Digital correction circuit processing method for two-step monoclinic ADC
By combining forward and reverse counting with an adder, the problem of excessive layout area and power consumption in traditional two-step single-slope ADCs is solved, realizing a low-cost, small-area digital correction circuit that can meet the needs of different operating conditions.
Patent Information
- Application Number
- CN202411125624.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-11-19
AI Technical Summary
The digital correction circuit of a traditional two-step single-slope ADC requires the introduction of a subtractor, which increases the layout area and power consumption, and has poor reusability. It is especially unsuitable for two-step single-slope ADCs with high coarse quantization bits.
By combining forward and reverse counting with an adder, and through blocking multi-input AND gates and register control, the coarse and fine quantization results are added in a staggered manner, avoiding the use of a subtractor and using an adder for correction processing.
It reduces layout area and power consumption, lowers design complexity, and improves circuit reusability and adaptability, especially when the number of finer bits is large, it significantly saves subtractor area.
Smart Images

Figure CN118984156A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of integrated circuit design, and in particular to a digital correction circuit processing method for a two-step single-slope ADC. Background Art
[0002] Analog-to-digital converters are an important part of modern integrated circuits. Different ADCs have their own characteristics. Single-slope ADCs are widely used in the field of image sensing due to their simple structure and small layout area. In order to improve its conversion speed, two-step single-slope ADCs have begun to attract attention. In two-step quantization, in order to overcome the "dead zone" phenomenon caused by deviations caused by irrational factors such as comparator offset, noise, and delay, the concept of redundant bit RSD is introduced. Therefore, after the quantization is completed, the results of coarse quantization and fine quantization need to be staggered and added to obtain the final quantization result.
[0003] In the traditional correction circuit design, it is necessary to complete the offset addition and subtract an offset value, which requires the introduction of a subtractor, increasing the layout area and power consumption, and has poor reusability, especially not suitable for two-step single-slope ADCs with a high number of coarse quantization bits. Summary of the invention
[0004] The purpose of the present invention is to design a digital correction processing circuit for a two-step single-slope ADC with a smaller layout area, thereby reducing the complexity and sophistication of a traditional digital correction circuit.
[0005] The object of the present invention can be achieved by the following technical solutions: A digital correction circuit processing method for a two-step single slope ADC, comprising step one: a coarse code value generating circuit is synchronously counted with a counter of a ramp generator of the single slope ADC, when a multi-input AND gate is blocked, a CK signal stops affecting a subsequent reverse counter, and transmits the signal to step two;
[0006] Step 2: When the fine quantization comparison result comp_f of the comparator flips from a high level to a low level, comp_f blocks the multi-input AND gate so that the CK signal no longer affects the fine quantization counter code value, and transmits the signal to step 3;
[0007] Step 3: The coarse quantization code value and RSD are sent to the adder, and the RSD and the coarse quantization lowest bit are added to obtain the final quantization result which is then stored in the register.
[0008] Preferably, the specific content of step one is as follows:
[0009] The coarse quantization result generating circuit is used to generate an m-bit coarse quantization code value, and the fine quantization result generating circuit is used to generate (n+1) bits; the quantization result, wherein the highest bit RSD is a redundant bit;
[0010] When the two-step single slope ADC starts quantization, the register of the coarse code value generation part is set to a logic high level, and the register of the fine code value generation part is set to a high level or a low level according to the setting, and is set to a binary code value corresponding to the number of steps of the upward extension and the downward extension. In the coarse quantization stage, coarse and comp_c without comparison results are set to a high level. comp_c is used to indicate the moment when the comparison is completed in the coarse quantization stage. The output of the multi-input AND gate of the coarse code value generation part is only related to the CK signal. The CK signal is the main clock of the ADC quantization. The clock cycle of CK represents the minimum time unit of the ramp voltage step during ADC quantization.
[0011] At the same time, comp_f is at a high level, and comp_f is used to indicate the moment when the comparison is completed in the fine quantization stage, and controls the multiplexer to connect the multi-input AND gate and the clock input terminal of the register, so that the coarse code value generation circuit and the counter of the ramp generator of the single slope ADC count synchronously;
[0012] When the coarse quantization result comp_c of the comparator in the coarse quantization stage flips, the multi-input AND gate is blocked, and the CK signal will no longer affect the subsequent reverse counter;
[0013] The fine quantization code value generation part is used to control the quantization state of the ADC during coarse quantization because the fine signal is at a low level during coarse quantization. When the Coarse signal and the fine signal are at a high level, the two-step single slope ADC is regulated to perform coarse quantization.
[0014] When the fi ne signal level is high and the coarse level is low, the two-step single slope ADC is regulated to perform fine quantization;
[0015] By blocking the multi-input AND gate, the code value of the fine quantization code value generation part is kept in the state set at the beginning of quantization. Comp_c and comp_f respectively represent the output of the ADC comparator. When the ramp signal at one input end of the comparator does not pass through the signal to be compared at the other end, the output of the comparator is high; when the ramp signal at one input end of the comparator passes through the signal to be compared at the other end, the output of the comparator is low. comp_c is used to indicate the moment when the comparison is completed in the coarse quantization stage, and comp_f is used to indicate the moment when the comparison is completed in the fine quantization stage.
[0016] Preferably, the specific process of step 2 is as follows:
[0017] The coarse signal is at a low level, blocking the coarse quantization code value generation part. The CK signal has no interference with the code value generated by the coarse quantization stage counting, while the output result of the multi-input AND gate of the fine quantization code value generation part is only related to CK. The counter of the fine quantization code value generation part starts counting forward from the set state at the beginning of quantization. When the fine quantization comparison result comp_f of the comparator flips from a high level to a low level, comp_f blocks the multi-input AND gate, so that the CK signal no longer affects the fine quantization counter code value.
[0018] The highest two bits QR and F l ag of the fine quantization code value generation part are redundant bits and flag bits respectively, which are set to high at the beginning of each cycle quantization. The function of the QR redundant bit is to expand the range of the extended fine slope to cover the dead zone. The function of the flag bit is to indicate whether the fine quantization code value state is in the lower extended slope, normal state or upper extended slope. The SC and H signals are used to control the initial state of the register. They are connected to the SET terminal or CLR terminal of the register to set the register to a high level or a low level.
[0019] Preferably, the specific process of step three is as follows:
[0020] When the output of the coarse quantization multi-input AND gate is in a low level state, and then the comparator flips in the extended state, the flag is high level, and the multiplexer MUX switches, the output of MUX generates a rising edge, so that the coarse quantization counter counts again, and the adder is used to perform offset addition to obtain the quantization code value after correction. Since the coarse quantization has ended at this time, the output of the coarse quantization multi-input AND gate must be in a low level state. At this time, if the comparator flips in the extended state, the flag is high level, and the multiplexer MUX switches, the output of MUX generates a rising edge, so that the coarse quantization counter counts again, which is equivalent to realizing subtraction.
[0021] The beneficial effects of the present invention are as follows:
[0022] The digital correction processing circuit for two-step single slope proposed by the present invention implements subtraction based on the forward and reverse counting of the counter and then uses the adder to perform offset addition to obtain the quantized code value after correction processing. No additional multi-bit subtractor is required, which saves layout area and power consumption, and has the advantages of simple structure and no need for complex design based on the digital integrated circuit design process.
[0023] At the same time, the number of bits for coarse and fine quantization and the starting state of fine quantization can be configured according to the specific situation. Therefore, this circuit can meet the digital correction requirements of different working conditions, has the advantage of easy expansion of applications, and reduces design costs and complexity. Especially when the number of bits for fine quantization is large, the area of the subtractor saved is more significant. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] The present invention will be further described below in conjunction with the accompanying drawings;
[0025] Figure 1 It is a flowchart of the system of the present invention;
[0026] Figure 2 is the timing diagram of this method;
[0027] Figure 3 This is the structure diagram of module 1;
[0028] Figure 4 This is the overall structure diagram of digital correction. DETAILED DESCRIPTION
[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.
[0030] Example:
[0031] See also Figures 1 to 4 As shown, the present invention is a digital correction circuit processing method for a two-step single slope ADC, comprising step 1: a coarse code value generating circuit and a counter of a slope generator of a single slope ADC are counted synchronously, when a multi-input AND gate is blocked, a CK signal stops affecting a subsequent reverse counter, and the signal is transmitted to step 2; step 2: when a fine quantization comparison result comp_f of a comparator flips from a high level to a low level, comp_f blocks a multi-input AND gate, so that the CK signal no longer affects the code value of a fine quantization counter, and the signal is transmitted to step 3; step 3: a coarse quantization code value and RSD are sent to an adder, and RSD and the lowest bit of the coarse quantization are added to each other, and the final quantization result is obtained and then stored in a register;
[0032] The present invention proposes a low-cost, small-area digital correction processing circuit for quantization results of a two-step single-slope ADC; based on the register set-reset function and the working principle of forward / direction counting, a register, a logic gate and a multiplexer are used to realize digital correction processing for a two-step single-slope ADC;
[0033] The timing sequence that the digital correction processing circuit must cooperate with is as follows Figure 2 As shown, the high level of the Coarse signal and the Fine signal is effective, and is used to control the quantization state of the ADC. When the coarse signal level is high and the Fine signal level is low, it means that the two-step single slope ADC is performing coarse quantization; when the Fine signal level is high and the coarse level is low, it means that the two-step single slope ADC is performing fine quantization.
[0034] Comp_c and comp_f represent the output of the ADC comparator respectively. When the ramp signal at one input end of the comparator does not cross the signal to be compared at the other end, the output of the comparator is high level; when the ramp signal at one input end of the comparator crosses the signal to be compared at the other end, the output of the comparator is low level. comp_c is used to indicate the moment when the comparison is completed in the coarse quantization stage, and comp_f is used to indicate the moment when the comparison is completed in the fine quantization stage. The SC and H signals are used to control the initial state of the register. They are connected to the SET end or CLR end of the register to set the register to a high level or a low level.
[0035] Part of the circuit architecture of the present invention is as follows Figure 3 As shown, Figure 3 The upper coarse quantization result generating circuit is used to generate an m-bit coarse quantization code value. Figure 3 The lower part of the fine quantization result generation circuit is used to generate (n+1) bits; the quantization result, wherein the highest bit RSD is a redundant bit; when the two-step single slope ADC starts quantization, all registers of the coarse code value generation part are set to logic high, and the registers of the fine code value generation part are set to high level or low level according to the setting; in the coarse quantization stage, the coarse level is high, and the comp_c is also high because the result is not compared, so the output of the multi-input AND gate of the coarse code value generation part is only related to the CK signal, and because the fine quantization has not started, comp_f is high, and the control multiplexer connects the multi-input AND gate and the clock input end of the register, so that the coarse code value generation circuit and the counter of the ramp generator of the single slope ADC count synchronously;
[0036] When the coarse quantization result comp_c of the comparator is flipped in the coarse quantization stage, the multi-input AND gate is blocked, and the CK signal can no longer affect the subsequent reverse counter; and the fine quantization code value generation part is in the coarse quantization, because the fine signal is at a low level during the coarse quantization, blocking the multi-input AND gate, so the code value of the fine quantization code value generation part always keeps the state set at the beginning of quantization; at the beginning of the fine quantization stage, because the coarse signal is at a low level, blocking the coarse quantization code value generation part, the CK signal has no interference with the code value generated by the coarse quantization stage counting;
[0037] The output result of the multi-input AND gate of the fine quantization code value generation part is only related to CK. The counter of the fine quantization code value generation part starts to count forward from the set state at the beginning of quantization. When the fine quantization comparison result comp_f of the comparator flips from a high level to a low level, comp_f blocks the multi-input AND gate, so that the CK signal no longer affects the fine quantization counter code value; the highest two bits QR and F l ag of the fine quantization code value generation part are redundant bits and flag bits respectively, which are set to high at the beginning of each cycle quantization. The role of the QR redundant bit is to expand the range of the extended fine slope to cover the dead zone; the role of the flag bit is to indicate whether the fine quantization code value state is in the lower extended slope, normal state or upper extended slope;
[0038] When in working state, when the coarse quantization is completed and the code value has been determined, fine quantization begins. The two-step single slope ADC with N-bit precision divides the two steps into fine quantization n+1 bits (one bit redundant) and coarse quantization Nn bits; during quantization, the code value generated by the coarse quantization code value is:
[0039] XXX…XXX (binary, m digits in total)
[0040] Then the state transition table of the fine-scale state machine is: (Z means to be determined)
[0041]
[0042] The function of the flag bit is that if the flip result of the fine quantization comparator comp_f arrives at the time of the next extension, Figure 2 , controls the multiplexer MUX to switch the data path to the fl ag signal. Since the coarse quantization has ended at this time, the output of the coarse quantization multi-input AND gate must be in a low level state. At this time, if the comparator flips in the extended state, the flag bit is high level, and the multiplexer MUX switches, then the output of MUX produces a rising edge, causing the coarse quantization counter to count again, which is equivalent to realizing subtraction.
[0043] The above formulas are obtained by collecting a large amount of data for software simulation and selecting a formula that is close to the real value. The coefficients in the formula are set by technical personnel in this field according to actual conditions. The above are only preferred specific implementations of the present invention, but the protection scope of the present invention is not limited to this. Any technical personnel familiar with the technical field within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, make equivalent replacement or change, which should be covered within the protection scope of the present invention.
Claims
1. A digital correction circuit processing method for a two-step single slope ADC, characterized in that: include: Step 1: The coarse code value generation circuit counts synchronously with the counter of the ramp generator of the single slope ADC. When the multi-input AND gate is blocked, the CK signal stops affecting the subsequent reverse counter and transmits the signal to step 2. Step 2: When the fine quantization comparison result comp_f of the comparator flips from a high level to a low level, comp_f blocks the multi-input AND gate so that the CK signal no longer affects the fine quantization counter code value, and transmits the signal to step 3; Step 3: The coarse quantization code value and RSD are sent to the adder, and the RSD and the coarse quantization lowest bit are added to obtain the final quantization result which is then stored in the register.
2. A digital correction circuit processing method for a two-step single slope ADC according to claim 1, characterized in that: The specific content of step one is as follows: The coarse quantization result generating circuit is used to generate an m-bit coarse quantization code value, and the fine quantization result generating circuit is used to generate (n+1) bits; the quantization result, wherein the highest bit RSD is a redundant bit; When the two-step single slope ADC starts quantization, the register of the coarse code value generation part is set to a logic high level, and the register of the fine code value generation part is set to a high level or a low level according to the setting, and is set to a binary code value corresponding to the number of steps of the upward extension and the downward extension. In the coarse quantization stage, coarse and comp_c that has not been compared are set to a high level, and comp_c is used to indicate the moment when the comparison is completed in the coarse quantization stage. The output of the multi-input AND gate of the coarse code value generation part is only related to the CK signal. The CK signal is the main clock of the ADC quantization. The clock cycle of CK represents the minimum time unit of the ramp voltage step during the ADC quantization. At the same time, comp_f is a high level, and comp_f is used to indicate the moment when the comparison is completed in the fine quantization stage. The control multiplexer connects the multi-input AND gate and the clock input end of the register, so that the coarse code value generation circuit and the counter of the ramp generator of the single slope ADC count synchronously; When the coarse quantization result comp_c of the comparator in the coarse quantization stage flips, the multi-input AND gate is blocked, and the CK signal will no longer affect the subsequent reverse counter; As for the fine quantization code value generation part, during the coarse quantization, since the fine signal is at a low level during the coarse quantization, when the coarse signal and the fine signal are at a high level and effective, it is used to control the quantization state of the ADC. When the coarse signal level is high and the fine signal level is low, the two-step single slope ADC is regulated to perform coarse quantization. When the fine signal level is high and the coarse level is low, the two-step single slope ADC is regulated for fine quantization; By blocking the multi-input AND gate, the code values of the part generating the fine quantization code values are kept in the state set at the beginning of quantization.
3. A digital correction circuit processing method for a two-step single slope ADC according to claim 2, characterized in that: The specific process of step 2 is as follows: The coarse signal is at a low level, blocking the coarse quantization code value generation part. The CK signal has no interference with the code value generated by the coarse quantization stage counting, while the output result of the multi-input AND gate of the fine quantization code value generation part is only related to CK. The counter of the fine quantization code value generation part starts to count forward from the set state at the beginning of quantization; When the fine quantization comparison result comp_f of the comparator flips from a high level to a low level, comp_f blocks the multi-input AND gate, so that the CK signal no longer affects the fine quantization counter code value; the highest two bits QR and Flag of the fine quantization code value generation part are redundant bits and flag bits respectively, which are set to high at the beginning of each cycle quantization. The role of the QR redundant bit is to expand the range of the extended fine slope to cover the dead zone, and the role of the flag bit is to indicate whether the fine quantization code value state is in the lower extended slope, normal state or upper extended slope.
4. A digital correction circuit processing method for a two-step single slope ADC according to claim 1, characterized in that: The specific process of step three is as follows: When the output of the coarse quantization multi-input AND gate is in a low level state, and then the comparator flips in the extended state, the flag is high level, and the multiplexer MUX switches, and the output of MUX produces a rising edge, so that the coarse quantization counter counts again, and the adder is used to perform offset addition to obtain the quantization code value after correction.
