[0076] In one possible implementation, please refer to Figure 6 In this embodiment, the step of determining the gain mismatch error estimation function of the current data transmission channel according to the judgment result at the current sampling moment and the judgment error at the current sampling moment includes step 3041-1 and step 3041-2.

[0077] Step 3041-1: construct a gain mismatch cost function according to the decision result at the current sampling moment and the decision error at the current sampling moment.

[0078] Step 3041 - 2: Filter the gain mismatch cost function to determine a gain mismatch error estimation function.

[0079] In this embodiment, the gain mismatch cost function can be constructed based on the decision result at the current sampling time and the decision error at the current sampling time. k Taking the data transmission channel as an example, the judgment result at the current sampling time can be expressed as: In order to reduce the signal loss of the Nyquist frequency, this embodiment adopts four-level pulse amplitude modulation. Correspondingly, adjacent decision symbols are uncorrelated, and the following relationship can be obtained: ;in, is the expected function, For p The corresponding kPAM-4 modulation symbol of the channel; For p The corresponding k+ 1 PAM-4 modulation symbol.

[0080] Based on this, the decision error at the same time can be expressed as: Among them, the decision error can be determined based on the digital signal, and the calculation formula of the decision error satisfies: .in, The sub-analog-to-digital converter converts the input analog signal into a digital signal, which corresponds to the input data of the sub-signal processing module.

[0081] Among them, k Data transmission channels in Digital signal at the moment satisfy: ; For the k Gain mismatch error of each data transmission channel; For the k Data transmission channels in The time corresponds to modulation symbols under symbol numbers; is the sampling period; For the k The timing mismatch error of each data transmission channel; For the k Data transmission channels in Additive noise at the moment.

[0082] In this embodiment, the sub-signal processing module can process the digital signal output by the sub-analog-to-digital converter. Equalization is performed so that inter-symbol interference (ISI) is eliminated. In this embodiment, the decision result is determined based on the PAM-4 modulation symbol after equalization. In this embodiment, the value corresponding to the decision result includes four possibilities, for example, -1, +1, -3 and +3.

[0083] In a possible implementation, the judgment result can be determined according to the value of the equalized digital signal. For example, when the digital signal output by the sub-signal processing module is When the corresponding value is 0.7, the corresponding most recent judgment result is +1, and we can get =1, then the corresponding decision error is 0.3. When the SerDes system is in good working condition, the decision symbol and the corresponding decision result satisfy: = .

[0084] Based on the judgment result at the current sampling time The decision error at the current sampling time The gain mismatch cost function is constructed by multiplying the expected value of the time. The gain mismatch cost function is expressed as: ;

[0085] in, is the total number of data transmission channels; For the k Data transmission channels in Gain mismatch error estimation function at time ; is the channel impulse response function; For the k The timing mismatch error of each data transmission channel; For the k Data transmission channels in The timing mismatch error estimation function at time instant.

[0086] Further, a gain mismatch error estimation function may be determined based on the gain mismatch cost function. In this embodiment, the gain mismatch cost function is related to the gain mismatch error of the data transmission channel. In a possible implementation, when the filter is an LMS adaptive filter, the corresponding gain mismatch error estimation function may be determined based on the minimum mean square error of the difference between the true value and the estimated value of the gain mismatch error in a statistical sense.

[0087] Please refer to Figure 7 , Figure 7 An implementation structure of the background channel mismatch calibration method in this embodiment is shown. Correspondingly, the expression of the gain mismatch error estimation function satisfies:

[0088] ;

[0089] in, For the k Data transmission channels in Gain mismatch error estimation function at time ; is the first step parameter, which may be the update step of the gain mismatch error LMS iteration, and its value is a constant.

[0090] Similarly, the timing mismatch error estimation function in this embodiment may be determined in the same manner as the gain mismatch error estimation function. Figure 5 Based on Figure 8 In this embodiment, the step of determining the timing mismatch error estimation function of the current data transmission channel according to the judgment result at the current sampling moment and the judgment error at the previous sampling moment includes step 3042-1 and step 3042-2.

[0091] Step 3042-1: construct a timing mismatch cost function according to the decision result at the current sampling time and the decision error at the previous sampling time.

[0092] Step 3042-2: Filter the timing mismatch cost function to determine the timing mismatch error estimation function of the current data transmission channel.

[0093] In this embodiment, a timing mismatch cost function may be constructed based on the decision result at the current sampling moment and the decision error at the previous sampling moment. For example, the timing mismatch cost function may be determined based on the expectation of the product between the decision result at the current sampling moment and the decision error at the previous sampling moment. The timing mismatch cost function is expressed as:

[0094] ;

[0095] in, For the Data transmission channels in The verdict at the moment.

[0096] Furthermore, a timing mismatch error estimation function can be determined based on the timing mismatch cost function. Since the timing mismatch cost function is related to the timing mismatch error of the data transmission channel, in one possible implementation, when the filter is an LMS adaptive filter, the corresponding timing mismatch error estimation function can be determined based on the minimum mean square error of the difference between the true value and the estimated value of the timing mismatch error in a statistical sense. Please continue to refer to Figure 7 , the expression of the timing mismatch error estimation function satisfies:

[0097] ;

[0098] in, For the k Data transmission channels in The timing mismatch error estimation function at time instant; is a second step parameter, which may be an update step of the timing mismatch error LMS iteration, and has a value that is a constant.

[0099] After obtaining the corresponding timing mismatch error estimation function and residual mismatch error estimation function, the sub-analog-to-digital converter can use the above-mentioned timing mismatch error estimation function and residual mismatch error estimation function to iteratively adjust its own operating parameters in sequence until the gain mismatch compensation value and timing mismatch compensation value corresponding to the digital signal meet the preset values, thereby obtaining the aligned digital signal.

[0100] Please continue to refer to Figure 2When the sub-ADC 201 includes a phase adjustment component 2011 and a gain adjustment component 2012, and the backstage channel mismatch calibration system 100 includes a processor 2013, please refer to Fig. 9 , according to the timing mismatch error estimation function and the gain mismatch error estimation function, the operating parameters of the corresponding sub-analog-to-digital converter are iteratively adjusted in sequence until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal meet the preset values, and the step of obtaining the aligned digital signal includes steps 3051 to 3053.

[0101] Step 3051: Iteratively adjust the operating parameters of the phase adjustment component according to the timing mismatch error estimation function.

[0102] Step 3052: Iteratively adjust the operating parameters of the gain adjustment component according to the gain mismatch error estimation function.

[0103] Step 3053, using the processor to determine whether the timing mismatch compensation value corresponding to the digital signal at the current sampling moment is equal to the first preset value; determine whether the gain mismatch compensation value corresponding to the digital signal at the current sampling moment is equal to the second preset value; and when the timing mismatch compensation value corresponding to the digital signal at the current sampling moment is equal to the first preset value, and the corresponding gain mismatch compensation value is equal to the second preset value, the digital signal in the current state is used as the aligned digital signal.

[0104] It should be noted that when executing the above steps, there is no specific limitation on the execution method of the specific sub-steps in step 305, that is, in this embodiment, the execution method of each sub-step in step 305 can be serial execution and / or parallel execution.

[0105] Please continue to refer to Fig. 9 In one possible implementation, when step 305 includes at least step 3051, step 3052, and step 3053, the execution method may be: after executing step 3051 and step 3052 in parallel, execute step 3052 according to the processing results of step 3051 and step 3052.

[0106] In one possible implementation, the processor can also adjust the corresponding working states of the sub-analog-to-digital conversion modules under each time-interleaved analog-to-digital converter and the sub-signal processor modules under the digital signal processing circuit according to the above-mentioned background channel mismatch calibration method, which will not be repeated here.

[0107] In this embodiment, when the timing mismatch compensation value corresponding to the digital signal at the current sampling time is equal to the first preset value, and the corresponding gain mismatch compensation value is equal to the second preset value, the aligned digital signal can be expressed as:

[0108] ;

[0109] Among them, the timing mismatch compensation value is expressed as: ; The gain mismatch compensation value is expressed as: After calibration by the above-mentioned background channel mismatch calibration method, the timing mismatch compensation value can satisfy: , that is, the timing mismatch compensation value approaches 1; the gain mismatch compensation value is expressed as: ; That is, the gain mismatch compensation value approaches 0.

[0110] In this embodiment, if the timing mismatch compensation value corresponding to the digital signal at the current sampling moment is not equal to the first preset value and / or the corresponding gain mismatch compensation value is not equal to the second preset value, at this time, the phase adjustment component is continuously used to iteratively adjust its own operating parameters based on the timing mismatch error and / or the gain adjustment component is continuously used to iteratively adjust its own operating parameters based on the gain mismatch error.

[0111] In summary, the present invention provides a background channel mismatch calibration method, which is applied to a background channel mismatch calibration system. The above-mentioned background channel mismatch calibration system includes a time-interleaved analog-to-digital converter, a digital signal processing circuit and a filter. Each sub-analog-to-digital converter under the time-interleaved analog-to-digital converter corresponds to and is connected to each sub-signal processing module under the digital signal processing circuit, and constitutes a plurality of data transmission channels. Among them, for each data transmission channel, the input analog signal is converted into a digital signal by using a sub-analog-to-digital converter. Then, the sub-signal processing module is calibrated using a preset mismatch calibration signal. On this basis, the calibrated sub-signal processing modules are used to determine the decision parameters of each data transmission channel at different sampling times according to the digital signal. Subsequently, the filter is used to determine the residual mismatch error estimation function of each data transmission channel according to the decision parameters at the current sampling time and the decision parameters at the previous sampling time; wherein, the residual mismatch error estimation function includes a gain mismatch error estimation function and a timing mismatch error estimation function. Each sub-ADC is used to iteratively adjust its own operating parameters according to the timing mismatch error and the gain mismatch error, until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal meet the preset values, thereby obtaining the aligned digital signal.

[0112] The present invention first calibrates the sub-signal processing module using a preset mismatch calibration signal, and then uses each calibrated sub-signal processing module to determine the decision parameters of each data transmission channel at different sampling times according to the digital signal, thereby determining the residual mismatch error (i.e., decision error) and the decision result under the corresponding data transmission channel. At the same time, the decision error change caused by the influence of process, voltage, temperature, etc. can be determined based on the decision result and the decision error at different sampling times, and then the residual mismatch error and the residual mismatch error change can be tracked and calibrated in real time without interrupting the normal operation of the SerDes system.

[0113] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A background channel mismatch calibration method, characterized in that: The invention is applied to a background channel mismatch calibration system, wherein the background channel mismatch calibration system comprises a time-interleaved analog-to-digital converter, a digital signal processing circuit and a filter, wherein the digital signal processing circuit is respectively connected to the time-interleaved analog-to-digital converter and the filter, and the filter is also connected to the time-interleaved analog-to-digital converter; the time-interleaved analog-to-digital converter comprises a plurality of sub-analog-to-digital converters; the digital signal processing circuit comprises a plurality of sub-signal processing modules; each of the sub-analog-to-digital converters is connected to one of the sub-signal processing modules to form a plurality of data transmission channels, and the method comprises: For each of the data transmission channels, converting the input analog signal into a digital signal using the sub-analog-to-digital converter; Sending a preset mismatch calibration signal to the sub-signal processing module, wherein the sub-signal processing module performs calibration according to the preset mismatch calibration signal; Using the calibrated sub-signal processing modules to obtain the decision parameters of each of the data transmission channels at different sampling times based on the digital signal; Determining the residual mismatch error estimation function of each of the data transmission channels by using the filter according to the decision parameter at the current sampling moment and the decision parameter at the previous sampling moment; the residual mismatch error estimation function includes a gain mismatch error estimation function and a timing mismatch error estimation function; The timing mismatch error estimation function and the gain mismatch error estimation function are used to iteratively adjust the operating parameters of the corresponding sub-analog-to-digital converters in sequence until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal both meet the preset values, thereby obtaining the aligned digital signal.

2. The background channel mismatch calibration method according to claim 1, characterized in that: The decision parameters include a decision result and a decision error. The steps of using the filter to determine the residual mismatch error estimation function of each data transmission channel according to the decision parameters at the current sampling moment and the decision parameters at the previous sampling moment include: For each of the data transmission channels, determining a gain mismatch error estimation function of the current data transmission channel according to a decision result at a current sampling moment and a decision error at the current sampling moment; The timing mismatch error estimation function of the current data transmission channel is determined according to the decision result at the current sampling moment and the decision error at the previous sampling moment.

3. The background channel mismatch calibration method according to claim 2, characterized in that: The step of determining the gain mismatch error estimation function of the current data transmission channel according to the decision result at the current sampling moment and the decision error at the current sampling moment comprises: Constructing a gain mismatch cost function according to the decision result at the current sampling moment and the decision error at the current sampling moment; The gain mismatch cost function is filtered to determine a gain mismatch error estimation function.

4. The background channel mismatch calibration method according to claim 3, characterized in that: The gain mismatch cost function satisfies: ; in, is the expected function; For the k Data transmission channels in The judgment error at the moment; For the k Data transmission channels in The judgment result at the moment; is the total number of data transmission channels; For the k Gain mismatch error of each data transmission channel; For the k Data transmission channels in Gain mismatch error estimation function at time ; is the channel impulse response function; For the k The timing mismatch error of each data transmission channel; For the k Data transmission channels in The timing mismatch error estimation function at time instant; For the k Data transmission channels in Additive noise at the moment.

5. The background channel mismatch calibration method according to claim 1, characterized in that: When the filter is an LMS adaptive filter, the expression of the gain mismatch error estimation function satisfies: ; in, For the k Data transmission channels in Gain mismatch error estimation function at time , For the k Data transmission channels in Gain mismatch error estimation function at time ; is the first step parameter; For the k Data transmission channels in The judgment error at the moment; For the k Data transmission channels in The judgment result at the moment; is the total number of data transmission channels.

6. The background channel mismatch calibration method according to claim 2, characterized in that: The step of determining the timing mismatch error estimation function of the current data transmission channel according to the decision result at the current sampling moment and the decision error at the previous sampling moment comprises: Constructing a timing mismatch cost function based on the decision result at the current sampling moment and the decision error at the previous sampling moment; The timing mismatch cost function is filtered to determine a timing mismatch error estimation function of a current data transmission channel.

7. The background channel mismatch calibration method according to claim 6, characterized in that: The timing mismatch cost function is expressed as: ; in, is the expected function; For the k Data transmission channels in The judgment error at the moment; For the k+ 1 data transmission channel in The judgment result at the moment; is the total number of data transmission channels; For the k Gain mismatch error of each data transmission channel; For the k Data transmission channels in Gain mismatch error estimation function at time ; is the channel impulse response function; is the sampling period; For the k The timing mismatch error of each data transmission channel; For the k Data transmission channels in The timing mismatch error estimation function at time instant; For the k Data transmission channels in Additive noise at the moment.

8. The background channel mismatch calibration method according to claim 1, characterized in that: When the sub-analog-to-digital converter includes a phase adjustment component and a gain adjustment component, and the background channel mismatch calibration system also includes a processor, the filter is also connected to the phase adjustment component and the gain adjustment component; the phase adjustment component is connected to the gain adjustment component and the processor; the gain adjustment component is also connected to the processor; the timing mismatch error estimation function and the gain mismatch error estimation function are used to iteratively adjust the operating parameters of the corresponding sub-analog-to-digital converter in sequence until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal meet the preset value, and the step of obtaining the aligned digital signal is: Iteratively adjusting the operating parameters of the phase adjustment component according to the timing mismatch error estimation function; Iteratively adjusting the operating parameters of the gain adjustment component according to the gain mismatch error estimation function; Using the processor to determine whether the timing mismatch compensation value corresponding to the digital signal at the current sampling moment is equal to a first preset value; Determine whether the gain mismatch compensation value at the current sampling moment is equal to the second preset value; When the timing mismatch compensation value corresponding to the digital signal at the current sampling moment is equal to the first preset value, and the corresponding gain mismatch compensation value is equal to the second preset value, the digital signal in the current state is used as the aligned digital signal.

9. The background channel mismatch calibration method according to any one of claims 1 to 8, characterized in that: The expression of the digital signal after the registration is: ; ; ; in, For the k Data transmission channels in The digital signal after registration at the moment; For the k Gain mismatch error of each data transmission channel; For the k Data transmission channels in Gain mismatch error estimation function at time ; is the channel impulse response function; is the sampling period; For the k The timing mismatch error of each data transmission channel; For the k Data transmission channels in The timing mismatch error estimation function at time instant; For the k Data transmission channels in Additive noise at the moment; For the k Data transmission channels in The time corresponds to modulation symbols under symbol numbers; is the total number of data transmission channels.

10. A background channel mismatch calibration system, characterized in that: The background channel mismatch calibration system includes a time-interleaved analog-to-digital converter, a digital signal processing circuit and a filter. The digital signal processing circuit is connected to the time-interleaved analog-to-digital converter and the filter respectively, and the filter is also connected to the time-interleaved analog-to-digital converter; the time-interleaved analog-to-digital converter includes a plurality of sub-analog-to-digital converters; the digital signal processing circuit includes a plurality of sub-signal processing modules; each of the sub-analog-to-digital converters is connected to one of the sub-signal processing modules to form a plurality of data transmission channels, wherein: For each of the data transmission channels, the sub-analog-to-digital converter is used to convert an input analog signal into a digital signal; The sub-signal processing module is used to perform calibration using a preset mismatch calibration signal, and obtain decision parameters of each of the data transmission channels at different sampling times based on the digital signal; The filter is used to determine the residual mismatch error estimation function of each of the data transmission channels according to the decision parameters at the current sampling time and the decision parameters at the previous sampling time; the residual mismatch error estimation function includes a gain mismatch error estimation function and a timing mismatch error estimation function; The sub-ADC is also used to iteratively adjust its own operating parameters according to the timing mismatch error and the gain mismatch error in sequence until the gain mismatch compensation value and the timing mismatch compensation value corresponding to the digital signal both meet preset values ​​to obtain an aligned digital signal.

Citation Information

Patent Citations

  • Digital background real-time compensating method for TIADC with reference channel

    CN104467842A

  • Measuring and correcting non-idealities of system

    CN107959499A