True Random Number Generators on IQM Spark Enable Enhanced Cryptography and Simulations
Summarize this article with:
True randomness underpins many critical technologies, from secure communications to accurate simulations and advanced machine learning, yet generating genuinely unpredictable numbers remains a significant challenge. Andrzej Gnatowski, Jarosław Rudy, and Teodor Niżyński, along with colleagues from Wrocław University of Science and Technology, now present a comprehensive study of true random number generation using superconducting quantum circuits. Their work addresses limitations in existing research, which often relies on simulations and tests only a small number of circuit designs, by utilising a real-life quantum computer, Odra 5. This represents the first investigation into true random number generators built on the IQM superconducting architecture, and the team’s analysis of 105 distinct circuit variations, each generating millions of random bits, offers a substantial step towards robust and reliable quantum-based randomness for future technologies. IBM Quantum Computers Generate Verified Randomness This research details a thorough evaluation of quantum random number generation (QRNG) using IBM quantum computers and the Odra 5 superconducting quantum computer, assessing the quality of randomness produced by various quantum circuits and configurations. Scientists implemented and tested different quantum circuits, employing Hadamard, rotation, and controlled-NOT gates, to generate random numbers, and rigorously tested 105 distinct circuit variations, each generating up to one million bits of random data. They then subjected these outputs to a comprehensive suite of statistical tests, including those defined by NIST SP 800-90B and SP 800-56C, as well as the robust Dieharder and TestU01 suites, providing a standardized measure of randomness. This work addresses a notable gap in previous research, which largely relied on simulations or limited testing of circuits on IBM quantum processors, and represents the first investigation of TRNG circuits on the IQM superconducting architecture. Researchers demonstrated that circuits mathematically equivalent can exhibit differing performance on real hardware, highlighting the importance of architecture-aware design, and isolated the impact of circuit design, gate choice, and temporal drift on statistical quality. The study explored various methods for preparing uniformly random measurement distributions, including the use of Hadamard, x-rotation, and y-rotation gates, and quantified randomness using entropy estimation.
Results demonstrate variations in circuit performance, with some designs generating more random sequences than others, and reveal that the number of qubits and their arrangement within a circuit impact the quality of the generated randomness. Statistical test results and entropy estimations provide a quantitative measure of randomness, and the research compares the performance of this QRNG implementation with existing methods and classical random number generators. Researchers acknowledge challenges associated with implementing QRNGs on noisy intermediate-scale quantum (NISQ) devices, such as decoherence and gate errors, and highlight the need for more research on alternative architectures to fully explore the capabilities of quantum TRNGs. Overall, this work contributes to the field by identifying promising approaches for building practical and reliable QRNGs, crucial for cryptography, simulations, and scientific research, and provides valuable insights into achieving high-quality randomness with quantum technology.
Superconducting Qubit Random Number Generation Study This work presents a comprehensive study of quantum random number generation (QRNG) using superconducting qubits, addressing limitations in previous research through a systematic, controlled approach on physical hardware. 9% for single-qubit gate fidelity and 97% for readout fidelity, and native gate flexibility. The study systematically evaluated 105 circuit subvariants across five distinct types of TRNG circuits, utilizing native Rx, Ry, and transpiled Hadamard gates. By performing all experiments on the same physical hardware under controlled conditions, scientists isolated the impact of circuit design and gate choice on statistical quality, eliminating variability introduced by different devices. This work represents the first investigation of TRNG circuits on the IQM superconducting architecture, enabling a detailed analysis unattainable with simulation alone.
The team meticulously examined 105 distinct circuit subvariants, each generating one million bits for rigorous statistical assessment. Experiments revealed that the quality of the random sequences generated could be comprehensively evaluated using the NIST SP 800-22 and NIST SP 800-90B test suites, providing a standardized measure of randomness. A key achievement of the study was the ability to isolate the impact of circuit design, gate choice, and temporal drift on statistical quality, ensuring observed differences stemmed from these factors rather than variations between devices. Researchers demonstrated that circuits mathematically equivalent can exhibit differing performance on real hardware, highlighting the importance of architecture-aware design. Measurements confirm that while these approaches are theoretically equivalent, their practical implementation differs due to the limitations of the IQM Spark processor, which natively supports only Rx, Ry, and CZ operations. Consequently, circuits requiring transpiled gates exhibit increased circuit depth and susceptibility to coherent errors.
The team’s analysis demonstrates that effective quantum algorithm design must minimize circuit depth and respect hardware constraints to maximize the reliability of random number generation. This work delivers a detailed understanding of the interplay between quantum circuit design, hardware limitations, and the generation of truly random numbers. Odra 5 Quantum Random Number Generation Tested This research presents a comprehensive study of True Random Number Generation (TRNG) circuits implemented on the Odra 5 superconducting quantum computer, representing a significant advancement in the field. Scientists rigorously tested 105 distinct circuit variations, each generating a substantial million bits of random data, and then assessed the quality of these sequences using established statistical tests from the NIST SP 800-22 and NIST SP 800-90B suites. This work addresses a notable gap in previous research, which largely relied on simulations or limited testing of circuits on IBM quantum processors.
The team’s investigation extends beyond simply generating random numbers; it provides a detailed analysis of circuit performance and contributes to a broader understanding of quantum-based randomness. Notably, this study is the first to utilize the IQM superconducting architecture for TRNG evaluation, opening new avenues for exploring different quantum platforms. While acknowledging the current limitations of qubit fidelity and the restricted gate sets available on existing hardware, the researchers demonstrate the potential of superconducting systems for generating high-quality random numbers. The authors recognize that the field is still developing and that most existing studies focus on IBM quantum processors. They highlight the need for more research on alternative architectures, such as those employing trapped ions or photonics, to fully explore the capabilities of quantum TRNGs. Future work could focus on optimizing circuit designs for specific hardware platforms and developing more robust statistical tests to evaluate the randomness of quantum-generated sequences. 👉 More information 🗞 True Random Number Generators on IQM Spark 🧠 ArXiv: https://arxiv.org/abs/2512.09862 Tags: