IEEE ITNEC (ISSN: 2693-3128)

FPGA-Based Real-Time Processing for Optical

Quantum Measurement and Control
Haozhi Wang1,2 ,Yaping Lu1,2 ,Yifei Chen1,2 ,Maojie Ma1,2 ,Jialin Xu2
2024 IEEE 7th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) | 979-8-3503-7080-5/24/$31.00 ©2024 IEEE | DOI: 10.1109/ITNEC60942.2024.10733136

1. School of Integrated Circuits and Electronics, Beijing Institute of Technology, Beijing, China
2. Beijing Naishu Electronics Co., Ltd., Beijing, China
[email protected], [email protected], [email protected], [email protected], [email protected]
Corresponding Author: Qiongzhi Wu Email: [email protected]

Abstract—In the context of rapid technological directions. Against this backdrop, fields such as quantum
advancements, progress in chip manufacturing technology computing, quantum communication, and quantum
is propelling electronic devices toward miniaturization and sensing have rapidly emerged, becoming frontiers of
high efficiency at an unprecedented pace. The fundamental scientific research and technological application.Quantum
principles of quantum mechanics are beginning to play a technologies demonstrate enormous potential in various
crucial role in modern electronic devices, providing the domains. For instance, they can significantly enhance
foundation for new technological methods and research information security, improve data processing speeds, and
directions. Optical quantum signal processing, as an
solve complex problems that are challenging for classical
important branch of quantum technology, utilizes the
computers [2]. These advantages drive the widespread
quantum properties of photons for information processing
and transmission, offering advantages such as high
application and in-depth study of quantum technologies
bandwidth, low noise, and strong resistance to interference. across multiple fields.
However, current optical quantum measurement and Optical quantum signal processing is one of the
control instruments face significant challenges in terms of important branches of quantum technology. It leverages
real-time performance and programming complexity, the quantum properties of photons for information
resulting in reduced data processing efficiency and processing and transmission, offering advantages such as
significantly increased development costs.
high bandwidth, low noise, and strong resistance to
This study implements a real-time processing algorithm
interference [3]. Consequently, it shows great potential in
for optical quantum signals on a Field-Programmable Gate fields like quantum communication, quantum computing,
Array (FPGA). The algorithm possesses advantages of real- and quantum precision measurement. However, this field
time performance, low power consumption, and high still faces numerous challenges, particularly in terms of
integration. It allows for rapid functionality switching and real-time performance and programming difficulty in
efficient optical quantum signal processing. The design practical applications. The current limitations of quantum
specifically aims to enhance the signal-to-noise ratio, measurement instruments in these aspects hinder their
significantly improving measurement accuracy and effectiveness and flexibility, posing obstacles to further
reliability. Experimental results indicate that this method research and practical application development [4].
substantially enhances the measurement precision and
signal-to-noise ratio of optical quantum signals. This To address these issues, this study implements a real-
research provides an efficient and reliable new method for time processing algorithm for optical quantum feedback
optical quantum signal processing, laying a solid foundation signals based on FPGA technology [5]. By leveraging the
for further studies in quantum communication, quantum high-performance computing and signal processing
computing, and quantum precision measurement. capabilities of FPGAs in conjunction with high-speed
Analog-to-Digital Converters (ADCs) and Digital-to-
Keywords—FPGA,Correlation Function,Optical Quantum Analog Converters (DACs), efficient real-time signal
Control,SNR processing and data analysis are achieved. This design not
only enhances the system's real-time performance and
I. INTRODUCTION integration but also improves the accuracy and reliability
of signal processing. Specifically, to address the typically
In the context of rapid technological advancements, low signal-to-noise ratio of optical quantum signals, we
progress in chip manufacturing technology continually developed specialized algorithms to enhance the signal-to-
drives the miniaturization and efficiency of electronic noise ratio (SNR), thereby improving measurement
devices [1]. According to Moore's Law, the number of accuracy. The outcomes of this research are detailed in
transistors on an integrated circuit continues to increase, subsequent sections, demonstrating its practical
leading to smaller component sizes and more pronounced application effectiveness in optical quantum signal
quantum effects. The fundamental principles of quantum processing.
mechanics play a critical role in modern electronic devices,
providing new technological approaches and research

979-8-3503-7080-5/24/$31.00 ©2024 IEEE 1922

Authorized licensed use limited to: Universidad Federal de Pernambuco. Downloaded on May 07,2025 at 14:07:15 UTC from IEEE Xplore. Restrictions apply.
 II. THEORETICAL ANALYSIS C. Second-Order Autocorrelation
To achieve efficient optical quantum signal processing, The second-order autocorrelation function is used to
this study designs several key technologies and algorithms, analyze the intensity fluctuations of a single optical field,
including cross-power spectral density, second-order serving as an important tool for studying photon statistical
autocorrelation function, second-order cross-correlation properties. It can reveal intrinsic properties of the optical
function, and cross-power autocorrelation function. field, such as photon bunching or anti-bunching effects
Additionally, to convert high-frequency modulated signals [10]. The second-order autocorrelation function g ( )
( 2)
into baseband signals that are easier to process, we
is represented as:
employ digital down-conversion and low-pass filtering
techniques [6]. We also utilize accumulation methods to I (t ) I (t   )
enhance the SNR. The following sections provide a g ( 2 ) ( )  2
(3)
detailed theoretical analysis: I (t )
A. Digital downconversion and low-pass filtering For example, lasers exhibit anti-bunching effects,
Optical quantum feedback signals are typically high- while thermal light sources show bunching effects. In
frequency modulated signals that need to be digitally quantum information processing, understanding and
down-converted within the FPGA to transform them into controlling photon statistical properties are fundamental to
baseband signals [7]. The digital down-conversion process achieving reliable quantum operations. The second-order
involves multiplying the received high-frequency signal autocorrelation function allows us to evaluate the purity of
by a local oscillator signal, resulting in a zero-carrier the quantum state of a light source, thereby optimizing
frequency baseband signal. The equation is as follows: quantum computing and communication systems. By
analyzing the second-order autocorrelation function, one
y(n)  x(n) cos( 2f c n)  j  x(n) sin( 2f c n) (1) can also distinguish between classical and non-classical
light sources. For instance, when g ( 2 ) (0)  1 , it indicates
Low-pass digital filtering is employed to enhance the that the light source is non-classical, which holds special
SNR as it effectively removes high-frequency noise application value in quantum technologies.
components, retaining only the primary components of the
baseband signal. The design parameters of the low-pass
D. Second-Order Cross-Correlation
filter are based on the spectral characteristics of the target
signal, ensuring that the main components of the signal The second-order cross-correlation function is used to
are preserved while suppressing high-frequency noise. analyze the correlation between two different optical
fields, serving as an essential tool for studying entangled
B. Cross-power spectral density photon states. This function can help determine whether
photon pairs are in an entangled state, thereby ensuring
The cross power spectral density(CPSD) is used to the correctness of quantum operations. The second-order
analyze the relationship between two different signals, ( 2)
with the notable advantage that uncorrelated noise cross-correlation function g xy ( ) is represented as:
approaches zero during the averaging process [8].
Suppose we have two independent detection chains ( 2)
I x (t ) I y (t   )
measuring the signal of the same optical field; these g xy ( )  (4)
signals will contain both the true signal and some noise. I x (t ) I y (t )
Given two signals x (t ) and y(t ) , their Fourier transforms
Ensuring good synchronization between photons is
are x ( ) and y( ) . The cross power spectral density crucial for the secure transmission and reception of
Sxy ( ) is represented as: quantum information. The second-order cross-correlation
function can be used to synchronously detect two different
S xy ( )  X ( )  Y  ( ) (2) photon signals, thereby improving the performance and
reliability of quantum communication [11]. In addition to
CPSD can accurately measure the spectral biphoton entanglement, the second-order cross-correlation
characteristics of an optical field, aiding in the function can also be used to study multiphoton entangled
understanding of the quantum states and coherence states, which is important for constructing quantum
properties of photons [9]. In quantum teleportation computers and complex quantum networks.
experiments, CPSD can confirm whether the entangled
state of photons has been successfully transmitted. By E. Cross Power Autocorrelation
ensuring that uncorrelated noise trends towards zero The cross power auto-correlation function combines
during averaging, the CPSD method effectively cross-channel and self-channel methods to achieve higher
suppresses noise and improves the SNR. This is measurement accuracy and reliability. By detecting
particularly important for experiments and applications signals from two optical quantum links S a [t ] and S b [t ] ,
requiring high precision measurements.

1923
Authorized licensed use limited to: Universidad Federal de Pernambuco. Downloaded on May 07,2025 at 14:07:15 UTC from IEEE Xplore. Restrictions apply.
 the second-order cross power auto-correlation function and consistency of ADC sampling, and examining and
Gab
( 2Y )
[ ] can be measured, which is represented as: optimizing the FPGA processing workflow to guarantee
the accuracy and efficiency of its results. Additionally, by
( 2Y )   adjusting the operating parameters of various modules,
Gab [ ]  ( S a S b )  ( S a S b ) [ ] (5) such as the cutoff frequency of filters and the carrier
frequency of mixers, the system performance can be
By integrating cross-channel and self-channel methods, optimized.
this approach significantly enhances measurement
precision, making it particularly suitable for tasks B. Signal generation and data acquisition
requiring high sensitivity and resolution, such as quantum This experiment utilizes a custom-designed board
metrology and high-precision spectral analysis. The cross based on RFSoC to transmit pre-set frequency and
power auto-correlation function effectively suppresses specific waveform signals to the optical quantum chip, as
uncorrelated noise, thereby improving the signal-to-noise illustrated in Figure 1. The board integrates an RF front-
ratio, which is highly beneficial for precise measurements end, DAC, and ADC, ensuring the consistency and
in noisy environments, such as in practical applications of accuracy of the input signals. The DAC is used to play
quantum sensors. This method enables the simultaneous waveform signals with specific frequencies and
analysis of relationships among multiple signal channels, amplitudes to the optical quantum chip, putting it in an
providing a more comprehensive and accurate system excited state [14]. Subsequently, the high-speed ADC on
description. For instance, in the analysis of multi-mode the RFSoC performs signal acquisition, converting the
optical fields, the cross power auto-correlation function analog feedback from the optical quantum chip into
can reveal complex interactions between different modes. digital signals. Additionally, multiple repeated samplings
are conducted to enhance data reliability and
F. Cumulative Processing for Enhanced SNR reproducibility.
Accumulation is an effective method for improving
the SNR. By repeatedly sampling and accumulating the
signal multiple times, the signal amplitude increases
linearly with the number of accumulations since it is
coherent, while the power of the noise, being random,
does not significantly increase [12]. Assuming the number
of accumulations is N, the improvement in SNR can be
expressed as:

SNRnew  N  SNRoriginal (6)

This method reduces the impact of random noise,

making the resulting signal clearer and statistically
significant [13]. These theoretical analyses directly guided
the subsequent implementation of algorithms on the
FPGA. Each algorithm and technique corresponds to
specific hardware implementation challenges, such as
high computational load or complex data flow control.
Fig. 1. Block diagram of the experiment.
III. EXPERIMENTAL METHODS AND RESULTS
C. FPGA processing algorithm
To validate the effectiveness of the aforementioned
theoretical methods, this study undertakes a detailed The process of handling high-frequency signals on an
experimental design and implementation. Through FPGA is as follows: As shown in Figure 2, the high-
systematic experimental procedures—including calibration, frequency signal is first downconverted to a baseband
signal using hardware multipliers and mixers.
data acquisition, FPGA processing, data storage and
Subsequently, an FIR filter is employed to remove noise.
transmission, followed by data analysis and result
In the core algorithm part, the FPGA's FFT IP core is used
comparison — this research covers all critical aspects of for Fast Fourier Transform, followed by conjugate
optical quantum signal processing. multiplication using hardware multipliers. The cross-
power spectral density is computed by accumulating the
A. System calibration results of multiple experiments. The autocorrelation and
To ensure the accuracy and reliability of the entire cross-correlation functions are calculated in the time
experimental system, system calibration is required both domain using multipliers and adders. To improve the SNR,
before and during the experiment. The calibration process the FPGA sums the results of each experiment using
includes verifying the frequency and waveform accuracy parallel accumulation registers and control logic to reduce
of the RFSoC transmitted signals, ensuring the precision random noise. The entire process begins with the high-

1924
Authorized licensed use limited to: Universidad Federal de Pernambuco. Downloaded on May 07,2025 at 14:07:15 UTC from IEEE Xplore. Restrictions apply.
 speed ADC capturing the analog signal, followed by IV. SUMMARY AND OUTLOOK
downconversion, low-pass filtering, FFT computation,
This study implements a real-time processing
and correlation operations. The final results are stored in
algorithm for optical quantum feedback signals on an
DDR memory and transmitted to a host computer for
FPGA, fully utilizing the flexible hardware configuration
analysis via the Peripheral Component Interconnect
and high computational efficiency of the FPGA, as well as
Express (PCIe) interface.
the high-precision acquisition and output capabilities of
high-speed ADC and DAC modules [16]. Experimental
validation demonstrates that this method significantly
enhances the measurement accuracy and SNR of optical
quantum signals. It improves the real-time performance
and reduces the programming complexity of optical
quantum measurement and control instruments, thereby
increasing data processing efficiency and reducing
Fig. 2. Algorithm flowchart. development costs.
Despite significant progress, this study still faces
D. Data Storage and Transmission several challenges and offers broad research prospects:
The data processed by the FPGA, along with the
originally acquired digital signals, will be packaged and 1) System Optimization: Further optimization of the
stored in Double Data Rate Synchronous Dynamic hardware architecture and algorithms on the FPGA is
Random Access Memory (DDR). The host computer required. This includes improving the utilization rate of
retrieves the stored data from the DDR via the PCIe, FPGA logic units, enhancing the computational
ensuring that the experimental data can be transmitted to efficiency of the ARM processor, and incorporating real-
the host computer quickly and accurately. time signal processing techniques to handle complex
optical quantum signal processing tasks.
E. Data analysis 2) Expanding Application Domains: Extending the
After receiving the data, the host computer compares existing technologies and methods to fields such as
and analyzes the results from different algorithms to quantum sensing, quantum imaging, and quantum
visually observe the distribution properties and correlation metrology is essential. This involves developing
characteristics of the optical quantum signals. The specialized algorithms and hardware modules tailored for
reliability and SNR of the data are enhanced through different application scenarios to enhance the system's
repeated experiments. At the end of each experiment, the
versatility and adaptability.
FPGA accumulates and sums the processing results of
each trial to further improve signal quality and reduce the 3) Expanding Application Domains: Extending the
impact of random noise [15]. As shown in Figure 3, this existing technologies and methods to fields such as
process results in an SNR improvement of 35 dB, making quantum sensing, quantum imaging, and quantum
the final signal clearer and statistically significant. metrology is essential. This involves developing
specialized algorithms and hardware modules tailored for
different application scenarios to enhance the system's
versatility and adaptability.
Despite significant progress, this study still faces
several challenges and offers broad research prospects:
In summary, this study provides an efficient and
reliable new method for optical quantum signal processing
and points the way for further research in the fields of
quantum communication, quantum computing, and
quantum precision measurement. This research aims to
continuously advance quantum technology development,
facilitating broader technological applications and
scientific exploration.

REFERENCES
[1] S. Xie et al., "Entangled Photon Pair Source Demonstrator Using
the Quantum Instrumentation Control Kit System," in IEEE
Fig.3. Quantum signal waveform.
Journal of Quantum Electronics, vol. 59, no. 5, pp. 1-7, Oct. 2023,
Art no. 9300307, doi: 10.1109/JQE.2023.3302926.
[2] Y. -L. Cheng, B. -H. Chen, Y. -P. Liao, R. -C. Lu, C. Y. Chen and
T. W. Huang, "FPGA Implementation of a Real-Time Quantum
Image Encryption System Based on Reversible Quantum Gate

1925
Authorized licensed use limited to: Universidad Federal de Pernambuco. Downloaded on May 07,2025 at 14:07:15 UTC from IEEE Xplore. Restrictions apply.
 Computing," 2024 10th International Conference on Applied Romeoville, IL, USA, 2023, pp. 406-411, doi:
System Innovation (ICASI), Kyoto, Japan, 2024, pp. 7-9, doi: 10.1109/eIT57321.2023.10187252.
10.1109/ICASI60819.2024.10547832. [10] J. Chen, L. Wang and B. Wang, "Quantum FPGA architecture
[3] X. Zhang et al., "The Quantum Shor Algorithm Simulated on design," 2013 International Conference on Field-Programmable
FPGA," 2019 IEEE Intl Conf on Parallel & Distributed Processing Technology (FPT), Kyoto, Japan, 2013, pp. 354-357, doi:
with Applications, Big Data & Cloud Computing, Sustainable 10.1109/FPT.2013.6718386.
Computing & Communications, Social Computing & Networking [11] Y. H. Lee, M. Khalil-Hani and M. N. Marsono, "FPGA-based
(ISPA/BDCloud/SocialCom/SustainCom), Xiamen, China, 2019, quantum circuit emulation: A case study on Quantum Fourier
pp. 542-546, doi: 10.1109/ISPA-BDCloud-SustainCom- transform," 2014 International Symposium on Integrated Circuits
SocialCom48970.2019.00082. (ISIC), Singapore, 2014, pp. 512-515, doi:
[4] R. D. Oliveira, S. Bahrani, E. Arabul, R. Wang, R. Nejabati and D. 10.1109/ISICIR.2014.7029495.
Simeonidou, "FPGA-based Deterministic and Low-Latency [12] M. Hart, J. M. Allister, L. Rogers and C. Gillan, "An Emulation of
Control for Distributed Quantum Computing," IEEE INFOCOM Quantum Error-Correction on an FPGA device," 2021 31st
2023 - IEEE Conference on Computer Communications International Conference on Field-Programmable Logic and
Workshops (INFOCOM WKSHPS), Hoboken, NJ, USA, 2023, pp. Applications (FPL), Dresden, Germany, 2021, pp. 104-108, doi:
1-6, doi: 10.1109/INFOCOMWKSHPS57453.2023.10226129. 10.1109/FPL53798.2021.00025.
[5] X. Qin et al., "An FPGA-Based Hardware Platform for the [13] Y. -P. Liao, Y. -L. Cheng, Y. -T. Zhang, H. -X. Wu and R. -C. Lu,
Control of Spin-Based Quantum Systems," in IEEE Transactions "The interactive system of Bloch sphere for quantum computing
on Instrumentation and Measurement, vol. 69, no. 4, pp. 1127- education," 2022 IEEE International Conference on Quantum
1139, April 2020, doi: 10.1109/TIM.2019.2910921. Computing and Engineering (QCE), Broomfield, CO, USA, 2022,
[6] Y. Hong, S. Jeon, S. Park and B. -S. Kim, "Quantum Circuit pp. 718-723, doi: 10.1109/QCE53715.2022.00097.
Simulator based on FPGA," 2022 13th International Conference [14] Y. Jungjarassub and K. Piromsopa, "A Performance Optimization
on Information and Communication Technology Convergence of Quantum Computing Simulation using FPGA," 2022 19th
(ICTC), Jeju Island, Korea, Republic of, 2022, pp. 1909-1911, doi: International Conference on Electrical Engineering/Electronics,
10.1109/ICTC55196.2022.9952408. Computer, Telecommunications and Information Technology
[7] Y. Xu et al., "QubiC: An Open-Source FPGA-Based Control and (ECTI-CON), Prachuap Khiri Khan, Thailand, 2022, pp. 1-4, doi:
Measurement System for Superconducting Quantum Information 10.1109/ECTI-CON54298.2022.9795571.
Processors," in IEEE Transactions on Quantum Engineering, vol. [15] A. G. Mazăre, L. Mihai Ionescu, I. Liță, G. Șerban and N. Belu,
2, pp. 1-11, 2021, Art no. 6003811, doi: "New FPGA design solution using quantum computation
10.1109/TQE.2021.3116540. concepts," 2021 IEEE 27th International Symposium for Design
[8] B. Branchini, D. Conficconi, D. Sciuto and M. D. Santambrogio, and Technology in Electronic Packaging (SIITME), Timisoara,
"The Hitchhiker's Guide to FPGA-Accelerated Quantum Error Romania, 2021, pp. 388-391, doi:
Correction," 2023 IEEE International Conference on Quantum 10.1109/SIITME53254.2021.9663653.
Computing and Engineering (QCE), Bellevue, WA, USA, 2023, [16] H. Johnson, S. Zorzetti and J. Saniie, "Improving Signal-to-Noise
pp. 338-339, doi: 10.1109/QCE57702.2023.10271. Ratio (SNR) for Readout Signals Using Adaptive Filters on
[9] H. Johnson, S. Zorzetti and J. Saniie, "Exploration of Optimizing Reconfigurable Controls Hardware," 2023 IEEE International
FPGA-based Qubit Controller for Experiments on Conference on Quantum Computing and Engineering (QCE),
Superconducting Quantum Computing Hardware," 2023 IEEE Bellevue, WA, USA, 2023, pp. 403-404, doi:
International Conference on Electro Information Technology (eIT), 10.1109/QCE57702.2023.10303.

1926
Authorized licensed use limited to: Universidad Federal de Pernambuco. Downloaded on May 07,2025 at 14:07:15 UTC from IEEE Xplore. Restrictions apply.