Power consumption constant type D flip-flop circuit based on improved three-stage logic
By introducing improved three-stage logic and internal discharge signal generation into the DyCML structure, the power consumption constant problem of existing power-resistant attack-type triggers under unbalanced load capacitance is solved, and the power-resistant attack-resistant capability of the circuit is improved and hardware security.
Patent Information
- Application Number
- CN202410995285.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-12
AI Technical Summary
Existing power-resistant attack-type flip-flops have poor power consumption constant in the case of unbalanced circuit load capacitance, and attackers can destroy the power-resistant attack performance of the circuit by changing the clock frequency or slowing down the discharge phase.
An improved three-stage logic power consumption constant D flip-flop based on DyCML structure is designed to generate discharge signals using internal nodes of the circuit to prevent attacks such as clock frequency and other attacks from affecting the power consumption attack performance of the overall circuit.
It effectively reduces the average power consumption of the flip-flop, improves the circuit's power-resistant attack capability and hardware security, and prevents safety problems caused by unbalanced load capacitors.
Smart Images

Figure CN118944640A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of integrated circuits, and in particular relates to a constant power consumption D flip-flop circuit based on improved three-phase logic. Background Art
[0002] Side channel attacks use physical information such as time, power consumption or electromagnetic radiation generated when the encryption device processes data to attack the circuit. There are many types of side channel attack techniques for different physical information. Among them, power consumption attack technology has become a serious threat to cryptographic chips due to its simple operation, low cost and wide attack range. The principle of power consumption attack is to use the correlation between the processing of relevant data by the cryptographic module in the chip circuit and the power consumption generated by the circuit to attack the circuit. The mainstream countermeasures against power consumption attacks include dual rail pre-charge logic (DRPL), sense amplifier based logic (SABL), wave dynamic differential logic (WDDL), three-phase dual-rail pre-charge logic (TDPL), etc. The main idea of different countermeasures is to eliminate the correlation between power consumption and processing related data. In addition to the traditional DRP logic types, dynamic current mode logic (DyCML) is also a power-resistant structure. Due to its dynamic and differential logic characteristics, dynamic current mode logic also has good advantages in resisting power-resistant attacks.
[0003] The trigger is an important component of the circuit operation cryptographic unit, and is also a part that is vulnerable to power consumption attacks. Applying anti-power consumption attack technology to trigger design is an important part of designing secure circuits.
[0004] Currently, the existing power consumption attack-resistant triggers are mainly based on SABL, TDPL, DyCML and other structures. When the circuit load capacitance is unbalanced, the power consumption stability of the trigger based on SABL or DyCML structure is poor. Figure 2 As shown, assuming that in the previous cycle, the logic values of TA and TB after the evaluation phase are (1, 0), in this pre-charge phase, TB is charged to a high level, and in this evaluation phase, is (1,0), TB discharges to 0 according to the input value, and TA remains unchanged. In this stage, TB undergoes the process of 0→1 and 1→0. If the logical values of TA and TB that have undergone the evaluation stage in the previous cycle are (0,1), is (1,0), TA goes through the process of 0→1 and 1→0, and TB remains unchanged. Considering the unbalanced load problem of the circuit, the initial values of TA and TB generated in the previous cycle will cause unbalanced power consumption of the circuit. In view of the above problems, the TDPL structure can solve the hardware security problem caused by the unbalanced load through the discharge stage, but the attacker may reduce the speed of the clock input, change the clock frequency, or slow down or even eliminate the discharge stage, which will destroy the circuit's anti-power attack performance. Therefore, the present invention proposes a new improved three-stage constant power consumption D flip-flop circuit. Summary of the invention
[0005] The purpose of the present invention is to provide a constant power consumption type D flip-flop circuit based on an improved three-stage logic. Based on the DyCML (dynamic current mode logic) structure, a constant power consumption type D flip-flop based on the DyCML structure is designed, and three-stage logic is introduced to prevent the circuit from being attacked by power consumption due to unbalanced load capacitance. At the same time, improvements are proposed to the three-stage logic, and a discharge signal is generated by the internal nodes of the circuit to prevent the circuit from being attacked by the clock frequency and the like, thereby eliminating the effect of the discharge stage and affecting the overall circuit's anti-power consumption attack performance.
[0006] The power consumption constant type D flip-flop circuit based on improved three-stage logic includes eight PMOS tubes P1-P8 and nine NMOS tubes N1-N9. The circuit can be divided into a middle structure, a left structure and a right structure. The middle includes P1, P2, P3, P4, N1, N2, N3, N4, N5, the left includes N8, P7, N9, P8, and the right includes N6, P5, N7, P6;
[0007] Middle: P1 gate and P4 gate are connected to clk signal to control the on and off of P1 and P4, P1 source and P4 source are connected to VDD, P1 drain and P4 drain are connected to N1 source and N2 source respectively; N1 gate and N2 gate are connected by input logic pair Control, N1 drain and N2 drain are connected to N3 source, while N3 drain is connected to N5 gate, N5 source and drain are connected; TA node is connected to P6 gate, P7 gate, N6 gate, N9 gate, TB node is connected to N8 gate, P8 gate, P5 gate, N7 gate, TC node is connected to N3 drain, N4 source, N5 gate; clk2 signal is connected to N4 gate through inverter to control N4 on and off;
[0008] Left side: N8 source is connected to VDD signal, N8 drain is connected to P7 source, P7 drain is connected to N9 source, N9 drain is connected to P8 source, P8 drain is grounded, and P7 drain is connected to N9 source to output signal Q;
[0009] Right side: N6 source connected to VDD signal, N6 drain connected to P5 source, P5 drain connected to N7 source, N7 drain connected to P6 source, P6 drain connected to ground, and P5 drain connected to N7 source to output signal
[0010] Further, the circuit includes a "precharge-evaluation" stage,
[0011] When clk=0, the circuit is in the pre-charge stage, and the clk signal connects the P1 gate and the P4 gate to control the conduction of P1 and P4, so that the TA node and the TB node are pulled to a high level; since the TA node is connected to the N6 gate, the P6 gate, the P7 gate, and the N9 gate, and the TB node is connected to the N7 gate, the N8 gate, the P5 gate, and the P8 gate, when the TA node and the TB node are both high levels, the output signal Q and the signal remain unchanged;
[0012] When clk = 1, the circuit is in the evaluation stage, and the input logic Connect N1 gate and N2 gate respectively. When N1 is turned on, the high-level signal at the TA node passes through the N1 source, N1 drain, N3 source, N3 drain in turn and is transmitted to the TC node, connected to the N5 gate. At the same time, the N5 source and drain are connected, acting as a capacitor. The signal is transmitted to the TC node to charge N5. At this time, the signals at the TA node and TB node are (0,1), P6 and N7 are turned on, and the signal It is discharged to a low level through the N7 source, N7 drain, P6 source, and P6 drain in sequence.
[0013] Furthermore, the circuit also includes two NMOS tubes N10 and N11, the source of N10 is connected to the TA node, the drain of N10 is grounded, and the gate of N10 is connected to the discharge signal dis to control the conduction and shutdown of N10; the source of N11 is connected to the TB node, the drain of N11 is grounded, and the gate of N11 is connected to the discharge signal dis to control the conduction and shutdown of N11.
[0014] Furthermore, the circuit is divided into a "precharge-evaluation-discharge" phase:
[0015] When clk=0, clk2=0, dis=0, the circuit is in the pre-charging stage, the clk signal connects the P1 gate and the P4 gate, controls the P1 and P4 to be turned on, and pulls the TA node and the TB node to a high level; since the TA node is connected to the N6 gate, the P6 gate, the P7 gate, and the N9 gate, the TB node is connected to the N7 gate, the N8 gate, the P5 gate, and the P8 gate, and the dis signal is connected to the N10 gate and the N11 gate, the TA node and the TB node are both high level, and the output signal Q and the signal remain unchanged;
[0016] When clk=clk2=1,dis=0,the circuit is in the evaluation stage, and the input logic Connect the gates of N1 and N2 respectively. When N1 is turned on, the high-level signal at the TA node is transmitted to the TC node through the N1 source, N1 drain, N3 source, and N3 drain in sequence, and then connected to the N5 gate. The signal is transmitted to the TC node to charge N5. At this time, the signals at the TA node and the TB node are (0,1), P6 and N7 are turned on, and the signal Pass through N7 source, N7 drain, P6 source, P6 drain to GND in sequence;
[0017] When clk=1, clk2=0, dis=1, the circuit is in the discharge stage, N10, N11, and N4 are turned on, the TA node is discharged through the N10 source and N10 drain in sequence, the TB node is discharged through the N11 source and N11 drain in sequence, and the TC node is discharged to a low level through the N4 source and N4 drain in sequence.
[0018] Furthermore, the circuit adds a dual-input NAND gate, the signals at the TA node and the TB node are used as inputs of the NAND gate, and the NAND gate outputs a discharge signal dis.
[0019] Furthermore, the circuit is divided into a "precharge-evaluation-discharge" phase:
[0020] When clk=0, clk2=0, dis=0, the circuit is in the pre-charging stage, the clk signal connects the P1 gate and the P4 gate, controls the P1 and P4 to be turned on, and pulls the TA node and the TB node to a high level; since the TA node is connected to the N6 gate, the P6 gate, the P7 gate, and the N9 gate, the TB node is connected to the N7 gate, the N8 gate, the P5 gate, and the P8 gate, and the dis signal is connected to the N10 gate and the N11 gate, the TA node and the TB node are both high level, and the output signal Q and the signal remain unchanged;
[0021] When clk=clk2=1,dis=0,the circuit is in the evaluation stage, and the input logic Connect N1 gate and N2 gate respectively. When N1 is turned on, the high-level signal at the TA node is transmitted to the TC node through the N1 source, N1 drain, N3 source, and N3 drain in sequence, and then connected to the N5 gate. The signal is transmitted to the TC node to charge N5. At this time, the signals at the TA node and the TB node are (0,1), P6 and N7 are turned on, and the signal Pass through N7 source, N7 drain, P6 source, P6 drain to GND in sequence;
[0022] When clk=1, clk2=0, dis=1, the circuit is in the discharge stage, N10, N11, and N4 are turned on, and the TA node is discharged through the N10 source and N10 drain in turn; the TB node is discharged through the N11 source and N11 drain in turn; the TC node is discharged to a low level through the N4 source and N4 drain in turn; and the discharge signal dis is generated by the TA node and TB node inside the circuit, the TA node and TB node are the inputs of the NAND gate, and the discharge signal dis is output as the NAND gate.
[0023] Compared with the prior art, the present invention has the following advantages: the present invention proposes a D flip-flop based on the DyCML structure, introduces an improved three-stage logic, designs a power-constant D flip-flop structure based on the improved three-stage logic, effectively reduces the average power consumption of the flip-flop and improves the circuit's ability to resist power consumption attacks, while improving the circuit's ability to resist hardware security issues caused by unbalanced load capacitance. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 This is a traditional three-phase dual-rail pre-charging logic structure diagram;
[0025] Figure 2 This is a traditional dynamic current mode logic structure diagram;
[0026] Figure 3 The invention is a constant power consumption D flip-flop circuit based on a DyCML structure;
[0027] Figure 4 The invention provides a constant power consumption D flip-flop circuit based on three-stage logic;
[0028] Figure 5 The structure diagram of the constant power consumption D flip-flop based on three-stage logic of the present invention;
[0029] Figure 6 The structure diagram of the constant power consumption D flip-flop based on the improved three-stage logic of the present invention;
[0030] Figure 7 The invention provides a constant power consumption D flip-flop circuit based on improved three-stage logic;
[0031] Figure 8 It is a timing diagram of the constant power consumption D flip-flop based on the improved three-phase logic of the present invention;
[0032] Fig. 9 It is an experimental test platform based on unbalanced load capacitance of the present invention;
[0033] Fig.10 The invention provides an experimental test platform based on balanced load capacitance. DETAILED DESCRIPTION
[0034] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, a constant power consumption D flip-flop circuit based on improved three-stage logic of the present invention is further described below in conjunction with the accompanying drawings.
[0035] Figure 1 This is the traditional three-phase dual-rail pre-charging logic structure diagram. Figure 2 This is the traditional dynamic current mode logic structure diagram.
[0036] like Figure 3 As shown, a constant power consumption D flip-flop circuit based on improved three-stage logic includes eight PMOS tubes P1-P8 and nine NMOS tubes N1-N9. The circuit can be divided into a middle structure, a left structure and a right structure. The middle includes P1, P2, P3, P4, N1, N2, N3, N4, N5, the left includes N8, P7, N9, P8, and the right includes N6, P5, N7, P6.
[0037] Middle: P1 gate and P4 gate are connected to clk signal to control the on and off of P1 and P4, P1 source and P4 source are connected to VDD, P1 drain and P4 drain are connected to N1 source and N2 source respectively; N1 gate and N2 gate are connected by input logic pair Control, N1 drain and N2 drain are connected to N3 source, while N3 drain is connected to N5 gate, N5 source and drain are connected; TA node is connected to P6 gate, P7 gate, N6 gate, N9 gate, TB node is connected to N8 gate, P8 gate, P5 gate, N7 gate, TC node is connected to N3 drain, N4 source, N5 gate; clk2 signal is connected to N4 gate through inverter to control N4 on and off;
[0038] Left side: N8 source is connected to VDD signal, N8 drain is connected to P7 source, P7 drain is connected to N9 source, N9 drain is connected to P8 source, P8 drain is grounded, and P7 drain is connected to N9 source to output signal Q;
[0039] Right side: N6 source connected to VDD signal, N6 drain connected to P5 source, P5 drain connected to N7 source, N7 drain connected to P6 source, P6 drain connected to ground, and P5 drain connected to N7 source to output signal
[0040] The above circuit includes a "precharge-evaluation" stage: when clk=0, the circuit is in the precharge stage, the clk signal connects the P1 gate and the P4 gate, controls P1 and P4 to be turned on, so that the TA node and the TB node are pulled to a high level; since the TA node is connected to the N6 gate, the P6 gate, the P7 gate, and the N9 gate, and the TB node is connected to the N7 gate, the N8 gate, the P5 gate, and the P8 gate, when the TA node and the TB node are both high levels, the output signal Q and the signal When clk = 1, the circuit is in the evaluation stage, and the input logic Connect N1 gate and N2 gate respectively. When N1 is turned on, the high-level signal at the TA node passes through the N1 source, N1 drain, N3 source, N3 drain in turn and is transmitted to the TC node, connected to the N5 gate. At the same time, the N5 source and drain are connected, acting as a capacitor. The signal is transmitted to the TC node to charge N5. At this time, the signals at the TA node and TB node are (0,1), P6 and N7 are turned on, and the signal It is discharged to a low level through the N7 source, N7 drain, P6 source, and P6 drain in sequence.
[0041] It can be understood that two output circuits are designed based on the DyCML structure, each consisting of 2 NMOS and 2 PMOS tubes, to realize the logic function of the D flip-flop. The circuit "precharge-evaluation" is controlled by the clk clock signal, and the input signal Controlling DyCML structure output Signal, at the same time The signal serves as an input to control the MOS tube of the output circuit to turn on or off, and output the correct D flip-flop logic value.
[0042] like Figure 4 As shown, the circuit Figure 3 On the basis of , two NMOS tubes N10 and N11 are added, the source of N10 is connected to the TA node, the drain of N10 is grounded, and the gate of N10 is connected to the discharge signal dis to control the on and off of N10; the source of N11 is connected to the TB node, the drain of N11 is grounded, and the gate of N11 is connected to the discharge signal dis to control the on and off of N11.
[0043] The circuit is divided into the "precharge-evaluation-discharge" stage: when clk = 0, clk2 = 0, dis = 0, the circuit is in the precharge stage, the clk signal connects the P1 gate and the P4 gate, controls P1 and P4 to turn on, so that the TA node and the TB node are pulled to a high level; because the TA node is connected to the N6 gate, the P6 gate, the P7 gate, and the N9 gate, the TB node is connected to the N7 gate, the N8 gate, the P5 gate, and the P8 gate, and the dis signal is connected to the N10 gate and the N11 gate, the TA node and the TB node are both high, and the output signal Q and the signal When clk=clk2=1, dis=0, the circuit is in the evaluation stage, and the input logic Connect the gates of N1 and N2 respectively. When N1 is turned on, the high-level signal at the TA node is transmitted to the TC node through the N1 source, N1 drain, N3 source, and N3 drain in sequence, and then connected to the N5 gate. The signal is transmitted to the TC node to charge N5. At this time, the signals at the TA node and the TB node are (0,1), P6 and N7 are turned on, and the signal It passes through the N7 source, N7 drain, P6 source, P6 drain to GND in sequence. When clk=1, clk2=0, dis=1, the circuit is in the discharge stage, N10, N11, N4 are turned on, the TA node is discharged through the N10 source and N10 drain in sequence, the TB node is discharged through the N11 source and N11 drain in sequence, and the TC node is discharged to a low level through the N4 source and N4 drain in sequence.
[0044] It can be understood that the three-stage logic is introduced on the basis of the constant power consumption D flip-flop based on the DyCML structure, and the circuit "precharge-evaluation-discharge" stage is controlled by the clk, clk2, and dis clock signals. In the precharge stage, TA and TB are charged to a high level; in the evaluation stage, according to the input signal Output Signal, at the same time The signal is used as the input value to connect the MOS tube gate, control the output circuit MOS tube to turn on or off, and realize the correct D flip-flop logic value. During the discharge phase, the two NMOSs controlled by dis are turned on to discharge the internal nodes of the circuit. Figure 5 The figure shows a three-stage D flip-flop structure diagram. The dotted box shows the internal structure of the circuit. The discharge signal dis is an external signal, which is easily attacked by different means to slow down the arrival time or be eliminated. Therefore, this paper proposes an improved three-stage D flip-flop structure diagram, as shown in Figure 6 As shown, the discharge signal is generated by the internal node of the circuit to avoid direct external connection, which can effectively prevent attackers from eliminating the discharge signal by means of separating the clock and the like.
[0045] like Figure 7 As shown, the circuit Figure 4 A dual-input NAND gate is added on the basis of the NAND gate, and the signals at the TA node and the TB node are used as the inputs of the NAND gate, and the NAND gate outputs a discharge signal dis.
[0046] The circuit is divided into the "precharge-evaluation-discharge" stage: when clk = 0, clk2 = 0, dis = 0, the circuit is in the precharge stage, the clk signal connects the P1 gate and the P4 gate, controls P1 and P4 to turn on, so that the TA node and the TB node are pulled to a high level; because the TA node is connected to the N6 gate, the P6 gate, the P7 gate, and the N9 gate, the TB node is connected to the N7 gate, the N8 gate, the P5 gate, and the P8 gate, and the dis signal is connected to the N10 gate and the N11 gate, the TA node and the TB node are both high, and the output signal Q and the signal When clk=clk2=1, dis=0, the circuit is in the evaluation stage, and the input logic Connect N1 gate and N2 gate respectively. When N1 is turned on, the high-level signal at the TA node is transmitted to the TC node through the N1 source, N1 drain, N3 source, and N3 drain in sequence, and then connected to the N5 gate. The signal is transmitted to the TC node to charge N5. At this time, the signals at the TA node and the TB node are (0,1), P6 and N7 are turned on, and the signal It passes through the N7 source, N7 drain, P6 source, P6 drain to GND in sequence. When clk=1, clk2=0, dis=1, the circuit is in the discharge stage, N10, N11, N4 are turned on, the TA node is discharged through the N10 source and N10 drain in sequence; the TB node is discharged through the N11 source and N11 drain in sequence; the TC node is discharged to a low level through the N4 source and N4 drain in sequence; and the discharge signal dis is generated by the TA node and TB node inside the circuit, the TA node and TB node are the inputs of the NAND gate, and the discharge signal dis is output as the NAND gate.
[0047] It can be understood that in the improved three-stage logic structure, in order to prevent attackers from eliminating the discharge stage and causing hardware security problems to the circuit, the dis signal is generated by the TA and TB nodes inside the circuit, TA and TB are the inputs of the NAND gate, and the dis signal is output by the NAND gate. That is, based on the D flip-flop based on the three-stage structure, the discharge signal dis is generated by the nodes inside the circuit, and an additional double-input NAND gate is added. The TA node and TB node inside the circuit are the input signals of the NAND gate, and the discharge signal dis is output to cancel the external discharge signal; the attacker cannot directly separate the discharge signal, preventing the attacker from eliminating the discharge signal by means such as separating the clock, thereby affecting the safety of the circuit.
[0048] In summary, in the circuit of the present invention, the sources of P1 to P4 are connected to VDD, the gates of P1 and P4 are connected to the clk signal to control the on and off of the PMOS tube, the drains of P1 and P2 are connected to the source of N1, the drains of P3 and P4 are connected to the source of N2, the gate of P2 is connected to the drains of P3, P4 and the source of N2 as the TB node, the TB node is connected to the gates of N8, P8, P5 and N7; the gate of P3 is connected to the drains of P1, P2 and the source of N1 as the TA node, the TA node is connected to the gates of P6, P7, N6 and N9; the gates of N1 and N2 are connected by the input logic pair Control, the drains of N1 and N2 are connected to the source of NMOS tube N3, while the drain of N3 is connected to the gate of N5, and the source of N5 is connected to the drain, and its logic function plays the role of a capacitor; the drain of N3 is connected to the source of N4, and the gate of N5 is connected to the connection between the drain of N3 and the source of N4, and the connection point is set as the TC node; the source of N4 is connected to the TC node, and clk is connected to the gate of N4 through an inverter to control the on and off of the MOS tube. The TA node is connected to the gates of P6, P7, N6, and N9, and the TB node is connected to the gates of N8, P8, P5, and N7. The source of N8 on the left side is connected to the VDD signal, the drain is connected to the source of P7, the drain of P7 is connected to the source of N9, the drain of N9 is connected to the source of P8, and the drain of P8 is grounded. The structures of N6, P5, N7, and P6 on the right side are obtained in the same way.
[0049] like Figure 8 Shown is a timing diagram of the constant power consumption D flip-flop based on the improved three-phase logic of the present invention.
[0050] like Fig. 9 and Fig.10 The figure shows the experimental environment platform used by the constant power consumption D flip-flop based on the improved three-stage logic of the present invention. Under the condition of 65nmCMOS, the voltage source is 1.5V, the input clock frequency is 500MHz, and the output loads are unbalanced 10uF and 1000uF capacitors and balanced load capacitors 10uF and 10uF.
[0051] The power consumption constant type D flip-flop based on improved three-stage logic designed by the present invention realizes low latency and low power consumption, and improves the circuit's resistance to power consumption attacks. NED (normalized energy deviation) and NSD (normalized energy variance) are used to measure the resistance to power consumption attacks of the comparison scheme and the proposed structure. The smaller the NED and NSD parameters are, the higher the level of resistance to power consumption attacks is. The proposed improved three-stage D flip-flop is compared with other schemes. Fig. 9 , Fig.10 In the experimental environment, the measured values are shown in Table 1 and Table 2.
[0052] Table 1 Comparison of experimental results of triggers under balanced load capacitance
[0053]
[0054] Table 2 Comparison of experimental results of triggers under unbalanced load capacitance
[0055]
[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A constant power consumption D flip-flop circuit based on improved three-stage logic, characterized in that: It includes eight PMOS tubes P1 to P8 and nine NMOS tubes N1 to N9. The circuit can be divided into a middle structure, a left structure and a right structure. The middle structure includes P1, P2, P3, P4, N1, N2, N3, N4, N5, the left structure includes N8, P7, N9, P8, and the right structure includes N6, P5, N7, P6; Middle: P1 gate and P4 gate are connected to clk signal to control the on and off of P1 and P4, P1 source and P4 source are connected to VDD, P1 drain and P4 drain are connected to N1 source and N2 source respectively; N1 gate and N2 gate are controlled by input logic pair (D, D), N1 drain and N2 drain are both connected to N3 source, while N3 drain is connected to N5 gate, N5 source and drain are connected; TA node is connected to P6 gate, P7 gate, N6 gate, N9 gate, TB node is connected to N8 gate, P8 gate, P5 gate, N7 gate, TC node is connected to N3 drain, N4 source, N5 gate; clk2 signal is connected to N4 gate through inverter to control the on and off of N4; Left side: N8 source is connected to VDD signal, N8 drain is connected to P7 source, P7 drain is connected to N9 source, N9 drain is connected to P8 source, P8 drain is grounded, and P7 drain is connected to N9 source to output signal Q; Right side: N6 source is connected to VDD signal, N6 drain is connected to P5 source, P5 drain is connected to N7 source, N7 drain is connected to P6 source, P6 drain is grounded, and P5 drain is connected to N7 source to output signal Q.
2. The constant power consumption D flip-flop circuit based on improved three-phase logic according to claim 1, characterized in that: The circuit includes a pre-charge phase and an evaluation phase, When clk=0, the circuit is in the pre-charging stage, and the clk signal connects the P1 gate and the P4 gate to control P1 and P4 to be turned on, so that the TA node and the TB node are pulled to a high level; since the TA node is connected to the N6 gate, the P6 gate, the P7 gate, and the N9 gate, and the TB node is connected to the N7 gate, the N8 gate, the P5 gate, and the P8 gate, when the TA node and the TB node are both at a high level, the output terminal signals Q and Q remain unchanged; When clk=1, the circuit is in the evaluation stage, and the input logic pair (D, D) is connected to the gate of N1 and the gate of N2 respectively. When (D, D)=(1,0), N1 is turned on, and the high-level signal at the TA node passes through the N1 source, N1 drain, N3 source, N3 drain in turn and is transmitted to the TC node, connected to the N5 gate. At the same time, the N5 source and drain are connected, acting as a capacitor. The signal is transmitted to the TC node to charge N5. At this time, the signals at the TA node and TB node are (0,1), P6 and N7 are turned on, and the signal Q passes through the N7 source, N7 drain, P6 source, and P6 drain in turn to discharge to a low level.
3. The constant power consumption D flip-flop circuit based on improved three-phase logic according to claim 1, characterized in that: The circuit also includes two NMOS tubes N10 and N11, wherein the source of N10 is connected to the TA node, the drain of N10 is grounded, and the gate of N10 is connected to the discharge signal dis to control the on and off of N10; the source of N11 is connected to the TB node, the drain of N11 is grounded, and the gate of N11 is connected to the discharge signal dis to control the on and off of N11.
4. The constant power consumption D flip-flop circuit based on improved three-phase logic according to claim 3, characterized in that: The circuit is divided into "precharge-evaluation-discharge" stages: When clk=0, clk2=0, dis=0, the circuit is in the pre-charging stage, the clk signal connects the P1 gate and the P4 gate, controls P1 and P4 to be turned on, so that the TA node and the TB node are pulled to a high level; since the TA node is connected to the N6 gate, the P6 gate, the P7 gate, and the N9 gate, the TB node is connected to the N7 gate, the N8 gate, the P5 gate, and the P8 gate, and the dis signal is connected to the N10 gate and the N11 gate, the TA node and the TB node are both high levels, and the output terminal signals Q and Q remain unchanged; When clk=clk2=1, dis=0, the circuit is in the evaluation stage, and the input logic pair (D, D) is connected to the gates of N1 and N2 respectively. When (D, D)=(1,0), N1 is turned on, and the high-level signal at the TA node is transmitted to the TC node through the N1 source, N1 drain, N3 source, and N3 drain in sequence, and then connected to the N5 gate. The signal is transmitted to the TC node to charge N5. At this time, the signals at the TA node and the TB node are (0,1), P6 and N7 are turned on, and the signal Q passes through the N7 source, N7 drain, P6 source, and P6 drain in sequence to GND; When clk=1, clk2=0, dis=1, the circuit is in the discharge stage, N10, N11, and N4 are turned on, the TA node is discharged through the N10 source and N10 drain in sequence, the TB node is discharged through the N11 source and N11 drain in sequence, and the TC node is discharged to a low level through the N4 source and N4 drain in sequence.
5. The constant power consumption D flip-flop circuit based on improved three-phase logic according to claim 3, characterized in that: The circuit is added with a double-input NAND gate, the signals at the TA node and the TB node are used as inputs of the NAND gate, and the NAND gate outputs a discharge signal dis.
6. The constant power consumption D flip-flop circuit based on improved three-phase logic according to claim 5, characterized in that: The circuit is divided into "precharge-evaluation-discharge" stages: When clk=0, clk2=0, dis=0, the circuit is in the pre-charging stage, the clk signal connects the P1 gate and the P4 gate, controls P1 and P4 to be turned on, so that the TA node and the TB node are pulled to a high level; since the TA node is connected to the N6 gate, the P6 gate, the P7 gate, and the N9 gate, the TB node is connected to the N7 gate, the N8 gate, the P5 gate, and the P8 gate, and the dis signal is connected to the N10 gate and the N11 gate, the TA node and the TB node are both high levels, and the output terminal signals Q and Q remain unchanged; When clk=clk2=1, dis=0, the circuit is in the evaluation stage, the input logic pair (D, D) is connected to the gate of N1 and the gate of N2 respectively, when (D, D)=(1,0), N1 is turned on, the high-level signal at the TA node is transmitted to the TC node through the N1 source, N1 drain, N3 source, and N3 drain in sequence, and then connected to the gate of N5. The signal is transmitted to the TC node to charge N5. At this time, the signals at the TA node and the TB node are (0,1), P6 and N7 are turned on, and the signal Q passes through the N7 source, N7 drain, P6 source, and P6 drain in sequence to GND; When clk=1, clk2=0, dis=1, the circuit is in the discharge stage, N10, N11, and N4 are turned on, and the TA node is discharged through the N10 source and N10 drain in turn; the TB node is discharged through the N11 source and N11 drain in turn; The TC node is discharged to a low level through the N4 source and the N4 drain in turn; and the discharge signal dis is generated by the TA node and the TB node inside the circuit. The TA node and the TB node are the inputs of the NAND gate, and the discharge signal dis is output as the NAND gate.
Citation Information
Cited By
Single event upset resistant reinforced CML latch, frequency-halving circuit and equipment
CN120567104A
Cml latch, divide-by-two circuit, and device hardened against single event upsets
CN120567104B
