SUPERCONDUCTING CIRCUIT AND METHOD FOR PEAK TIMING DETECTION OF SINGLE PHOTON DETECTOR SIGNALS
The bidirectional SQUID quantizer topology in the superconducting circuit addresses amplitude-dependent timing errors in SNSPDs by detecting the zero slope point, enabling reliable and scalable photon detection without room temperature signal transfer.
Patent Information
- Application Number
- TR202603093
- Authority / Receiving Office
- TR · TR
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2026-03-02
- Publication Date
- 2026-10-21
AI Technical Summary
Existing superconducting nanowire single-photon detectors (SNSPDs) face timing drift issues due to amplitude-dependent timing errors, which are exacerbated by the need to transfer signals from cryogenic to room temperature, leading to increased jitter and complexity.
A superconducting circuit using a bidirectional SQUID quantizer topology determines the peak moment of input pulses by detecting the zero slope point, eliminating the need for threshold-based triggering and reducing signal transmission to room temperature.
This approach provides amplitude-independent timing with reduced jitter and complexity, facilitating scalable and integrated photon detection systems without additional room temperature components.
Smart Images

Figure 00000045_0000 
Figure 00000045_0001 
Figure 00000046_0000
Abstract
Description
1 TARIFF PEAK TIMING DETECTION OF SINGLE PHOTON DETECTOR SIGNALS SUPERCONDUCTING CIRCUIT AND METHOD FOR Technical Area This invention provides a 5% improvement over signal timing separation generally achieved with superconducting electronic circuits. It is related. More specifically, the invention involves single-photon detectors (SNSPDs) made of superconducting nanowires. A SQUID is used to determine the peak moment of short pulses from photon detectors. This relates to the circuit based on and related methodology. In this context, the circuit involves the rise and fall of the input pulse. using the transition between Single Flux Quantum (SFQ) pulses generated during the fall phases Detects the zero slope (peak) point of the pulse. 10 State of the Art Superconducting nanowire single-photon detectors (SNSPDs) offer high detection efficiency and low darkness. High count rate and low timing jitter values for single photon detection. These are detectors that provide performance [1][2][3]. These detectors operate at cryogenic temperatures. quantum key distribution (QKD) systems [2], photonic quantum computing 15 Basic components in applications such as architectures and optical communication in space environment [4] It is among them. An SNSPD produces an electrical output pulse when a single photon is absorbed. This output pulse... The amplitude and shape of the pulse determine the polarization current state of the nanowire at the moment of photon absorption. Depending on the photon energy and the detector's recovery status from previous detection events, 20 It can vary [1][5]. At high photon count rates, the detector between successive events It may not switch completely and this leads to low amplitude output pulses [5]. Low-amplitude pulses lead to a later trigger threshold exceedance in threshold-based discriminators. Opening it can increase timing errors caused by time drift. A common method for timing determination is to set the SNSPD output pulse to a specific voltage or current. It involves the use of threshold-based discriminators that are triggered when the threshold is exceeded. However, SNSPD Since the amplitude of the pulses can vary, the moment when the pulse crosses the same threshold also changes, and this situation This creates an amplitude-dependent timing error ("time drift"). As a result, low amplitude A pulse threshold can be crossed later, and different triggers can be used for photon events occurring simultaneously. The timings may vary. This amplitude-dependent error increases the timing jitter, making detection difficult. 30 It reduces the overall timing performance of the system [5]. 2 One of the known established solutions for obtaining amplitude-independent timing is the constant fraction. is the discriminator (CFD) [4][6]. In conventional CFD, the input signal is split into two arms: one of the arms is The signal is attenuated to a certain extent on one arm, while on the other arm, the signal is delayed for a certain period of time. When the attenuated signal is inverted and combined with the delayed signal, a bipolarized system is formed. The signal is obtained, and the zero-crossing moment of this signal is used as the timing point. The 5 in question... The zero-crossing point is a similar pattern on the rising edge of the pulse, even if the signal amplitude changes. Since it corresponds to position, the CFD approach provides amplitude-independent timing. It is suitable. However, traditional constant fraction discriminators (CFDs) are not suitable for semiconductor circuits that operate at room temperature. However, it is not suitable for direct operation at cryogenic temperatures. Therefore, SNSPDs and 10 When used together, the output signal produced by the detector in the cryogenic layer can be converted to CFD. It must be transmitted to room temperature via coaxial cables and amplifiers in order to reach room temperature. This is necessary. The transmission path in question involves cable attenuation, impedance mismatches, and thermal noise. and due to effects such as amplifier bandwidth limitations, signal distortions and additional This can lead to timing jitter. In the literature, optimized CFD 15 at room temperature Even though low jitter values can be obtained with cryogenic-room temperature circuits, the signal... It has been reported that the path constitutes a significant performance limitation [7]. In SNSPDs operating at cryogenic temperatures, the problem arises from transferring the signal to room temperature. To reduce distortions, a superconducting readout that can operate in the same cryogenic environment as the detector is used. circuits have been developed for this purpose, which convert the SNSPD output into superconducting digital pulses. Single Flux Quantum (SFQ) based circuits have been proposed. For example, circuits using a DC / SFQ converter. read circuits [8]
[28] , SFQ comparator-based timing discriminators [9] and multi-channel Multiplexing architectures have been developed for systems
[10]
[11]
[29] . In addition, different scalability For their targets, cryotron-based readout
[13] , cryogenic AQFP / RSFQ hybrid interfaces
[12] CMOS-based ASIC solutions operating at temperatures
[14] , photonic readout techniques
[14] and 25 Alternative reading approaches such as cryogenic opto-electronic interfaces
[15] have also been investigated. Various architectures related to SNSPD reading have also been discussed in patent publications.
[16]
[17]
[18]
[19] . All previous SFQ-based read circuits used threshold-based triggering mechanisms. It uses, for example, a DC / SFQ converter or SFQ comparator, where the input current is a certain 30V. It generates an SFQ pulse when a threshold value is exceeded. Such circuits allow the signal to pass through the cryogenic layer. 3 by reducing the need to bring it to room temperature, it could eliminate the long cable route. However, the triggering moment depends on the amplitude of the input pulse. Therefore, a low-amplitude SNSPD pulse may cross the threshold later, thus preventing threshold-based response. The amplitude-dependent time shift effect observed in the discriminators may persist. Bidirectional SFQ Topology-based read cells
[30] and digital SQUID designs
[31] are also in this context. These approaches have been examined and are also based on the threshold-based triggering principle. Software-based correction approaches are also used to reduce the time shift effect. It has been suggested [5]. These approaches reduce timing delays with inter-arrival times. It is based on the principle of estimating and compensating for time displacement by correlating it. This Although such approaches can reduce the tails in the jitter distribution in some cases, detection 10 It does not provide real-time amplitude-independent timing at this point and at room temperature. This requires additional digital processing. Furthermore, since the correction is statistical in nature, it affects the single photon. It directly eliminates the amplitude-dependent fundamental timing error in terms of events. It does not remove it. A superconducting circuit topology associated with the present invention is, for the first time, a superconducting analog-to-digital 15 It is a bidirectional two-junction SQUID quantizer defined in the context of adverter converters (ADCs).
[20] . In this topology, a two-junction SQUID ring; magnetic flux quantum (Φ0) in the ring When the magnetic single flux quantum increases, one junction switches, and when it decreases, the other junction switches. It is designed to generate an SFQ pulse by doing so. The aforementioned bidirectional quantizer, 20 experimentally implemented with reversible counting structures in monitoring ADC architectures
[21] ; the history and development of topology have been discussed in detail in various studies
[22] -
[23] . Use of bidirectional SFQ structures as read cells and digital SQUID optimization has also been investigated
[30]
[31] . Recently, the same bidirectional two-joint SQUID Quantizer topology, an analog-to-digital converter for high-frequency signal acquisition. (ADC) has been experimentally demonstrated
[33] . In the said study, two Josephson 25 The SFQ pulses produced by the joint during its positive and negative tilt phases are asynchronous toggle flip- The input signal is counted via floppy disk (TFF) counters and synchronous cumulative adders. Amplitude information has been digitized. Majority voting and error correction were used in the study. Mechanisms have also been applied. This study examines the role of topology in high-frequency signal acquisition. Although we have experimentally confirmed that it is functional, the 30 between SQUID outputs He did not propose using temporal transition as a timing reference, but rather amplitude. It did not define the concept of independent limited timing error and described the circuit as an open loop. 4 It has not been evaluated in a (counterless) configuration. It uses SQUID-based quantization. There are also patent publications relating to superconducting ADC circuits
[24]
[25] . Previous In applications, the bidirectional SQUID quantizer is primarily used in analog-to-digital converters (ADCs). It has been used as a component in their architectures. In this context, for increasing flux and decreasing flux. Two generated SFQ pulse sequences are applied to an up / down counter, and the resulting cumulative 5 The count represents the digitized amplitude value of the input signal. In other words... The goal is to convert an analog waveform into a digital value representing its amplitude. (Previous) the technique, including recent experimental verification
[33] its subject is the timing reference of the transition point between two series of coups. We do not recommend its use; furthermore, we do not recommend using a bidirectional SQUID topology with a single-photon detector. an alternative approach to reading, timing discrimination or constant fraction discrimination It does not implement it as such. Of particular note is the fact that the applicant is listed as a co-author
[33] His work experimentally generated, tested, and Although he published it, he neither proposed nor discussed the potential for a time separation. 15 The article in question uses the terms "timing" and "time-walk". "jitter", "peak detection" or "timing" There are no references to the concepts of "discrimination" in the application section of the article. It uses only the phrase "cryogenic computing and sensing applications". Application The owner is among those involved in both SFQ circuit design and SNSPD reading 20 timing functionality of the research group with direct experience in their fields The fact that it hasn't been defined is a strong objective indication that this usage isn't obvious to an expert in the field. It is an indicator. All previous SFQ-based readout circuits developed for SNSPDs were DC / SFQ converters. circuits using [8]
[28] , SFQ comparator-based circuits [9]
[10]
[11] and address encoders 25
[29] including threshold-based triggering mechanisms. The SQUID quantizer topology was first described by Lee and Petersen
[20] in 1989. defined and in the 36 years that have passed since then, Lee and Petersen
[20] , Fillipenko et al.
[21] , Mukhanov et al.
[22]
[23] , Jabbari et al.
[30] , Foroughi et al.
[31] and Ucpinar et al.
[33] included Independent research groups, including those mentioned above, have studied this topology using various ADCs and 30 It has been used in digitization applications. Notably, the applicant has used this He was a co-author in five of the studies (
[28] ,
[29] ,
[30] ,
[31] ,
[33] ), None of the studies have identified the potential for time separation of the topology. This 36 In the annual literature, it is suggested that this topology can be used for timing discrimination purposes. or there are no sources indicating that this is possible; whereas topology The SNSPD time shift problem exists and has been documented in the literature since at least 2009 [8]. was the case [5]. Similarly, the SNSPD has been reading for over 15 years since 2009 [8]. In the same period, 5 such as Ortlepp et al. [8], Terai et al. [9], Miyajima et al.
[10]
[11] and Takeuchi et al.
[12] Independent groups consistently preferred threshold-based approaches, A quantization-based timing detection approach has not been proposed. Directly related to topology. This includes applicants with experience, as well as researchers in both fields. The consistent choice and the combination of expertise in the relevant technical field in the present invention are evident. This is an objective indication that they will not find it. This situation shows that experts in the field believe the SQUID quantizer 10 he considered its topology only as an amplitude digitization tool and timing detection a technical bias indicating that it did not foresee that it could be used for that purpose It reflects prejudice. Quantum key distribution (QKD) systems and photonic quantum computing architectures at scale. And with increasing performance requirements, minimized timing jitter 15 The need for cryogenic and scalable SNSPD readout solutions is growing. is growing. As these applications mature, threshold-based SNSPD readout circuits Amplitude-dependent timing error (time drift) stands out as a critical performance limitation. Software-based correction approaches [5] only partially address this problem. This can eliminate the need for additional room temperature processing steps. The present invention addresses these 20 It addresses the limitation with a fundamentally different approach: replacing threshold-based triggering with, Peak timing detection is performed entirely on a cryogenic substrate. This allows for QKD (Quick Detection and Control). and becoming increasingly evident with the scaling of photonic quantum computing systems Timing performance requirements without the need for additional room temperature components. can be met. 25 Therefore, a timing system with the following characteristics is needed, especially in terms of SNSPD output pulses. The need for a separation circuit continues: (a) pulse amplitude independent timing. (b) ability to operate at cryogenic temperatures together with SNSPD, (c) timing (d) random (e) 30 circuit physics and peaks, without relying on comparator threshold level or signal-to-noise ratio. It can offer limited low timing uncertainty depending on the signal behavior in its vicinity. The present invention It meets this need. 6 SOURCES [1] CM Natarajan, MG Tanner, and RH Hadfield, “Superconducting nanowire single- photon detectors: physics and applications," Supercond. Sci. Technol., vol. 25, no. 6, p. 063001, 2012. [2] B. Korzh et al., “Demonstrating sub-3 ps temporal resolution in a superconducting nanowire 5 single-photon detector," Nature Photonics, vol. 14, pp. 250-255, 2020. [3] D. V. Morozov, A. Casaburi, and R. H. Hadfield, "Superconducting photon detectors," in Contemporary Physics, 2021. [4] E. E. Wollman, J. P. Allmaras, A. D. Beyer, B. A. Korzh, M. C. Runyan, L. Narvaez, W. H. Farr, F. Marsili, R. M. Briggs, G. J. Miles, and M. D. Shaw, "SNSPD-based detector system 10 for NASA's Deep Space Optical Communications project," Opt. Express, vol. 32, no. 27, pp. 48185-48198, 2024. [5] A. Mueller et al., "Time-walk and jitter correction in SNSPDs at high count rates," Appl. Phys. Lett., vol. 122, no. 4, p. 044001, Jan. 2023. [6] ORTEC, "Choosing the Right Timing Discriminator for the Application," Application Note. 15 [7] I. Esmaeil Zadeh, J. W. N. Los, R. B. M. Gourgues, V. Steinmetz, G. Bulgarini, S. M. Dobrovolskiy, V. Zwiller, and S. N. Dorenbos, "Efficient Single-Photon Detection with 7.7 ps Time Resolution for Photon-Correlation Measurements," ACS Photonics, vol. 7, no. 7, pp. 1780-1787, 2020. DOI: 10.1021 / acsphotonics.0c00433. [8] T. Ortlepp, M. Hofherr, L. Fritzsch, S. Engert, K. Ilin, D. Rall, H. Toepfer, H.-G. Meyer, 20 and M. Siegel, "Demonstration of digital readout circuit for superconducting nanowire single photon detector," Opt. Express, vol. 19, no. 19, pp. 18593-18601, 2011. [9] H. Terai, T. Yamashita, S. Miki, K. Makise, and Z. Wang, "Low-jitter single flux quantum signal readout from superconducting single photon detector," Opt. Express, vol. 20, no. 18, pp. 20115-20123, 2012. 25
[10] S. Miyajima, M. Yabuno, S. Miki, T. Yamashita, and H. Terai, "Current sensitivity enhancement of a quasi-one-junction SQUID comparator as an input circuit of SFQ readout circuit for a superconducting detector," J. Low Temp. Phys., vol. 176, pp. 465-470, 2014. DOI: 10.1007 / s10909-014-1119-3. 7
[11] S. Miyajima, M. Yabuno, S. Miki, T. Yamashita, and H. Terai, "High-time-resolved 64- channel single-flux quantum-based address encoder integrated with a multi-pixel superconducting nanowire single-photon detector," Opt. Express, vol. 26, no. 22, pp. 29045- 29054, 2018.
[12] N. Takeuchi et al., "Scalable readout interface for superconducting nanowire single-photon 5 detectors using AQFP and RSFQ logic families," Opt. Express, vol. 28, no. 8, pp. 12047-12057, 2020.
[13] K. Zheng, Q. Zhao, H. Lu, L. Zhang, L. Kang, J. Chen, and P. Wu, "Characterize the switching performance of a superconducting nanowire cryotron for reading superconducting nanowire single photon detectors," Sci. Rep., vol. 9, art. 16345, 2019. DOI: 10.1038 / s41598-10 019-52874-3.
[14] Oak Ridge National Laboratory / Fermilab, "A 32-Channel Cryo-CMOS ASIC for SNSPD Biasing and Readout with Picosecond Timing," 2024.
[15] M. de Cea, E. E. Wollman, A. H. Atabaki, M. D. Shaw, and R. J. Ram, "Photonic readout of superconducting nanowire single photon counting detectors," Sci. Rep., vol. 10, art. 9500, 15 2020. DOI: 10.1038 / s41598-020-65971-5.
[16] L. Gao et al., "Interfacing superconducting nanowire single photon detectors with cryogenic opto-electronics for quantum photonic applications," arXiv:2501.08125, Jan. 2025.
[17] S. N. Dorenbos, V. G. Zwiller, and R. N. Schouten, "Read-out of superconducting single photon detectors," Patent NL2003572C2, Mar. 30, 2011. 20
[18] D. Zhu, M. Colangelo, B. Korzh, M. Shaw, and K. K. Berggren, "Impedance matched superconducting nanowire photodetector for single- and multi-photon detection," U.S. Patent US11522115B2, Dec. 6, 2022.
[19] "Device and system for single photon detection using a plurality of superconducting detection means connected in parallel," U.S. Patent US11988551, May 21, 2024. 25
[20] G. S. Lee and D. A. Petersen, "Superconductive A / D converters," Proc. IEEE, vol. 77, no. 8, pp. 1264-1273, Aug. 1989.
[21] L. V. Fillipenko, S. V. Rylov, O. A. Mukhanov, V. K. Semenov, and K. K. Likharev, "Experimental implementation of analog-to-digital converter based on the reversible ripple counter," IEEE Trans. Magn., vol. MAG-27, no. 2, pp. 2464-2467, Mar. 1991. 30 8
[22] O. A. Mukhanov, "History of Superconductor Analog-to-Digital Converters," in 100 Years of Superconductivity, H. Rogalla and P. Kes, Eds. London, UK: Taylor & Francis, 2011, pp. 440-458.
[23] O. A. Mukhanov, D. Gupta, A. M. Kadin, and V. K. Semenov, "Superconductor analog- to-digital converters," Proc. IEEE, vol. 92, no. 10, pp. 1564-1584, Oct. 2004. 5
[24] "Superconducting analog-to-digital converter and digital magnetometer and related method for its use," European Patent EP0222470A2, 1986. (TRW Inc.)
[25] "High-performance superconducting digital-to-analog converter employing double- junction SQUID flip-flops," U.S. Patent US5398030A.
[26] R. L. Fagaly, "Superconducting quantum interference device instruments and 10 applications," Rev. Sci. Instrum., vol. 77, 101101, 2006.
[27] J. Clarke and A. I. Braginski, The SQUID Handbook. Weinheim: Wiley-VCH, 2004.
[28] A. Bozbey, S. Miyajima, H. Akaike, and A. Fujimaki, "Single-flux-quantum circuit based readout system for detector arrays by using time to digital conversion," IEEE Trans. Appl. Supercond., vol. 19, no. 3, pp. 509-513, 2009. 15
[29] A. Bozbey, E. C. Aydogan, K. Usenmez, S. Razmkhah, M. Tanaka, and A. Fujimaki, "Synchronous SFQ address encoder for superconductor detector arrays," IEEE Trans. Appl. Supercond., vol. 31, no. 5, pp. 1-1, 2021.
[30] T. Jabbari, H. Zandi, F. Foroughi, A. Bozbey, and M. Fardmanesh, "Investigation of readout cell configuration and parameters on functionality and stability of bi-directional RSFQ 20 TFF," IEEE Trans. Appl. Supercond., vol. 26, no. 3, pp. 1-5, 2016. DOI: 10.1109 / TASC.2016.2517145.
[31] F. Foroughi, A. Bozbey, and M. Fardmanesh, “Design and optimization of fully digital SQUID based on bi-directional RSFQ," J. Supercond. Novel Magn., vol. 27, no. 7, pp. 1623- 1628, 2014. DOI: 10.1007 / s10948-014-2501-8. 25
[32] JA Delport, K. Jackman, P. le Roux, and CJ Fourie, “JoSIM - Superconductor SPICE Simulator," IEEE Trans. Appl. Supercond., vol. 29, no. 5, art. 1300905, Aug. 2019. DOI: 10.1109 / TASC.2019.2897312. 9
[33] BZ Ucpinar, S. Razmkhah, MA Karamuftuoglu and A. Bozbey, "Experimental "Demonstration of a Superconductor SFQ-Based ADC for High-Frequency Signal Acquisition," Supercond. Sci. Technol., 2025. arXiv:2510.14191v1. Purpose of the Invention In the previous technique, the SNSPD timing drift problem consistently reached "threshold level 5". This has been addressed within the framework of "optimization": setting a fixed threshold level, impedance. such as matching methods or the use of constant fraction discriminant (CFD) at room temperature [5][6][7]. The present invention addresses the problem in question in a fundamentally different way. The framework formulates it as follows: instead of optimizing the threshold level, the peak of the input signal... Using the point (the moment of zero slope) as a direct timing reference. This 10 The reformulation leads to an entirely different family of solutions (flux quantization-based timing). (detection) opens the door, and this frame was not suggested in the previous technique. The aim of the present invention is to define a timing mechanism that describes the peak moment of a transient input signal pulse. A superconducting timing separation circuit and related methods for obtaining the reference and to provide a system. For this purpose, the invention adapts the known bidirectional SQUID topology for timing 15 It uses a new way for the separation. Accordingly, in the rising phase of the input pulse (320) The transition between the impact generation occurring at the exit, the falling phase (340) and the other exit is the basis. By taking this, the zero slope (peak) point (330) of the input signal pulse is determined. This reduces the time-shifting effect caused by changes in pulse amplitude. This allows for a more reliable and repeatable determination of photon arrival time. This is ensured. By simplifying the timing chain, a distinctive timing system used at room temperature is achieved. The need for additional signal processing components can be reduced. Furthermore, the system architecture... simplifying the implementation by reducing cabling and interconnection requirements. This helps to reduce complexity and associated costs. In previous implementations of the bidirectional SQUID topology, the aim was to represent the amplitude of the input signal. While the aim in the present invention is to count the cumulative SFQ pulses generated in the first output, the aim in the current invention is to count the SFQ pulses generated in the first output. It appears as the cessation of pulse generation and the commencement of SFQ pulse generation in the second exit phase. The goal is to detect the resulting temporal transition. This transition refers to the change in the input signal pulse. It corresponds to the point where its velocity passes through zero, i.e., the peak moment of the pulse. The peak moment of the pulse is the amplitude. Even if it changes, it expresses the same relative temporal position on the impact pattern, so this transition is 30 A timing reference based on this can be produced in the cryogenic layer. Thus, timing determination can be made. Its architecture can be simplified and long signal transmission can be achieved with additional room temperature timing electronics. The need for lines can be reduced. Additionally, multi-channel or array-based detectors can be used. Integration and scalability can be facilitated in applications. The circuit is suitable for asynchronous operation without requiring an external clock signal and is one of the SNSPDs. It is suitable for detecting photon events that occur at random times. The circuit consists of an SNSPD and a It can be located in the same cryogenic temperature stage together with the timing capture circuit; 5 Thus, the need to transmit the signal to room temperature electronics for timing determination. It eliminates this. In this context, the invention applies to constant fraction distinguishers used at room temperature. It provides a cryogenic alternative and detects the peak by zeroing out a generated bipolar signal. Instead of relying on the transition itself, SQUID accomplishes this through the transition between outputs. This In this way, the scheduling chain is simplified, resulting in low latency and low distortion. A workable architecture can be obtained; long signal with additional discrimination / processing components. Integration and scalability by reducing dependence on transmission lines. It can be made easier. Brief Description of the Invention This invention is one of 15 single-photon detectors such as the superconducting nanowire single-photon detector (SNSPD). the peak moment of an incoming transient input pulse, the temporal interval between successive SFQ pulses by providing adjustable timing resolution based on changes in pulse amplitude by limiting the resulting timing errors with a Φ0 / M flux quantization step. It includes a circuit capable of determining this, along with the related methods and systems. Within this scope, the circuit... a superconductor 20 magnetically coupled to the input pulse and containing two Josephson junctions It includes a quantum interference device (SQUID) ring. While the circuit is running; during the rise phase of the input pulse (320) at the first output, the fall phase of the pulse In the second stage, Single Flux Quantum (SFQ) pulses are generated at the output. The input pulse... Since the rise turned into a fall at the peak, the SFQ impulse activity in the first breakout ended. and the temporal transition that occurs as the initiation of SFQ pulse activity in the second output, peak 25 It creates a reliable time reference showing the moment. The invention includes a circuit for applications where photon events occur at random times. It is suitable for asynchronous operation without requiring an external clock signal. Furthermore, the circuit is suitable for SNSPD (Synchronous No- ... Thanks to its ability to be positioned in the cryogenic temperature stage, it can be stored at room temperature. 30 The need can be reduced or eliminated. The invention relates to quantum switch distribution and photonics. It is applicable in fields such as quantum computing. 11 Explanation of the Figures The invention, when considered together with the detailed description below and the accompanying drawings... It will be better understood through this: FIGURE 1 shows a conceptual illustration of the timing jitter (time shift) problem. The diagram shows a fixed-threshold discriminant with different amplitudes for SNSPD output pulses. how it is triggered at certain times and how peak (zero slope) detection is used as a timing reference This shows that it is being used. FIGURE 1(a) shows the fixed threshold of SNSPD output pulses of different amplitudes in a discriminant. The triggering moments differ because the threshold level Vth is exceeded at different times, and This shows that the time shift is represented by Δt1 and Δt2. FIGURE 1(b) shows 10 If the timing point is determined based on the peak moment (t_peak) for both pulses, the timing reference can be obtained independently of changes in pulse amplitude This shows that amplitude-dependent time shift can be reduced in this way. FIGURE 2 is a conceptual diagram of a double-output bidirectional SQUID ring (110), the first The output (OUTPUT1, 130) generates SFQ pulses during the positive slope of the input signal, and 15 The second output (OUTPUT2, 140) generates SFQ pulses during the negative slope of the input signal. The figure also shows the input coupling structure to the SQUID ring via (120). Depending on the applied input signal, the Josephson joints (J1, J2) have the corresponding slope phases. As a result of switching, the SFQ pulse sequences generated at the two outputs will follow each other. It is conceptually shown that it is formed in this way. 20 FIGURE 3 shows a detailed explanation of the preferred implementation method for the superconducting zero slope detection circuit. The circuit diagram shows two Josephson junctions (J1, J2), input coupling inductances (LC, LIN), bias source (VSQUID_BIAS), bias resistors (R1, R2), parasitic inductances It shows a SQUID ring containing the output inductances (LPR1, LPR2) and output inductances (LP1, LP2). Each of the Josephson joints (J1, J2) has a shunt resistor (RS1, RS2) included in the joint model. It includes resistors which are not shown as separate components in the diagram. Also, the first output... via route (130) to the first exit (OUTPUT1) and via the second exit route (140) to the second exit With the circuit structure connected to the output (OUTPUT2), the SFQ generated on OUTPUT1 / OUTPUT2 the transmission of pulses to downstream circuits via Josephsonian transmission line (JTL) chains The connections to these points are shown schematically. 30 12 FIGURE 4 shows a realistic SNSPD output pulse (200 ps linear rise, 5 ns exponential fall, 3 waveform showing simulation via superconducting zero slope detection circuit (mA peak amplitude) It is a set of forms. (a) shows the input current signal and the rise phase of the input pulse. (320), peak region (330) and fall phase (340) are marked in the figure; (b) ascent phase This shows the SFQ pulses generated on OUTPUT1 during the process, and the magnification addendum shows individual 5 pulses. It presents the resolved view of the SFQ impacts; and (c) during the fall phase. The SFQ pulses generated on OUTPUT2 are shown. The pulse activity of OUTPUT1 is also shown. The temporal transition between the termination and initiation of OUTPUT2 pulse activity, input This corresponds to the peak (zero slope) point of the signal (330). FIGURE 5, an SNSPD (310), superconducting zero slope detection circuit (100), a timing 10 single photon detection system consisting of capture circuit (200) and room temperature electronics (240) (300) is the system block diagram. In the figure, the cryogenic temperature of the components in question is shown (230) The current obtained from the SNSPD (310) output, in which it is located together, has zero slope. Timing capture is performed via detection circuit (100) and OUTPUT1 / OUTPUT2 lines applied to the circuit (200) and the digital time created by the timing capture circuit 15 It is shown that the output of the stamp (210) is transmitted to the room temperature electronics (240). FIGURE 6, a QKD transmitter (510), quantum channel (520), SNSPD array (310), superconducting zero slope detection circuits (100), timing capture circuit (200), classic processing unit (530) and invention containing classical channel (540) in a quantum key distribution (QKD) system It is a system diagram showing the application. In the figure, the SNSPD array (310), zero slope detection 20 cryogenic temperature stage (230) of circuits (100) and timing capture circuit (200) digital time obtained from the timing capture circuit in which it is located The stamp (210) information is suitable for processing by transferring it to the classical processing unit (530). and classical communication regarding QKD was carried out via the classical channel (540) It is shown. 25 FIGURE 7, a photon source (610), optical network (620), SNSPD array (310), superconducting zero slope including detection circuits (100), timing capture circuit (200) and classical controller (630) A system diagram showing the application of the invention in a photonic quantum computing system. This includes peak timing synchronization measurement, measurement-based feedforward, and photon count. It shows the role of photon detection for its resolution. In the figure, 30 from the optical network (620). Photons directed in a channel-based (ch1-chN) manner were detected by the SNSPD array (310); SNSPD array (310), zero slope detection circuits (100) and timing capture circuit (200) 13 The cryogenic temperature is located in the solid (230); obtained from the timing capture circuit. The digital timestamp (210) information obtained can be transferred to the classical controller (630) and the classical It is shown that the controller (630) can perform the relevant control / processing functions. FIGURE 8, Thermal noise at T = 4.2 K with parameters from Table 1 (K = 0.3, M = 13.6 pH). The timing of 1000 independently conducted Monte Carlo simulation trials is 5. It shows the performance of (a) the input current waveform, (b) each in the amplitude range of 0.1-1.5 mA. A raster of superimposed SFQ pulse times from 1000 trials for a given amplitude level. diagram (thin lines: exit A, rising slope; thick lines: exit B, falling slope), (c) detection The dependence of the obtained peak time on amplitude for threshold-based (A) and SFQ-based (B) methods comparison (error bars show the standard deviation of 1000 trials) and (d) every 10 This presents a comparison of the timing errors of the two methods in picoseconds. FIGURE 9 shows the second parameter set (K = 0.5, M ≈ 22.7 pH, VSQUID_BIAS = 1.0 mV) with T = 4.2 Timing performance under thermal noise at temperature K is shown in the same format as FIGURE 8. It shows. FIGURE 10, the third parameter set optimized for SNSPD reading (K = 0.494, LIN = 15 Thermal noise is effectively detected at a temperature of 45 nH, M = 387.6 pH, Φ0 / M = 5.34 μA, and T = 4.2 K. The timing of the 1000 independent Monte Carlo trials at each location where they were performed. Its performance is shown in the same format as FIGURE 8. Tested amplitude range: 5-30 μA. For six different amplitude levels, the SFQ pulse count (N) increases monotonically from 1 to 6. No bifurcation was observed at any amplitude point. 20 Explanation of References in Figures 100: Superconducting zero slope detection circuit 110: SQUID ring 120: Input coupling structure 130: First exit route 25 140: The second exit route OUTPUT1 / OUTPUT2: First / second exit route J1, J2: Josephson joints LC: SQUID ring inductance (coupled arm) LIN: Input coupling inductance 30 R1, R2: Polarity resistors RS1, RS2: Josephson junction shunt resistors 14 LPR1, LPR2: Parasitic inductances LP1, LP2: Output inductances 200: Timing capture circuit 210: Digital timestamp 230: Cryogenic temperature solid 5 240: Room temperature electronics 300: Single photon detection system 310: Superconducting nanowire single-photon detector (SNSPD) 320: Uphill phase (positive slope zone) 330: Peak point (zero slope point) 10 340: Falling phase (negative slope region) 400: Multiplexing Circuit 510: QKD transmitter 520: Quantum channel 530: Classical processing unit 15 540: Classic channel 610: Photon source 620: Optical network 630: Classic controller Description of the Invention 20 The present invention is a superconducting quantum interference device containing two Josephson junctions (J1, J2). Using the bidirectional response of the (SQUID) ring, the peak moment of a transient input signal pulse can be determined. a superconducting timing separation circuit that determines the (zero slope point) (hereinafter referred to as "zero") (also referred to as the "slope detection circuit") provides. Referring to FIGURE 2, the SQUID ring input It is designed to be magnetically coupled to the signal. Input signal current 25 When the magnetic flux in the SQUID ring increases, the change in the magnetic flux of the first Josephson junction (J1) by causing it to switch, Single Flux Quantum (SFQ) at the first output path (OUTPUT1) It enables the production of pulses. In the present invention, the quantizer circuit can be used with any counter, cumulative adder, or feedback circuit. It is executed in an open-loop configuration without the need for a loop. 30 When the input signal current decreases (negative rate of change / negative slope), the flux change reverses. This happens and the second Josephson junction (J2) switches on the second exit path. (OUTPUT2) causes the generation of SFQ pulses. At the peak of the input signal pulse, Since the rate of change of the input signal passes through zero, the SFQ pulse on OUTPUT1 the cessation of activity and the initiation of SFQ pulse activity on OUTPUT2 The resulting temporal transition defines the zero slope (peak) point. This allows the aforementioned The timing signal generated from the transition is obtained as a timing reference. 5 The present invention differs from the conventional constant fraction discriminator (CFD) approach in that it introduces an input Generating a bipolar signal from delayed and attenuated copies of a signal pulse. instead, the natural bidirectional quantization of the Josephson junctions (J1, J2) in the SQUID ring (110) It determines the peak timing by using the behavior of the input signal. In traditional CFD, the input signal is a The copy is delayed and attenuated, then reversed, and combined with the delayed signal to create a double 10. A polarized signal is obtained; the zero-crossing of this signal is used as the timing point. The current In the invention, the first and second Josephson joints (J1, J2) in the SQUID ring (110) the temporal transition between the SFQ pulses it produces in the positive and negative slope phases, respectively, It enables direct detection of the peak moment. This approach involves signal copying, attenuation, and delay. by eliminating analog processing steps such as 15, a completely digital and superconducting timing system It makes the distinction. In terms of distinguishing itself from bidirectional tracking ADCs in the present invention, bidirectional two The articulated SQUID quantizer topology has long been used in analog-to-digital conversion applications. Although it is an approach that has been known for a long time
[20] -
[23] , the present invention fundamentally changes the topology in question. It is applied for a different purpose. The operating principles of SQUID devices have been described in the literature for 20 years. This has been explained in detail
[21]
[22] . In monitoring ADC architectures, for increasing flux and decreasing flux. Two generated SFQ pulse sequences are applied to an up / down counter, and the resulting cumulative count, It represents the digitized amplitude value of the input signal. In other words, the ADC. This approach uses the cumulative effect (integral) of flux changes to determine "what is the value of the signal?" It aims to produce an answer to the question. 25 In contrast, the present invention uses the signal of the time derivative (rate of change) of the circuit input signal. It extracts timing information from the change. Specifically, from the activity of OUTPUT1 to OUTPUT2. The transition to activity is the moment when the sign of the time derivative of the input signal changes (the zero slope point). It determines. As mentioned above, the monitoring ADC typically asks, "What is the value of the signal?" It produces an answer to the question. The current invention, however, provides an answer to the question "When is the peak moment of the signal?" with 30 It aims to provide a timing reference. This distinction will be explained in detail below. This leads to fundamentally different circuit design requirements and operating modes. 16 Circuit Topology and Component Values Referring to FIGURE 3, a preferred application of the superconducting zero slope detection circuit (100) The form is a SQUID containing the first Josephson joint (J1) and the second Josephson joint (J2). It includes the ring (110). In the preferred implementation, (J1) and (J2) are positive and 5 with significantly equal critical currents to provide a symmetrical response to negative slope conditions. For example, the critical current Ic = 150 μA is determined for each joint. The meeting is not limited to this, but also includes critical current values in different application forms. Different options can be selected depending on the desired working conditions. The circuit (100) is an input coupling structure configured to receive the transient input signal pulse. (120) contains. The input coupling structure (120) is an input inductance 10 that carries the input signal. magnetic between (LIN) and an inductance (LC) forming part of the SQUID ring (110). It is designed to provide coupling. This coupling allows for changes in the input signal to... The conversion of the flux on the SQUID ring (110) and this flux change between (J1) and (J2) It is selected in such a way as to enable the creation of switching conditions on it. The coupling coefficient and Inductance values can be varied depending on the application. 15 Circuit (100) also requires that junctions (J1) and (J2) be polarized below their critical currents. It includes a configured power supply (VSQUID_BIAS). Preferably, each joint is standard. It is polarized at a DC current level below its critical current under operating conditions. Input pulse. When applied, the additional current contribution generated via magnetic coupling ensures the appropriate pulse flow. It ensures that the total current passing through the relevant joint in its phases exceeds the critical threshold. Polarization 20 The current can be routed through polarization elements (e.g. (R1), (R2)) depending on the application, or These can be achieved through equivalent arrangements. The first exit path (130) passes through the exit inductance (LP1) from the first Josephson junction (J1). It is connected and is connected to the first output (OUTPUT1). The second output path (140) is the second It is connected from the Josephson junction (J2) via the output inductance (LP2) and to the second output 25 (OUTPUT2) is connected. The first output is (OUTPUT1) and the second output is (OUTPUT2). The generated SFQ pulses are time-captured via Josephson transmission line (JTL) chains. The outputs of the circuit can be transmitted to downstream digital processing circuits such as (200). Since SFQ is in pulse form, existing SFQ cell libraries (DFF, T flip-flop, It is directly compatible with logic elements (combining, separating, etc.); this allows the output pulses to be 30 No additional signal conversion is required for recording, counting, or routing. 17 In a practical application, the Josephson transmission line (JTL) chains on the exit routes (130, 140) Some of the polarization currents can leak back into the SQUID loop (110). This parasitic leakage The currents determine the effective polarity ratio across the Josephson junctions (J1, J2) at the nominal setting. It raises its value above a certain threshold. If the effective polarization ratio exceeds a certain threshold, The joints generate continuous oscillations by spontaneously producing SFQ pulses without an input signal. 5 It enters this mode and the timing separation function disappears. In the preferred application mode, The nominal value of the polarization source (VSQUID_BIAS) is effective, including JTL leakage currents. The polarization ratio has been adjusted to remain below the oscillation threshold. In simulations, in the JTL configuration used, the JTL junctions with critical current Ic < 100 μA were found to have a JTL configuration. It was observed that it constantly oscillated due to leakage, and the safe operating zone was determined as Ic ≥ 10. 150 μA has been determined. These threshold values depend on the production process used and the output. This may vary depending on the JTL design of the roads. Reducing the critical current of Josephson junctions is a method for detecting low-amplitude signals. It can be considered as an alternative; however, the JTL leakage currents described above are low critical. It can cause current-carrying junctions to oscillate even at zero nominal polarity. Furthermore, 15 At very low Ic values, the Josephson binding energy (EJ = IcΦ0 / 2π) is the thermal energy (kBT). Approaching this level creates a risk of unwanted switching due to thermal noise. In a preferred application, (LPR1) and (LPR2) are equal within the manufacturing tolerances. can be determined in the values; (LP1) and (LP2) are equal values within the production tolerances. can be determined (LP1 = LP2 = 12.75 pH, Table 1). However, the invention is not limited to this, 20 the inductance values in question depend on the circuit layout, transmission line implementation method and Depending on the desired working conditions, different options can be selected. See the sub-options in FIGURE 3. The index / labeling scheme is used in accordance with the representation shown in the figure. The table below shows the component in the preferred application mode as shown in FIGURE 3. The values and circuit parameters used in the simulation in FIGURE 4 are summarized. This 25 The values were supported using a superconducting circuit simulator. Preferred application. The circuit was designed in this form using the Leibniz IPHT CJ2 process (Jc = 1.7 kA / cm²) and It has been confirmed. The invention can also be used in the fabrication process of HSTP, SFQ5ee or another superconductor. It is obvious that it can be applied. 18 Component Description Value J1 First Josephson junction, critical current Ic = 150 μA J2 Second Josephson junction, critical current Ic = 150 μA LC SQUID ring inductance (coupled arm) 13.68 pH LIN Input coupling inductance 150 pH R1 Primary polarity resistor 15 Ω R2 Secondary polarity resistor 15 Ω RS1 J1 shunt resistor is 1.86 Ω RS2 J2 shunt resistor 1.86 Ω LPR1 J1 parasitic inductance 0.1 pH LPR2 J2 parasitic inductance 0.1 pH LP1 First output inductance 12.75 pH LP2 Second output inductance 12.75 pH Coupling coefficient (between LIN and LC) is 0.3 Mutual inductance (K√(LIN·LC)) 13.6 pH VSQUID_BIAS SQUID polarity voltage 0.5 mV Josephson junction model parameters: Vg = 2.6 mV, Cap = 0.368 pF, R0 = 555 Ω, Rn = 15.5 Ω, Icrit = 0.1 mA, Rs = 1.86 Ω (shunt resistance, IcRs = 0.279 mV). The junction area is 1.5, and each The critical current for the junction is Ic = 1.5 × 0.1 mA = 150 μA. In the preferred application method, the Josephson joints (J1, J2) are fitted with Stewart-McCumber 5. The parameter βC = (2π Ic RS² C) / Φ0 is preferably kept below one (βC < 1), for the joints Fast operation in overdamped mode without latching after switching. This condition ensures that the input signal pulse returns to the superconducting state. It is preferred for reliably generating sequential SFQ pulses throughout the phase. 19 However, the invention is not limited to this, but also applies to configurations in the βC ~ 1-5 range. The circuit can work; however, in this case, as βC increases, the junction reset time increases. Due to the prolongation and ringing effect of successive SFQ pulses The time interval between them can be increased, and the number of pulses that can be generated can be decreased. Preferred In the application mode, Josephson joints operate in an overdamped regime with shunt resistances of 5. Although it is implemented, the basic principle of the method is to move from the first exit point to the second exit point. Using the SFQ pulse transition as a timing reference for the shunt resistor configuration It applies independently. Mutual inductance M = K × √(LIN × LC) = 0.3 × √(150 pH × 13.68 pH) ≈ 13.6 pH This is calculated. This value corresponds to a quantization step of Φ0 / M ≈ 152 μA. 10 The above results are based on a total of 30,000 samples taken at 30 different points under thermal noise at a temperature of 4.2 K. Verified by trial-based (30×1000) Monte Carlo simulation. 0.3 mA to 1.5 mA input. SFQ pulses were generated on the rising edge in the amplitude range between N = 2 and 9, and these values This is consistent with the theoretical formula N = I_peak / (Φ0 / M). Timing at stable operating points. The standard deviation of the error is less than 1 ps, and the average absolute timing error is 34 ps, not 15. It does not exceed the quantization boundary points where the value of N changes from one integer to another. The standard deviation increases (up to 11.5 ps) due to the effect of thermal noise, and the average timing also increases. The error reaches -44 ps. This transition limit effect is a phenomenon present in every quantizer; In practice, the input pulse amplitude is located to the right or left of these limit points, Steady-point performance is valid under practical operating conditions. 20 In the preferred application configuration, (J1) and (J2) have equal critical currents, positive and negative. It responds symmetrically to slope conditions. Similarly, shunt resistors (RS1) and (RS2) also by maintaining equal values, both joints should have the same reset dynamic. Output inductances (LP1) and (LP2) and polarization resistors (R1) and (R2) are provided. 25 by keeping the values equal between the first exit route (130) and the second exit route (140) The propagation characteristic has been ensured. This symmetry is achieved at the moment of transition from OUTPUT1 to OUTPUT2. It is important because it reliably reflects peak time. However, this invention... Component values are not limited to specific values but vary according to different application requirements. Adjustable. Circuit operation in asymmetric configurations (J1 ≠ J2 or RS1 ≠ RS2). While possible, in this case the switching between positive and negative slope phases is 30 Since the thresholds and reset times will differ, there will be a systematic asymmetry offset at the transition point. This can occur. This systematic offset can be determined from the circuit parameters and is repeatable. Therefore, it can be compensated for through calibration. The first output is also present in an asymmetric configuration. Timing reference of SFQ pulse transition from path (130) to second exit path (140) The principle of its use remains unchanged. Working Principle The circuit (100) operates as follows. In the static state, the first Josephson junction (J1) and the second 5 The Josephson junction (J2) is polarized at a value below its critical currents, SQUID The ring (110) is in a stable operating state. A transient input signal pulse (e.g. SNSPD) When the output pulse (310) is applied to the input coupling structure (120), the change in the input signal Switching of the circuit by creating a change in magnetic flux on the SQUID ring (110). It produces an effect that is conducive to inducing the behavior. 10 In the branch where the flux change (J1) occurs during the rise phase (320) of the input pulse It predominantly occurs in the direction that will support switching. In this phase, it passes through (J1). If the total current exceeds the critical threshold, (J1) switches and outputs the first output (OUTPUT1). Single Flux Quantum (SFQ) pulses are generated. Thanks to the resistive elements (J1) in the circuit. By quickly returning it to the superconducting state, it can be switched back on in a short time. 15 It can be made favorable. As long as the upward phase continues, it depends on whether the conditions are met. An SFQ pulse sequence can be generated on (OUTPUT1). When the input pulse reaches the peak region (330), the upward tendency of the input signal decreases and During the transition from rising to falling, the SFQ pulse activity at the first output (OUTPUT1) There is a 20-minute interval between the end of the first phase and the start of SFQ pulse activity on the second output (OUTPUT2). A temporal transition occurs; this transition corresponds to the peak (zero slope) moment of the input pulse. It will appear in the future. During the fall phase of the input pulse (340), the direction of flux change reverses, and this time Conditions are created on the arm where (J2) is located that will support switching. At this stage, (J2) If the total current passing through it exceeds the critical threshold, it switches (J2) and the second 25 SFQ pulses occur at the output (OUTPUT2). As the fall phase continues, the conditions change. Depending on the configuration, an SFQ pulse sequence can be generated on (OUTPUT2). Thus, the temporal sequence of the outputs is as follows: during the rise phase of the input pulse (OUTPUT1) SFQ pulse activity, a transition zone around the peak and the decay phase of the entry pulse. SFQ pulse activity occurs on (OUTPUT2). 30 from (OUTPUT1) activity. 21 Detecting the transition to (OUTPUT2) activity indicates the zero slope (peak) point of the input pulse. definitions. The moment of transition from the first to the second exit path coincides with the peak of the input signal pulse. This creates a corresponding timing reference. In this approach, the SFQ pulse at the outputs. It's not the numbers that are being processed, but the temporal transition between the two outputs. 5 Simulation Verification The operating principle described above uses a realistic SNSPD pulse shape to implement JoSIM. This has been numerically supported in the superconducting circuit simulator
[32] . Presented below The simulation results represent a preferred application form of the invention, and the invention This is not limited to these specific parameters. 10 In the simulation, a typical SNSPD (310) output pulse was used as the input signal: 200 ps A linear rise phase followed by an exponential fall phase with a time constant of 5 ns. Impulse t = It starts at 1.0 ns and reaches a peak value of 3 mA at t = 1.2 ns. This pulse type is described in the literature. It reflects the reported typical SNSPD output pulse characteristics [1][5]. In the simulation The circuit used (100), component 15 in the preferred application form shown in FIGURE 3. It was modeled with the following values (Ic = 150 μA, M = 13.6 pH, K = 0.3). Figure 4(a) shows the input current signal. Two characteristic phases of the SNSPD pulse. It is clearly distinguished: positive slope region (320), peak (330) and negative slope region (340) shown input current signal. This asymmetric pulse shape is the enhancement of the present invention and An ideal test signal to demonstrate its ability to independently distinguish between the phases of a fall is 20. It constitutes. FIGURE 4(b) shows the SFQ pulses generated on the first output path (OUTPUT1, 130). This is shown. The first Josephson during the positively sloped rising phase of the input pulse. By switching the junction (J1) in succession, a total of 18 SFQ pulses are applied to OUTPUT1. It produced these blows all at approximately 11 ps intervals within a 200 ps boost window. The temporal resolution of individual pulses is clearly shown in the magnification supplement in the figure. It is observed that the number of pulses, N = Itepe / (Φ0 / M), is consistent with the theoretical formula. FIGURE 4(c) shows the SFQ pulses generated on the second output path (OUTPUT2, 140). It is shown that during the fall phase, the second Josephson joint (J2) switches. OUTPUT2 generated a total of 18 SFQ pulses. Unlike the rise phase, these 30 The pulses are distributed at progressively less frequent intervals over a period of approximately 25 ns. This situation, 22 The exponential drop is directly related to the progressively decreasing slope value (|dI / dt|): at the onset of the impact. The impacts are frequent, but as the slope decreases while the fall progresses, the impacts are produced less frequently. As clearly shown in FIGURE 4, there is a correlation between the OUTPUT1 and OUTPUT2 impact activities. A distinct temporal discrepancy exists. The last SFQ pulse on OUTPUT1 is the same as the input signal. It occurs very close to the peak moment (t = 1.2 ns) and is immediately followed by OUTPUT2 5 Pulse generation begins on it. This transition is the moment when the slope sign of the input signal changes. In other words, it directly reflects the peak (zero slope) point. Against the pulse amplitude of the transition moment. The restricted behavior is quantitatively validated at different amplitudes in FIGURE 8. Dead Time and Recovery After the SNSPD (310) output pulse is damped and the SFQ pulse on the second output (OUTPUT2) is 10 After the activity ends, the zero slope detection circuit (100) detects both Josephson junctions (J1, J2) Static operating condition where the polarized current is below the critical current. It returns. At this stage, the time it takes for the circuit to become ready again is the SQUID loop (110) rather than its own switching dynamics, the SNSPD (310) output that forms the input signal It is determined by the duration and damping characteristics of the pulse. In other words, the SQUID side 15 Although it can be reswitched back in very short periods of time, the system's next event becoming ready to detect reliably essentially depends on the rescue behavior of the SNSPD(310). and depends on the input signal returning to its baseline. Consequently, in applications with high photon count rates, the system's dead time, not from the limitations of the weighted discriminative circuit, but from the consecutive events of SNSPD(310) 20 It results from the recovery process between them. When SNSPD (310) is recovered and the input signal When it returns to its baseline, the circuit (100) is ready to respond to the next photon event. income. Due to the low rate of change of the input signal around the peak (zero slope) region, Noise may occur in some cases on (OUTPUT1) and / or (OUTPUT2) and may be faulty or infrequently "false". 25 It can lead to the formation of SFQ pulses. However, in typical SNSPD (310) pulse patterns The peak region is traversed quickly, and the joint switching exhibits a specific threshold behavior. When considered together, these incidental shocks remain practically limited in number. and at most one or a few additions, provided that this does not fundamentally alter the timing reference. It is limited by the coup. 30 23 This assessment is also supported by thermal noise analysis at an operating temperature of 4.2 K. Josephson junction energy is three orders of magnitude greater than thermal energy (E_J / kT > 10³), The probability of noise-induced faults generating an SFQ pulse is negligible. Post-peak OUTPUT1 pulses observed in simulations (1 to 10 depending on amplitude) (number, all within 55 ps from the actual peak moment) not from thermal noise, but from the 5 in the SQUID ring. This is due to the inability of the flux dynamics to respond instantaneously to the zero-slope transition. The deterministic effect is limited by the circuit parameters. Peak Detection with Limited Timing Error The timing reference derived from the zero slope transition is proportional to the amplitude of the input pulse. non-growing, circuit quantization parameters (Φ0 / M) and SQUID switching dynamics 10 It is obtained within a limited margin of error. The peak region of the input pulse, Even if the magnitude of the pulse changes, it corresponds to the same relative temporal position on the waveform. a higher amplitude pulse results in more intense SFQ pulse activity at each output. It can open; a lower amplitude pulse, however, may result in less frequent SFQ pulses. However, the activity on the first output (OUTPUT1) ends and the second output 15 The transition in which activity begins on (OUTPUT2) is always at the peak (zero slope). It occurs in a way that corresponds to the moment. In threshold-based methods, the timing error amplitude While it grows proportionally and indefinitely, the error in the current invention is related to the circuit parameters. It is limited. This aspect differs fundamentally from threshold-based discrimination. This limited timing error characteristic is a universal result dependent on specific component values. 20 For any given value of mutual inductance (M), the flux quantization step Φ0 / M distinguishes the circuit. It determines the minimum current change it can withstand. The last SFQ in the first output path (OUTPUT1). The timing interval between the first SFQ pulse and the second output path (OUTPUT2), This is limited by the quantization step. Therefore, the timing error |Δt| ≤ Φ0 / (M · |dI / dt|peak) It is limited by the correlation and is independent of the amplitude of the input pulse. This feature, threshold 25 In amplitude-based methods, the timing error increases structurally in proportion to the amplitude. It is different and constitutes the fundamental difference between this invention and previous techniques. Notably, the bidirectional SQUID quantizer topology was developed in 1989 by Lee and First described by Petersen
[20] and recently experimentally as an ADC Although verified in this manner
[33] , with the knowledge of the applicants, these 30 The limiting characteristic of Φ0 / M on the timing error of the topology is the subject of these studies. It was not defined in any of them or used for the purpose of timing distinction.
[33] 24 In the previous technique, including experimental studies, the bidirectional SQUID structure was only used. The amplitude is considered for digitization purposes, and the temporal transition between outputs is defined as Φ0 / M. It presents a timing error independent of the limited amplitude, within the scope of the present invention. It has been defined and used for the first time. This limitation was not defined in the previous technique or It has not been characterized. Error 5 in ADC applications of bidirectional SQUID quantizer. The metrics are called integral nonlinearity (INL) and differential nonlinearity (DNL). The concept of timing error is defined as
[22]
[23]
[33] and is included in these applications. It does not include the timing error limited to Φ0 / M, which is a design flaw in the present invention. It is a new metric used for optimization and performance evaluation, unlike the previous one. There is no equivalent in technology. This feature is merely the discovery of a natural consequence of topology. 10 no, the circuit is consciously restructured for a different purpose and the results obtained The result is characterized in a way that was unpredictable with the previous technique. The timing resolution (tres) of the present invention is the sequential SFQ generated in the first exit path (130). It is determined by the temporal distance between the pulses. This distance is approximately expressed as trise / N. can be determined; here, trise is the rise time of the input pulse and N is the 15 produced during the rise phase. SFQ is the pulse count. The value of N is the mutual inductance (M), biasing current (Ibias), and input. It is determined by the amplitude of the pulse. In general, SFQ is generated during the rising phase. The number of pulses can be approximately expressed by the relation N ≈ Itepe · M / Φ0, where Itepe is the input. The peak current value of the pulse is M, the mutual inductance is Φ0, and the magnetic flux quantum is Φ0. The correlation is consistent across all three parameter sets: in the first set (M = 13.6 pH), 20 with an amplitude of 1.5 mA. N ≈ 10, in the second set (M = 22.7 pH) the same amplitude N ≈ 16, in the third set (M = 387.6 pH) 30 μA The values calculated for amplitude N ≈ 6 are consistent with the simulation results. (Reciprocal) Increasing the inductance (M) or increasing the biasing current (Ibias) with the same input pulse This allows for the generation of more SFQ pulses, and consequently, finer timing resolution. This leads to the following: The slope (dI / dt) of the input signal is not constant during the rise phase; it reaches its peak at 25. As we get closer, the slope decreases, and consequently, the temporal interval between successive SFQ pulses also decreases. It expands. The largest quantization step, which occurs around the peak, is the actual timing of the circuit. It determines the resolution. Figure 8 (N = 1-9, K = 0.3) and Figure 9 (N = 1-16, K = 0.5) show this relationship. This is demonstrated with two different sets of parameters. This adjustability affects the timing of the circuit. The resolution is optimized during the design phase according to the requirements of the target application. It makes it possible to do so. When the input signal pulse reaches its peak (330), the rate of change of the input signal starts from zero. It passes. However, the first Josephson joint (J1) passes for a limited time even after the apex. SFQ can continue to generate pulses. This situation affects the total on the SQUID ring (110). a transition window determined by the time it takes for the current to drop below J1's reset threshold This results from the creation of a circuit. The transition window in question is not a random error, but a circuit error. A repeatable and calibratable switching system with fixed parameters (M, Ic, Ibias). This creates the offset. This deterministic offset takes similar values for different pulse amplitudes and is 5 It ensures the consistency of the timing reference. Because this offset is systematic and repeatable, This can be corrected by measuring it during system calibration and subtracting it from the timing reference. For example, the first output can be obtained by applying a test pulse with a known timing reference. The difference between the timing of the last SFQ pulse (130) on the path and the known peak moment is measured. and this difference is recorded as the systematic offset correction value. This calibration process is performed on the circuit's 10 This is performed at operating temperature and polarization conditions. Observed in simulations. Post-peak OUTPUT1 pulses (1 to 10 depending on amplitude, all true peaks) (from that moment onwards, within 55 ps), this directly demonstrates the transition window behavior. In tracking ADC architectures, the SFQ pulse counts generated on the first and second output paths are the input. It represents the amplitude information (cumulative count) of the signal; an increase in the pulse count indicates a larger 15 a decrease in amplitude corresponds to a smaller amplitude. In the present invention, however, in the first exit path (130) The temporal distance between successive SFQ pulses generated determines the timing of the circuit. It determines the resolution. In other words, the tracking ADC asks "how many pulses were generated?" In answering the question, the information in the current invention is encoded as "what is the time interval between pulses?" This fundamental functional difference is encoded in the answer to the question "how much?". This is a different 20 of the same topology. This stems from its use for various purposes and with different design optimizations. Another In other words, the current invention represents the information extraction paradigm of the bidirectional SQUID quantizer. It fundamentally changes the previous technique (ADC applications) using cumulative SFQ pulses of information. While encoded in number, in the present invention, information transitions from the first exit route to the second exit route. It is being coded in a timely manner. This paradigm shift leads to design in a different and even opposite direction. 25 This requires optimizations: mutual inductance (M) amplitude in ADC applications While being scaled down for resolution, the current invention uses M timing resolution. is being scaled up. Similarly, closed-loop or counter-based structures are preferred in ADCs, The current invention uses an open-loop configuration. This reverse optimization is the same. This shows that an unpredictable and different result was obtained from the topology. 30 In a practical application, the timing reference takes different forms depending on the character of the transition. This can be determined. For example, the timing reference can be set immediately before the transition (OUTPUT1). Based on the last SFQ pulse generated, an upper limit for the peak moment or just before the transition is determined. 26 then a lower bound for the peak moment is determined based on the first SFQ pulse generated on (OUTPUT2). It can be defined as follows. Timing resolution is the last SFQ pulse in the first output path. It is limited to the time interval between the first SFQ pulse on the second exit path and the second exit path. The subject range refers to the input signal shape around the peak region and the SFQ pulse during the transition. It may vary depending on production intensity. 5 Typical SNSPD (310) pulse shapes and fast switching of superconducting Josephson junctions. When these characteristics are considered together, this approach allows for timing in the picosecond range. It is anticipated that it will be suitable for providing resolution. This will allow, especially, impact resistance. In cases where the amplitude is variable, amplitude-dependent approaches emerge in threshold-based approaches. Time drift effects can be significantly reduced and a more consistent timing output can be obtained. It can be done. The timing performance of the present invention is demonstrated by using realistic SNSPD pulse parameters. an open-source superconductor independently verified and supported by peer-reviewed publications This is numerically supported in a circuit simulator (JoSIM, Stellenbosch University). Below are the simulation results obtained for two different sets of parameters. Both 15 The parameter set also includes a linear rise time of 200 ps and an exponential drop time of 5 ns as the input signal. A typical SNSPD pulse shape with a time constant (τ) is between 0.1 mA and 1.5 mA (0.1 mA It was tested at fifteen different amplitudes (in steps). First Parameter Set (Table 1) FIGURE 8 shows the first-choice application method given in Table 1 (K = 0.3, M = 13.6 pH, 20 Timing performance (Φ0 / M ≈ 152 μA) effectively reduces thermal noise at a temperature of T = 4.2 K. This is demonstrated by 1000 independent Monte Carlo trials that have been carried out. In FIGURE 8(a), two timing methods are shown on an SNSPD pulse with an amplitude of 1.0 mA. They are compared. In A, a conventional threshold-based discrimination triggered at a threshold level of 0.1 mA is used. In case B, the method of the existing invention was applied, resulting in the first output of 25. (OUTPUT1) The time of the last SFQ pulse is taken as the peak reference. The input is shown in the figure. The rising phase (320), peak (330) and falling phase (340) of the coup are marked. In FIGURE 8(b), the first output is shown for fifteen different amplitude levels in the amplitude range of 0.1-1.5 mA. Individual SFQs produced on exit line 1 (OUTPUT1, 130) and exit line 2 (OUTPUT2, 140) The temporal distribution of the pulses is shown as a raster diagram. Each amplitude is 30 The level is represented as a horizontal line, showing all SFQ pulse times for 1000 attempts. 27 They are superimposed. The thin lines represent the pulses generated at exit A (during the ascending slope), The thick lines indicate the pulses generated at exit B (during the descending slope). At the deterministic operating points, the pulses of all trials coincide exactly (bold, (sharp lines). This representation visually reveals the deterministic nature of SFQ quantization. It states that: thermal noise alters pulse times to the sub-picosecond level, but 5 It does not change the pulse count. In the first parameter set (K = 0.3), N values range from 1 to 9. It varies between them. Figure 8(c) shows the dependence of the detected peak time on the pulse amplitude. Threshold In the method based on (A, open squares), the detected time pulse amplitude is strongly correlated. It depends: low-amplitude pulses are triggered later because they reach the threshold level later, 10 High-amplitude pulses reach the threshold earlier, triggering the system much sooner. In contrast, SFQ-based method (B, filled tiles) close to the true peak time (330) across the entire amplitude range. and produces a stable timing reference. Method B averages 1000 trials (tiles) It is shown with (markers) and standard deviation (error bars). Error bars indicate stable It is very small at the operating points (standard deviation < 1 ps), but 15 at the quantization limit points. It increases due to the effect of thermal noise (maximum 11.5 ps). In FIGURE 8(d), the timing errors of both methods are compared in picoseconds. A (threshold-based): Timing error from low amplitudes to high amplitudes is approximately 100 ps. It increases monotonically up to 187 ps. B (slope-based): Timing error is positive and It oscillates between negative values, and this oscillation occurs when the SFQ pulse number (N) is 20. This is due to the amplitude increasing discretely. The largest absolute error is 44 at 0.9 mA. This was observed as ps. The standard deviation is typically less than 2 ps. Gray dots represent individual values. The experiments show that the black tile markers represent the average value. This error is due to the pulse amplitude. It does not show a systematic dependence and is determined by the quantization step Φ0 / M. It is a structural limit. 25 Second Parameter Set FIGURE 9 shows a second preference configured with stronger coupling and higher polarization. the timing performance of the implemented application method, thermal noise effective at T = 4.2 K This is demonstrated by 1000 independent Monte Carlo experiments conducted as a parameter. In the set, the coupling coefficient K is increased to 0.5, and the mutual inductance M ≈ 22.7 pH (Φ0 / M ≈ 91 30 The polarity (μA) value has been increased; the biasing voltage VSQUID_BIAS has been increased to 1.0 mV, thus increasing the junctions. 28 The polarization was achieved at approximately 48% of the critical current. Other circuit parameters... This is the same as Table 1. In FIGURE 9(b), 1000 for fifteen different amplitude levels in the amplitude range of 0.1-1.5 mA. The entire SFQ pulse times of the experiment are shown as an overlaid raster diagram. This is shown. In the second parameter set (K = 0.5), the N values range from 1 to 16. and, due to increased coupling, higher pulse counts compared to the first parameter set. It is observed that the pulses of all trials are precisely at the deterministic operating points. They overlap. The transition region remains within a narrow time window across the entire amplitude range. As shown in FIGURE 9(c), the SFQ generated in the first output path in the second parameter set. The number of pulses N varies between 1 and 16, and even at the lowest test amplitude (0.1 mA) it is 10. Pass detection is achieved by generating a number of SFQ pulses determined by the circuit parameters. B The method's error bars (standard deviation of 1000 trials) are at most amplitude points. It is invisibly small. At a specific amplitude point (0.7 mA) compared to other points. Wider error bars are observed. This situation indicates that the input current is precisely at that amplitude. 15 as the limit of whether SQUID will fire an additional SFQ pulse during its ascent phase. This is explained by the presence of (bifurcation point). Some experiments with thermal noise involve N pulses. Some produce N±1 pulses, and this results in a jump of approximately three times the time of the last pulse. This is why. This observation shows that even at the bifurcation point, the timing error is quantized. This confirms that it is limited to the step (tres). In FIGURE 9(d), the absolute value of the B (slope-based) timing error is 20 in the range of 0.2-0.6 mA. It remains low at stable operating points (< 21 ps). 0.7 mA from the bifurcation point. A systematic negative shift is observed from this point, and the absolute error is highest at the amplitude (1.5 The amplitude (mA) reaches approximately 43 ps. At the lowest amplitude point (0.1 mA; N = 1 pulse), however... An error of approximately 55 ps in the positive direction was measured. Gray dots represent 1000 individual trials, black tiles... The signals indicate the average value. The error signal is positive at low amplitudes and negative at high amplitudes. The amplitudes are negative, and this situation occurs shortly after the actual peak of the last SFQ pulse. This is due to the fact that it can occur before or after; the fluctuating current in question This is an expected result given the nature of quantization. The first and second sets of parameters define the basic operating principle of the invention (open-loop timing). (separation, limited timing error, thermal noise tolerance) accessible circuit 30 This is proven by its parameters. The third set of parameters presented below is the SNSPD of the circuit. It shows its configuration optimized for reading. 29 Third Parameter Set (Optimized for SNSPD Reading) The coupling coefficient (K) directly influences the progression of the SFQ pulse number (N) in the operating amplitude range. It has an effect. Since the mutual inductance M = K × √(LIN × LC), increasing the value of K increases M. It amplifies and reduces the quantization step Φ0 / M; this results in more SFQs in the same amplitude range. This leads to the generation of a pulse. However, the choice of the K value only increases the number of pulses to 5. It also determines whether N increases monotonically at successive amplitude levels. In the third preferred application method, K = 0.494, LIN = 45 nH and LC = 13.68 pH. The following parameters were used (M = 387.6 pH, Φ0 / M = 5.34 μA). In this parameter set, 5-30 μA SFQ pulse generated in the first output path (130) for every 5 μA amplitude increase in the input amplitude range The number (N) increases by exactly one unit, resulting in a monotonic progression of N = 1, 2, 3, 4, 5, 6. It has been determined that K = 0.494 is the minimum coupling that provides a monotonic N increase in this amplitude range. is the coefficient that best balances both timing accuracy and amplitude resolution. It represents the working point. In this parameter set, the input inductance (LIN = 45 nH) is different from the SQUID loop inductance (LC = The pH (13.68) is approximately 3289 times larger. This asymmetric scaling occurs at a level of Φ0 / M = 5.34 μA. By providing a quantization step, an external source of SNSPD pulses with amplitudes as low as 15 μA is possible. This enables direct detection without the need for a signal amplifier. Typical SNSPD. Since their kinetic inductances Lk are in the range of 200-1000 nH, LIN / Lk is between 1 / 4 and 1 / 18. It remains and is confirmed that the input inductance creates a low impedance load on the detector. FIGURE 10, timing 20 of the third parameter set (K = 0.494, M = 387.6 pH, Φ0 / M = 5.34 μA). Its performance was effectively achieved with 1000 independent thermal noise tests at a temperature of T = 4.2 K. This is demonstrated by a Monte Carlo experiment. The tested amplitude range is 5-30 μA, in 5 μA steps. It includes six different amplitude levels. At each amplitude level, the first exit path (130) The number of SFQ pulses generated (N) was determined as 1, 2, 3, 4, 5, 6 respectively, and no amplitude was specified. No bifurcation was observed at point 25; in all 6000 trials, the pulse count was the same. The timing was obtained deterministically. The timing was standard at all amplitude points. The deviation remained below 2.3 ps. A noteworthy observation emerges in FIGURE 10: in the N ≥ 2 regime (10-30 μA amplitude interval), timing error of the last SFQ pulse in the first exit path (130) all operation At these points, |Δtlast| remains at < 20 ps (the highest value is 19.7 ps at 10 μA). The same 30 The timing error of the threshold-based method in the amplitude range is between 110 ps and 170 ps. It is changing. In the multiple SFQ pulse regime, the present invention offers at least the same efficiency as the threshold-based method. It provides a 5.6-fold improvement (at 10 μA). At the N = 1 operating point (5 μA), however, only a single Since an SFQ pulse has been generated, the convergence effect has not yet occurred and the timing error (64.1 ps) is higher compared to the threshold-based method (20 ps); this means the circuit has at least two SFQs. It emphasizes the importance of configuring it to generate a pulse (N ≥ 2). This observation reflects the fundamental mechanism of multiple SFQ pulse convergence: rise 5 Each successive SFQ pulse generated during the phase reduces the accumulated flux in the SQUID ring (110). By reducing Φ0, the firing threshold of the next pulse is positioned closer to the peak. It carries. Consequently, the final SFQ pulse on the first exit path peaks due to successive convergence. It occurs at the temporal position closest to the moment. In threshold-based methods, amplitude increase is time-related. While the time walk amplitude is increased monotonically, the SFQ-based method increases the amplitude by 10. This leads to the generation of more pulses and therefore to a better convergence of the final pulse to its peak. This inverse relationship opens up, previously in the context of the timing separation of SFQ quantization. It is an undefined property. Parameter-Independent Result Simulation results obtained for the three different parameter sets above, with limited time 15 The qualitative nature of the error characteristic (being limited and amplitude independent) depends on parameter selection. This confirms that they are independent. Any mutual inductance (M) value and any For a single-peak transient input pulse shape, the timing error in the slope-based approach is Φ0 / M. While the timing error remains limited, in the threshold-based approach, it is proportional to the amplitude change. It grows. This qualitative distinction is due more to the flux quantization of the circuit than to an empirical discovery of the simulation. This is a mathematical consequence of its architecture. These results demonstrate that the present invention's threshold-based discrimination It quantitatively demonstrates its significant advantages compared to other methods: (i) timing error It does not grow proportionally with the pulse amplitude and is limited by the circuit parameters, (ii) the absolute error value is lower even compared to the best case of the threshold-based method and (iii) Since the timing resolution of the system is determined by the circuit parameters (Φ0 / M), 25 It is a characteristic that can be controlled and optimized during the design phase. The amplitude range tested in the simulations was 0.1-1.5 for both the first and second parameter sets. The first parameter set is mA (15:1 dynamic range), and the third parameter set is 5-30 μA (6:1 dynamic range), each... Both values represent typical amplitude variations (approximately 2-3 times) observed in real SNSPD systems. [5]) satisfies it more than adequately. The reliable operating condition of the circuit is that the input pulse rises by 30 generating a number of SFQ pulses determined by the circuit parameters throughout the phase This requires and in preferred application methods the minimum input amplitude is approximately 31 Imin is determined by ≈ Φ0 / M. As shown in FIGURE 9, the coupling coefficient (K) or By increasing the polarization level, the minimum operating amplitude can be reduced to lower values. By selecting the mutual inductance (M) value differently depending on the application, the minimum The working amplitude can be adjusted according to the target SNSPD amplitude range; this allows for different applications of the invention. It is possible to adapt the detector types to different operating conditions. 5 The simulations above were performed in the JoSIM superconducting circuit simulator
[32] using the ".temp 4.2" command. This was achieved by enabling thermal noise models. This command applies to all circuits. Johnson-Nyquist thermal noise on resistors and Josephson junctions It enables it. In each configuration (Table 1, second parameter set and third parameter set) and at each amplitude point (between 0.1 and 1.5 mA in the first and second sets, 15 10 (point; 5 to 30 μA in the third set, 6 points) 1000 independent Monte Carlo experiments This was done. In total, 2 × 15 × 1000 + 6 × 1000 = 36,000 simulations were run. FIGURE Figures 8, 9 and 10 show the results of these thermal noise simulations. Monte Carlo results from all three configurations revealed the following three key findings: It includes: 15 (i) Deterministic points: Thermal noise timing at almost all operating points. It does not affect the sensitivity (standard deviation < 2.5 ps). Number of pulses (N) and transit time. It's the same in all trials. (ii) Bifurcation point: In the first and second parameter sets, at specific amplitude values N A stochastic transition is observed between N±1. However, even in this case, the timing error is 20. It is limited by ±tres; that is, the error is quantized and limited, and not continuous. The third parameter In the set (K = 0.494), no bifurcation was observed at any of the tested amplitude points (5-30 μA). Deterministic behavior was obtained in all 6000 trials (FIGURE 10). This result indicates that... that bifurcation regions can be eliminated by appropriate selection of the coupling coefficient (K) It shows this. In addition, by selecting appropriate circuit parameters according to the SNSPD output values, 25 The circuit can be configured to operate to the right or left of the bifurcation region. (iii) Comparison with threshold-based method: In method A, thermal noise is added at the threshold crossing moment. By adding uncertainty, the timing error increases at each amplitude. In method B, however, the noise effect... SFQ is suppressed due to the deterministic nature of quantization. These results correspond to section 30 of the specification, "Peak Detection with Limited Timing Error". The stated limit |Δt| ≤ Φ0 / M is also valid under 4.2 K thermal noise conditions. 32 This confirms the observation previously made in the section "Dead Time and Rescue". The stated Josephson junction energy is three orders of magnitude greater than thermal energy (E_J / kT > This is consistent with 10³) and this theoretical prediction has been quantitatively proven with 36,000 independent simulation trials. It supports. This deterministic behavior under thermal noise makes the present invention superior to threshold-based methods. This constitutes an unpredictable technical advantage. In threshold-based methods, thermal Noise adds continuous and unlimited additional timing error at each amplitude, while SFQ-based In this method, the deterministic nature of flux quantization naturally reduces the noise effect. It suppresses. Asynchronous Work 10 A key feature of the present invention is the SQUID-based zero slope with timing separation front-end. The detection circuit (100) can operate asynchronously without using any external clock signal. It is convenient. The first and second Josephson junctions (J1, J2), polarization current and input. conditions formed by the summation of the effects resulting from magnetic coupling of the signal Below this, it naturally switches when the critical threshold is exceeded. In this context, zero slope (peak) is 15. "Gating" of switching events related to detection or according to a specific clock phase Synchronization is not mandatory. However, recording the timing reference or transferring it to later circuits downstream read / capture circuits used for transmission (for example, a (trigger / latch element) depending on its architecture, it can take 20 from a clock signal. It should be considered that it can be used. However, the critical timing discrimination function, that is... Determining when the input pulse reaches its peak (zero slope) moment, SQUID-based front end This is performed asynchronously, independently of any clock phase. This asynchronous front-end operating mode is particularly advantageous in terms of SNSPD (310) reading; Because photon arrivals are by nature random, the circuit has 25 at any given time. It must be prepared to respond to and differentiate a potential photon detection event. The process is expected to not carry any additional uncertainty depending on the clock phase. It is necessarily triggered by the clock. One differentiating approach limits the resolution by restricting timing determination to the clock period. This can lead to a quantization effect that is conducive to lowering the system's dead time. In this context, the system's dead time also... rather than the separation circuit itself, the consecutive events of the SNSPD (310) 30 the recovery time between these steps and, if applicable, the readout of the downstream reading circuit. This is determined by the reset / return requirements. 33 In the previous technique, known bidirectional tracking in ADC applications also involved the input current. SQUID-based systems that respond to changes and exhibit asynchronous switching behavior. It should be noted that quantizers are used. However, the distinction of the present invention is SQUID. Rather than the ability of the joints to perform asynchronous switching, the purpose of this asynchronous operation is and It is related to the context of application. 5 The current circuit uses isolated and random transient pulses, with a single timing for each event. It is structured in a way that facilitates the extraction of the reference. In contrast, the tracking ADC Their architectures typically operate on signals that are continuous and change relatively slowly. and cumulative accumulation and, in most cases, clockwise down to obtain amplitude information. It is based on a flow counting / processing structure. The present invention, however, specifically targets 10 individual transient events. To be able to perform timing separation independently of the clock phase, asynchronous It utilizes a front-end approach. Open Loop Operation In order to expand the dynamic range and keep the SQUID at its most precise operating point, the previous 15 of the flux-locked feedback loops (FLL) in bidirectional tracking ADC systems can be used
[28]
[29] . In contrast, in the present invention, superconducting zero slope detection circuit (100) in open loop configuration without any feedback loop This approach is particularly well-suited for event-based responses to transient (single) input pulses. It is suitable in situations where feedback needs to be given; because the settling / settling behavior of the feedback, the input This may not be consistent with the short-lived character of the coup. 20 Furthermore, the magnitude of input pulses from sources similar to SNSPD affects the sensitivity of the flux. positive and negative slope phases by triggering joint switching rather than measurement. It is at a level that allows for differentiation between them. In this context, the function of the circuit is the precise current. Rather than measuring its value, it detects the transition between the rising and falling phases of the input pulse. Since it relies on this, open loop 25 without needing dynamic range expansion with FLL. The desired timing reference can be obtained through this study. In the previous technique, the established function of the bidirectional SQUID quantizer was continuous. Up / down counter or cumulative adder for the purpose of amplitude digitization of signals. It is used in conjunction with their structures. The present invention abandons this established function: a circuit It does not include an up / down counter, cumulative adder, or flux-locked feedback loop. These 30 Instead, the second output is triggered by the termination of SFQ pulse generation on the first output path (OUTPUT1). Determining the temporal transition between the start of SFQ pulse generation (OUTPUT2) and the path. 34 This constitutes the only functional output of the circuit. This structural difference allows for monitoring of the present invention. ADC architectures differ not only in terms of application area but also in terms of circuit components and operation. This also shows that they differ in terms of their form. Timing-Optimized Circuit Design The superconducting zero slope detection circuit (100) within the scope of the present invention is similar to the previous bidirectional 5 SQUID quantizer arrangements prioritize timing over amplitude measurement. It is distinguished by its pursuit of accuracy, and there are multiple design choices that serve this purpose. has. In the preferred application methods, the first and second Josephson joints ((J1), (J2)) To ensure a rapid return to the starting state after switching, the polarity resistor must be 10Ω. (R1, R2) values are lower values seen in some tracking devices. A higher level can be selected compared. This reduces the tendency of the joints to "lock," thus improving entry. the pulse remains ready for successive switching throughout its rise phase, thus enabling the input This makes it easier to accurately track changes in the slope. In other words, the priority here is... Rather than maximizing amplitude sensitivity, rapid recovery and timing stability 15 to provide. First exit route (130) and second exit route (140), the first exit (OUTPUT1) and the second exit (OUTPUT2) temporal response should be designed to be as symmetrical as possible. is suitable. In the preferred application methods, for this purpose, the exit routes (130, 140) They show similar characteristics in terms of impedance and propagation delay. 20 This can be ensured. Thus, SFQ pulse generation on (OUTPUT1) ends and (OUTPUT2) is activated. The timing represents the transition where pulse generation begins, marking the peak (zero slope) moment. It becomes more reliable in terms of accuracy. Entry joint structure (120), joints (J1, during the rising and falling phases of the entry impact J2) is arranged to provide a magnetic coupling sufficient to initiate the switching. 25 In the preferred implementation modes, coupling is limited throughout each phase of the input pulse. It can be selected to generate a certain number of SFQ pulses; thus, facilitating the detection of the transition. While obtaining the output pattern, transition blurring or SQUID due to excessive pulse generation occurs. This prevents the negative impact on the working regime of the ring (110). This approach is particularly important for SNSPD (310) In cases where the amplitude of the input pulses from the source can vary depending on the application, 30 It contributes to the stability of peak timing. 35 In the previous technique, the bidirectional two-junction SQUID quantizer topology was used in tracking ADCs
[20] -
[25] and cumulative SFQ pulse of input magnitude in digital SQUID magnetometers
[30]
[31] It was used for the purpose of digitization through counting. These applications are continuously monitored. In this mode, it operates with a flux-locked feedback loop (FLL) and uses cumulative pulse information. It is coded with awareness. In the present invention, the circuit is event-triggered and open-loop 5 working in its configuration, information not in cumulative counting but from the first exit route (130) second It is coded at the time of transition to the exit route (140). Simultaneous Amplitude Information The number of SFQ pulses (N) generated in the first exit path (130) is a natural characteristic of flux quantization physics. As a result, it is proportional to the amplitude of the input pulse. However, within the scope of the present invention, circuit 10 Not designed or optimized for amplitude measurement, mutual inductance (M) and shunt resistance (RS1, RS2) values prioritize timing accuracy and speed over amplitude resolution. It has been selected for reset. These parameter selections are suitable for the circuit's use as an ADC. It creates a non-existent point of focus. Timing Capture Circuit 15 As shown in FIGURE 5, in a preferred application configuration, the zero slope detection circuit (100), SFQ pulse generation ends at the first output (OUTPUT1) and at the second output (OUTPUT2) A timestamp at the transition moment defined by the start of SFQ pulse generation. It is connected to a timing capture circuit (200) suitable for recording. Timing capture circuit (200), e.g. a D flip-flop (DFF) or non-destructive readout 20 It may contain a superconducting digital logic element, such as an (NDRO) latch, and at the second output Triggered by the first SFQ pulse (OUTPUT2), it passes through a time measurement circuit (e.g., a TDC (or high-speed counter) enables the capture of time information corresponding to the peak moment. However, the invention is not limited to this, but also includes the transition moment at the end of the first output (OUTPUT1). SFQ pulse or OUTPUT1 stopping and OUTPUT2 starting simultaneously 25 It is also possible to determine this through a logical framework that evaluates the situation. This eliminates the need for additional intermediate processing between peak moment detection and time data recording. without being heard, low-delay time stamp generation for each photon event (210). This is possible, and reliable timing output is obtained by simplifying the system's timing chain. is obtained. 30 36 System Integration with SNSPD As shown in FIGURES 5 and 6, the single photon detection system (300) according to the present invention; detects a photon a superconducting nanowire capable of generating a transient output pulse in response to detection single photon detector (SNSPD) (310), magnetically coupled to the output of SNSPD (310) Based on the superconducting zero slope detection circuit (100) and the detected zero slope (peak) transition 5 a timing capture circuit suitable for recording a timestamp of photon arrival It contains (200). In a preferred application form, SNSPD (310), zero slope detection circuit (100) and timing capture circuit (200) at the same cryogenic temperature level (e.g. 4.2 K and / or (to operate under) a single superconducting chip or the same cryogenic package 10 This can be implemented on multiple chips positioned close together. Thanks to co-positioning, the signal is extracted from the cryogenic environment for timing determination. The need to transfer room temperature electronics (240) in analog form can be reduced or eliminated. This eliminates factors such as long cable routes, impedance transmission, and thermal noise. This helps prevent potential timing disruptions. 15 The output of the timing capture circuit (200) provides an additional analog discrimination at room temperature. instead of an analog signal requiring the process, timing information determined in the cryogenic layer room temperature electronics (240) as a corresponding digital timestamp (210) This digital output approach allows for the transmission of timing accuracy achieved in the cryogenic layer. It is suitable for preservation at the system level. Thanks to this, the time separation is 20 in the cryogenic layer. Once completed, it is sufficient to transfer only the timestamp information to the room temperature side. This reduces the need for additional analog discrimination and intermediate signal processing components. and contributes to the simplification of the system architecture. In an alternative application form, the timing capture circuit (200) cryogenic temperature solid (230) can be positioned at room temperature (240). In this configuration, zero slope detection 25 The circuit (100) and the detector (310) continue to work together in the cryogenic environment and Timing separation is performed in the cryogenic layer; however, the timing capture circuit... (200) The timestamp recording function is performed on the room temperature side. In this case, zero slope SFQ pulse transfer information from the detection circuit (100), cryogenic connection lines (e.g. digital 30 via superconducting microstrip transmission lines or cryogenic coaxial cables) The critical timing is transmitted to the room temperature timing capture circuit (200). Since the separation was completed in a cryogenic environment by the SQUID-based front end (100), 37 It is necessary to transmit the transient output pulse of the detector (310) to room temperature in analog form. This reduces signal degradation and additional sources of timing jitter. Multi-Pixel Array Configuration In another application, the system consists of two superconducting nanowires, each a single-photon detector. (SNSPD) (310) and will have a zero slope detection circuit (100) connected to it. It may contain multiple duplicated detection channels. In this application format, each channel... The obtained photon arrival timestamps are combined via the multiplexing circuit (400) to form a common It can be collected at an output and / or transmitted appropriately. Thus, timing for each pixel. High pixel count SNSPD array structures are supported while maintaining performance; this configuration, large-scale quantum switch 10 requiring multi-channel receiver architectures Quantum photonic computing requiring QKD distribution (QKD) systems and / or multiple detection channels. It provides a scalable solution for applications. Production Superconducting zero slope detection circuit (100) is widely used in the market and research infrastructures. These can be produced using niobium (Nb)-based superconducting circuit manufacturing processes. There are 15 such examples. Josephson joints (e.g., Nb / AlOx / Nb layered joint structures) are commonly used in these processes. It is used; however, the invention is not tied to a specific type of joint or a single production line. In practice, standard Nb-based facilities / processes suitable for the production of SFQ circuits (e.g. (AIST STP2, MIT Lincoln Laboratory SFQ5ee, SEEQC / Hypres and similar infrastructures) It can serve as an example for the implementation of the circuit. 20 The circuit (100) is operated below the temperatures at which niobium remains superconducting; this is practically Suitable for operation at approximately 4.2 K (liquid helium) or lower temperatures. This means that SNSPDs (310) can typically operate in the range of 0.8-2.5 K. Co-positioning of circuit (100) and SNSPD (310) on the same cryogenic temperature level In terms of temperature requirements, they are compatible. 25 In addition, the circuit (100) is limited to Nb only, depending on the application and manufacturing compatibility. This can also be achieved with other superconducting materials such as NbN or NbTiN without any limitations. In this way, the invention can be adapted to different production infrastructures and SNSPD-based systems. Integration can be facilitated. 38 Applications The invention is suitable for use in the applications exemplified below, and its areas of application are as follows: These are not the only ones. Timing for single-photon detection in Quantum Key Distribution (QKD) systems. Its sensitivity affects the correct setting of time windows, increasing the speed of reliable key production by 5. and is one of the factors that directly determines the achievable communication distance. The present invention, SNSPD (310) provides a suitable timing for determining the peak moment of the output pulse, It particularly helps reduce timing uncertainty in the cryogenic layer. This is why QKD Timing performance can be improved in the receiver architecture and at longer distances. Efficient key production may become possible. Referring to FIGURE 6, an example QKD 10 incorporating the invention is shown. in the system, the only one sent by the QKD transmitter (510) over the quantum channel (520) Photons are detected by one or more SNSPD (310), SNSPD (310) output The superconducting zero slope detection circuit (100) is connected and the time corresponding to the transition moment information is obtained as a digital timestamp (210) through the timing capture circuit (200). by performing the key elimination and error correction operations, the classical processing unit (530) 15 It is transmitted via the classical channel (540). Referring to FIGURE 7, linear optical quantum computing (LOQC) and measurement-based quantum computing Timing accuracy for single-photon detection in MBQC (Metal-Quick Conversion) architectures improves system performance. It is a critical parameter that directly affects [the system]. For example, quantum gates based on photon interference. In these processes, photons arriving from different paths are expected to be temporally indistinguishable; 20 As timing jitter in the detection chain increases, interference visibility may decrease, and this As a result, gate fidelity may be negatively affected. The present invention, with its zero-fidelity (passage) Using a slope / peak-based approach, it provides timing accuracy in the picosecond range. its suitability, the fact that the detection timing uncertainty limits the indistinguishability of photons It contributes to its reduction. 25 Architectures like MBQC also require fast feedforwarding; Information about a photon detection event occurring in one part of an optical circuit, transmitted to another... Delay for conditional driving of phase shifter / switching elements in the section It is desired that it be used without creating a zero slope detection circuit (100) SNSPD. The fact that the series (310) can be positioned together with the same cryogenic temperature level, the timing is 30 propagation resulting from the transmission of the signal to room temperature electronics and the associated additional It is suitable for reducing delay / uncertainty components. 39 In addition, the present invention has applications suitable for dual-function operation (e.g., Claims). 4) Single Flux Quantum (SFQ) generated at the first output (OUTPUT1) during positive slope by counting the impacts (the rise zone is shown enlarged) to the entry impact It is possible to obtain a rough amplitude indicator of the related information. This additional information can be used in some photonics. Classification of detection events in quantum computing protocols, post-selection 5 or it may provide an additional detection parameter that is useful in terms of auxiliary decision-making mechanisms. In this context, an example photonic quantum computing system that includes the invention is a photon source (610), an optical network consisting of linear optical elements (620), at the outputs of the optical network positioned and timing capture with a corresponding zero slope detection circuit (100) for each. 10 timestamps with an SNSPD array (310) connected to the circuit (200) concurrency assessment, feedforward and / or audit processes based on It can be arranged to include a classical controller (630) that performs the function. In space-to-ground or space-to-space optical communication systems, weak laser pulses at the receiver side. reliable detection and precise determination of pulse arrival time, connection budget. and is important in terms of achievable data rate. NASA's Deep Space Optical Communications 15 SNSPD in the (DSOC) system for receiving laser pulses from spacecraft. It has been reported that the series of SNSPD output pulses are used [4]. The present invention utilizes the output pulses of the SNSPD. Timing reference corresponding to the peak moment, unaffected by changes in amplitude. Because it can generate this, it helps reduce timing uncertainty in such systems and consequently... This can contribute to a more stable operation of the purchase-timing chain; thus, threshold 20 By reducing performance limitations related to time drift caused by trigger-based triggering. An architecture suitable for higher data rates can be achieved. The invention allows for the amplitude of transient signal pulses to vary from event to event, yet remain precise. It can be used in applications that require scheduling. In this context, time... Correlational single-photon counting (TCSPC), fluorescence lifetime imaging (FLIM), and single-photon 25 In scenarios such as light detection and ranging (LiDAR) with high sensitivity, amplitude-dependent Enables more consistent timestamping by reducing the time drift effect. This can provide additional room temperature discrimination to improve timing performance. The need for transmitter / signal processing components can be reduced, and the system architecture can be simplified. Industrial Applications of the Invention 30 The superconducting zero slope detection circuit (100), which is the subject of the present invention, is available on the market and in research standard niobium (Nb)-based superconducting circuit production available in their infrastructures 40 processes (e.g., AIST STP2, MIT Lincoln Laboratory SFQ5ee, SEEQC / Hypres) It is manufacturable. The basic components of the circuit are the Josephson junctions (J1, J2), inductances (LC, LIN, LP1, LP2) and polarization resistors (R1, R2) in a superconducting circuit These are standard components used in the design. Special materials or components are not required for the production of the circuit. No process development is needed. The circuit, together with SNSPD (310), has a cryogenic temperature of 5 Quantum key distribution (QKD), photonic quantum computing, located on the (230) floor, Time-correlated single-photon counting (TCSPC), LiDAR, and space optical communications, among others. It can be used in industrial settings.
Claims
1 REQUESTS 1. A SQUID containing a first Josephson joint (J1) and a second Josephson joint (J2). The ring (110) is magnetic to the SQUID ring (110) via a mutual inductance (M). an input coupling structure with a coupled input inductance (LIN) (120), A 5 that polarizes the first and second Josephson junctions (J1, J2) below their critical currents. The polarization source (VSQUID_BIAS) is a first output connected to the first Josephson junction (J1). including route (130) and a second exit route (140) connected to the second Josephson junction (J2), a timing reference corresponding to the peak moment of a transient input signal pulse a superconducting timing separation circuit (100) is configured to provide, Feature; 10 - each of the first and second Josephson joints (J1, J2), after switching Shunts that provide rapid zeroing in an overdamped regime without latching. being shunted with resistors (RS1, RS2), - input inductance (LIN) in the input coupling structure (120), in the SQUID loop (110) Having an inductive value that is asymmetrically large compared to the ring inductance (LC), 15 - feedback connection of the circuit from the output paths (130, 140) to the SQUID loop (110) The single flux quantum (SFQ) pulse that is not found and is in the first exit path (130) Temporal transition from activity to SFQ pulse activity in the second exit pathway (140) It has an open-loop topology that provides timing reference through it, - Timing error determined by mutual inductance (M) Φ0 / M current 20 It is limited by the quantization step and the upper bound of the timing error is the input. independent of the amplitude of the signal pulse It includes.
2. According to claim 1, the circuit is (100) and its characteristic is that the first and second shunt resistors are (RS1, RS2) each of them has a reset time of 25 of the input signal pulse for the corresponding Josephson junction. It is sized in such a way as to keep the rise time short.
3. According to claim 1, the circuit is (100) and its characteristic is that the timing reference is temporal. its sensitivity is the timing of the last SFQ pulse in the first exit path (130) or determined by the timing of the first SFQ pulse in the second exit path (140) That is. 30 4. According to claim 1, the circuit is (100) and its characteristic is; the input inductance (LIN) and the mutual inductance (M) in the first output path along the rising edge of the input signal pulse (130) It must be sized to generate at least two consecutive SFQ pulses. 2 5. According to claim 1, the circuit is (100) and its characteristic is that the circuit can receive any external clock signal. It is configured to operate asynchronously without requiring any additional resources.
6. According to claim 1, the circuit is (100) and its characteristic is that the nominal value of the polarization source is; including parasitic leakage currents leaking from the exit paths (130, 140) into the SQUID ring (110) the effective polarization ratio will remain below the oscillation threshold of Josephson joints 5 It is adjusted in this way.
7. According to claim 1, the circuit is (100) and its characteristic is that the input signal pulse is delayed and Instead of generating a bipolar signal from attenuated copies, a transient input signal in the first and second exit routes of the temporal transition corresponding to the peak moment of the coup (130, 140) SFQ pulse activities generated are determined via amplitude-independent 10 A functional alternative to the constant fraction discriminator (CFD) by providing a timing reference. It is the creation of.
8. According to claim 1, the circuit is (100) and its characteristic is; from the peak point (130) on the first exit path (330) then a limited number of SFQ pulses are generated, determined by the circuit parameters It has a repeatable and calibratable switching offset. 15 According to claim 1, the circuit is (100) and its characteristic is; the first SFQ pulse in the second output path (140). a timing system containing a superconducting digital logic element triggered by It is equipped with a capture circuit (200).
10. The peak moment of a transient input signal pulse applied to a superconducting circuit (100). is a method of establishing a corresponding timing reference, 20 feature; - the input signal pulse is an asymmetric large input to a SQUID ring (110) coupled via inductance (LIN) and mutual inductance (M) implementation, - a first Josephson 25 in response to the positive rate of change of the input signal pulse Generating SFQ pulses (130) in the first exit pathway from the joint (J1, - a second Josephson wave in response to the negative rate of change of the input signal pulse. Generating SFQ pulses in the second exit pathway (140) from the joint (J2), - from the SFQ pulse activity in the first exit path (130) to the second exit path (140) The SFQ pulse activity transition moment is determined as the timing reference, 30 - Timing error determined by the mutual inductance (M) Φ0 / M current It is limited by the quantization step and the upper bound of the timing error is the input. independent of the amplitude of the signal pulse, 3 - the method as an open loop without applying feedback to the SQUID loop (110) implementation It includes the steps.
11. The method according to claim 10, its characteristic is that the input signal pulse is passed through a superconducting nanowire. It is a pulse from a single photon detector (310). 5 12. The method according to claim 10, its characteristic is that the timing reference is the quantum key. distribution (QKD), positron emission tomography (PET), laser rangefinding (LIDAR) or the photon arrival time in one of the photonic quantum computing applications It is used in determining...
13. At least one superconducting single-photon detector (310) has a transient output pulse of 10⁻¹⁰ of the detector (310). a superconductor timing directly in the field via an asymmetric inductive coupling separation circuit (100) and timing information from the separation circuit (100) a single photon detection system (300) which includes a timing capture circuit (200), feature; - timing separation circuit (100), first Josephson junction (J1) and second Josephson 15 including joint (J2) and shunted with a shunt resistor (RS1, RS2) at each of the joints having a SQUID ring (110), - feedback connection of the timing separation circuit (100) to the SQUID ring (110) having an open loop structure which is not present and the first exit path (130) and the second exit Timing reference 20 via SFQ pulse activity transition between path (140). having a structure that provides - Timing error determined by the mutual inductance (M) Φ0 / M current being limited by the quantization step and the upper limit of the timing error independent of the output pulse amplitude of the detector (310), - detector (310) and timing separation circuit (100) in a common cryogenic environment 25 positioned It includes.
14. The system according to claim 13, and its characteristic is; having multiple superconducting single-photon detectors. (310) each of them is connected to a separate timing separation circuit (100).
15. The system is according to claim 13, and its feature is; the timing capture circuit (200) cryogenic 30 Being located outside the ambient temperature, the detector (310) temporary output timing pulse transmitted analogously to room temperature electronics the separation was carried out in a cryogenic environment and the timing of the separation 4 cryogenic connection lines of SFQ pulse transition information from circuit (100) It is transmitted digitally to the timing capture circuit (200) via.
