 Research
 Open access
 Published:
Application of the QDST algorithm for the Schrödinger particle simulation in the infinite potential well
EPJ Quantum Technology volume 11, Article number: 12 (2024)
Abstract
This paper examines whether a quantum computer can efficiently simulate the time evolution of the Schrödinger particle in a onedimensional infinite potential well. In order to solve the Schrödinger equation in the quantum register, an algorithm based on the Quantum Discrete Sine Transform (QDST) is applied. The paper compares the results obtained in this way with the results given by the previous method (based on the QFT algorithm).
1 Introduction
In the near future, quantum calculations can make a major contribution to the development of informatics [1]. Nowadays some institutions claim to have a quantum computer and offer its computing power. Therefore, it is worth examining the properties of such machines.
For many years we have known Shor [2] and Grover [3, 4] algorithms which are of lower computational complexity than their best classical counterparts. Another promising application of quantum computer are quantum simulations [5–8], i.e. the computer modeling of behavior of physical quantum systems. It gives the possibility of effective modeling quantum processes, which is not possible using classical computers [9]. Quantum computers can simulate a wide variety of quantum systems, including fermionic lattice models [10–12], Heisenberg chains [13], Lattice Boltzmann methods [14] quantum chemistry [15, 16], field theories [17, 18], gravitational waves [19], and paradoxes in quantum mechanics [20].
As is well known, simulations of quantum systems performed using conventional computers are not effective. This means that for classical computer the memory resources and time required to simulate grow exponentially with the size of quantum system. In the case of a quantum computer, the situation is different. The relationship between the size of quantum computer (register) and the size of the simulated quantum system is linear. Therefore, a very important task is to find the appropriate algorithms that can properly simulate complex quantum systems and nontrivial interactions between them. This is a difficult issue, because most of the interesting quantum systems is feasible in infinitelydimensional Hilbert spaces. In such situations, we can use the technique of sampling the wave function and build an algorithm based on the Quantum Fourier Transform (QFT). This case was tested in [21–24], which examined the free particle and the harmonic oscillator.
In our previous works algorithms based on the QFT have been used. We have successfully simulated both the Schrödinger particle with a continuous spectrum [25], and a particle trapped in a potential well with finite length [26]. In this paper we examine a modification of that approach. We show that algorithm based on the Quantum Discrete Sine Transform (QDST) can be successfully used for simulation of the Schrödinger particle inside infinite potential well. We also demonstrate that this approach has significant advantages.
The physical system we simulate here is simple and is successfully simulated on classical computers. However, the presented algorithm can be easily generalized to the threedimensional case. Moreover, it may be part of a larger simulation of multimolecular processes. In such situations, the use of quantum computer offers significant advantages.
The possibility of using a quantum register to simulate the Schrödinger particle has already been discussed in the literature. For example, in [27] simulation of noisy Schrödniger equation is examined. It is also worth mentioning about [28, 29], where technique based on the quantum latticegas model was examined. Our approach is completely different. We examine the problem on purely algorithmic grounds, using abstract model of quantum system. We abstract completely from specified physical implementation of the quantum register.
As is well known, quantum transforms are widely used in quantum processing. We can mention here the Quantum Fourier Transform [30, 31] and the cosine transform [32], which is used in image compression issues [33, 34]. The wavelet [35, 36] (e.g. in image processing [37, 38]) and the WalshHadamard [39] transforms are also considered.
In order to simulate a quantum register, we used an environment written in C++ language. We also performed several tests of the algorithm with the use of IBM hardware [40]. In both cases we obtained the same results. Python (Qiskit) code of the algorithm main blocks can be found in Appendix E.
2 Description of the simulated system
In this work we consider the onedimensional Schrödinger particle in a infinite potential well, as shown in Fig. 1. Our main purpose is the simulation of time evolution of such system in the quantum register.
Time evolution of the Schrödinger particle in a onedimensional case takes the form:
where \(H_{0}=p^{2}/(2m)\) is free Hamiltonian and \(V(x)\) is the potential of the well from Fig. 1. The time evolution operator corresponding to Eq. (1) can be approximated as follows:
where \(\mathcal{F}\) is the Quantum Fourier Transform. This method has been tested in paper [26]. In this work we show that algorithm based on the Quantum Discrete Sine Transform (QDST) can be used as well. Moreover, there are advantages to this approach. Both the QFT and the QDST algorithms applied to the quantum particle state allow the transition from the position to the momentum representation (stationary states). However, the QDST transformation is dedicated to states that meet the Dirichlet boundary conditions. Thus, it maps much more accurately the state of the particle to stationary states in the well (Eq. (C.4)). The accuracy of the mapping is not the only advantage. The fulfillment of the condition \(\psi _{S}(a)=\psi _{S}(a)=0\) means that the potential \(V(x)\) from Fig. 1 is implemented automatically. The block implementing \(\exp (i V(x)\Delta t/\hbar )\) operator is not required. For this reason the algorithm can be run only in the energy domain and for time step of any value (i.e. the Trotter time step dt with error \(O(\text{d}t^{2})\) [41] is not required). The algorithm based on the QFT (examined in an earlier work [26]) does not show the advantages described above.
3 Description of the algorithm
In the simulation a \(n_{q}\)qubit register has been used. The state of the Schrödinger particle is encoded in all qubits exept the last one. Initial state of the last qubit is \(\vert 1\rangle \), as shown in Fig. 2. It is an auxiliary qubit for the QDST algorithm (see Appendix B). The state of the Schrödinger particle is sampled and encoded in the subregister state in the following way:
where \(n_{x}=n_{q}1\), \(\psi _{S}(x)\) is the Schrödinger particle wave function, \(\Delta{x}=2a/2^{n_{x}}\) is the distance between adjacent spatial samples. The \(\psi _{i}\) numbers are probability amplitudes of ith subregister base state (i.e. \(\psi _{0}= _{n_{x}1} \langle 0\vert \dots _{1} \langle 0\vert _{0} \langle 0\vert \Psi \rangle \), \(\psi _{1}= _{n_{x}1} \langle 0\vert \dots _{1} \langle 0\vert _{0} \langle 1\vert \Psi \rangle \), \(\psi _{2}= _{n_{x}1} \langle 0\vert \dots _{1} \langle 1\vert _{0} \langle 0\vert \Psi \rangle \), etc., where \(\vert \Psi \rangle \) is the subregister state). Thus, amplitude \(\psi _{0}\) corresponds to \(\psi _{S}(a)\), while amplitude \(\psi _{2^{n_{x}}1}\) corresponds to \(\psi _{S}(+a\Delta{x})\).
The scheme of the algorithm is shown in Fig. 2. The first QDST block implements the QDST algorithm which changes the state of the particle from the position to the energy representation (the stationary states given by Eq. (C.4)). A detailed description of the QDST algorithm implementation can be found in [32] (see also Appendix B). The FES block implements free evolution of the particle (\(\hat{U}(t)=\exp (\text{i}t\hat{p}^{2})/(2\hbar m)\) operator). Its implementation is shown in Fig. 3. Gates \(P_{\phi}\) are phase shift gates (\(\vert 1\rangle \rightarrow \text{e}^{i\phi}\vert 1\rangle \)), and controlled phase shift gates (\(\vert 1,1\rangle \rightarrow \text{e}^{i \phi}\vert 1,1\rangle \)). Implementation of the FES block is based on decomposition of the \(\hat{U}(t)\) operator in the following way:
where \(\alpha =\pi ^{2}\hbar t/(8ma^{2})\), n is level number in the well (see Eq. (C.3)), and \(i_{k}\) are binary digits of n (\(n= \sum_{k}2^{k}i_{k}\)). The first product from Eq. (4) corresponds to onequbit gates from Fig. 3, while the second one (over \(k_{1}\) and \(k_{2}\)) corresponds to twoqubits gates (\(k_{1}\) is position of the gate, while \(k_{2}\) is its control qubit number). The second QDST block implements the QDST^{−1} transformation which is identical with the QDST transformation.
4 The simulation results
The algorithm has been tested in the \(n_{q}=9\) qubit quantum register. We examined both the QDST algorithm from Fig. 2 and the algorithm based on the QFT transformation presented in [26]. In both cases we simulated a particle with the mass of electron in a well with length \(2a=4\) nm. In the case of the QFT algorithm time step is \(dt=10^{18}s\). The results of simulation are shown in Figs. 4–7. In all plots only single curves are visible because the results of both algorithms coincide with each other.
Plots from Fig. 4 show example of the stationary state given by Eq. (C.4) for \(n=3\). In the next figures propagation of the localized Gaussian state (given by Eq. (D.1)) is tested. In the case from Fig. 5 initial packet parameters take the following values: \(dx=0.04a\), \(\langle{x}\rangle =0.6a\) and \(T=25\) eV. In the case from Fig. 6 packet with parameters: \(dx=0.04a\), \(\langle{x}\rangle =+0.5a\) and \(T=150\) eV has been used. In all figures the left plots show the particle state in the position representation. Numbers on the horizontal axes correspond to spatial sample values (sample no. 0 corresponds to \(x=a\) while sample no. 256 corresponds to \(x=+a\)). The right plots show the particle state in the momentum representation. Numbers on the horizontal axes correspond to n (number of the stationary state in the well).
Plot from Fig. 7 shows results from the IBM simulator (ibmq_qasm_simulator) [40]. We examine exactly the same case as in Fig. 6. As we can see, the results from both simulations are similar.
5 Conclusions

Both methods (the QDST and the QFT) give convergent results. The states in the position representation coincide with each other for all three cases from Figs. 46.

The QFT method does not calculate exactly the number of the stationary level inside the well (the right plot from Fig. 4). This affects the accuracy of the simulation as shown in the left plot from Fig. 4.

Preparation of the initial state was excluded from the discussion. A simple algorithm for inputting these types of states into the quantum register has been proposed in [42].

Implementation of the QDST transform requires more gates than implementation of the QFT transform (according to [32]). Therefore, in the case of a free Schrödinger particle, a QFTbased simulation is faster. In the case of a particle trapped inside a well (regardless of its initial state), a simulation based on QDST is definitely better because it does not require the Trotter time step.

In the QDST method the well potential \(V(x)\) is implemented automatically. For this reason the algorithm can be run only in the momentum representation and the simulation can be performed in one time step. This is a particularly important advantage for simulations on currently available quantum computers, which are characterized by high error rates.
Data availability
No datasets were generated or analysed during the current study.
Code availability
Not applicable.
Abbreviations
 DCT:

Discrete Cosine Transform
 DST:

Discrete Sine Transform
 FES:

free evolution block
 QDCT:

Quantum Discrete Cosine Transform
 QDST:

Quantum Discrete Sine Transform
 QFT:

Quantum Fourier Transform
References
Feynman R. Int J Theor Phys. 1982;21:467–88.
Shor PW. Algorithms for quantum computation: discrete logarithms and factoring. In: Proc 35th Ann. Symp. Found. Comp. Sci. Los Alamitos: IEEE Comput. Soc.; 1994. p. 124. https://doi.org/10.1109/SFCS.1994.365700.
Grover LK. From Schrödinger equation to the quantum search algorithm. Am J Phys. 2001;69:769–77. https://doi.org/10.1119/1.1359518.
Sakhouf H, Daoud M, Laamara RA. Implementation of Grover’s search algorithm in the QED circuit for two superconducting qubits. Int J Theor Phys. 2020;59:3436–48. https://doi.org/10.1007/s10773020046021.
Lloyd S. Universal quantum simulators. Science. 1996;273(5278):1073–8. https://doi.org/10.1126/science.273.5278.1073.
Johnson TH, Clark SR, Jaksch D. What is a quantum simulator? EPJ Quantum Technol. 2014;1:10. https://doi.org/10.1140/epjqt10.
Schaetz T, Monroe CR, Esslinger T. Focus on quantum simulation. New J Phys. 2013;15:085009. https://doi.org/10.1088/13672630/15/8/085009.
Lanyon BP, Hempel C, Nigg D, Müller M, Gerritsma R, Zähringer F, Schindler P, Barreiro JT, Rambach M, Kirchmair G, Hennrich M, Zoller P, Blatt R, Roos CF. Universal digital quantum simulation with tapped ions. 2011. https://doi.org/10.1126/science.1208001. http://xxx.lanl.gov/arXiv:1109.1512v2.
Childs AM, Maslov D, Nam Y, Ross NJ, Su Y. Toward the first quantum simulation with quantum speedup. Proc Natl Acad Sci. 2018;115(38):9456–61. https://doi.org/10.1073/pnas.1801723115.
Wecker D, Hastings MB, Wiebe N, Clark BK, Nayak C, Troyer M. Solving strongly correlated electron models on a quantum computer. Phys Rev A. 2015;92:062318. https://doi.org/10.1103/PhysRevA.92.062318.
Kokail C, Maier C, van Bijnen R, Brydges T, Joshi MK, Jurcevic P, Muschik CA, Silvi P, Blatt R, Roos CF, Zoller P. Selfverifying variational quantum simulation of lattice models. Nature. 2019;569:16. https://doi.org/10.1038/s4158601911774.
Lamata L, Mezzacapo A, Casanova J et al.. Efficient quantum simulation of fermionic and bosonic models in trapped ions. EPJ Quantum Technol. 2014;1:9. https://doi.org/10.1140/epjqt9.
Grass T, Lewenstein M. Trappedion quantum simulation of tunablerange Heisenberg chains. EPJ Quantum Technol. 2014;1:8. https://doi.org/10.1140/epjqt8.
Succi S, FillionGourdeau F, Palpacelli S. Quantum lattice Boltzmann is a quantum walk. EPJ Quantum Technol. 2015;2:12. https://doi.org/10.1140/epjqt/s4050701500251.
Wecker D, Bauer B, Clark BK, Hastings MB, Troyer M. Gate count estimates for performing quantum chemistry on small quantum computers. Phys Rev A. 2014;90:022305. https://doi.org/10.1103/PhysRevA.90.022305.
Hempel C, Maier C, Romero J, McClean J, Monz T, Shen H, Jurcevic P, Lanyon BP, Love P, Babbush R, Guzik AA, Blatt R, Roos CF. Quantum chemistry calculations on a trappedion quantum simulator. Phys Rev X. 2018;8:031022. https://doi.org/10.1103/PhysRevX.8.031022.
Jordan SP, Lee KSM, Preskill J. Quantum algorithms for quantum field theories. Science. 2012;336:1130–3. https://doi.org/10.1126/science.1217069.
Sinha S, Russer P. Quantum computing algorithm for electromagnetic field simulation. Quantum Inf Process. 2010;9:385–404.
Bravo T, Sabín C, Fuentes I. Analog quantum simulation of gravitational waves in a BoseEinstein condensate. EPJ Quantum Technol. 2015;2:3. https://doi.org/10.1140/epjqt16.
Hou T, Ding D, Wang C et al.. Quantum simulation of generalized Hardy’s paradox and corresponding Hardy’s inequality via quantum programming. Int J Theor Phys. 2021;60:1972–9. https://doi.org/10.1007/s1077302104815y.
Wiesner S. Simulation of manybody quantum systems by a quantum computer. http://xxx.lanl.gov/quantph/9603028.
Zalka C. Efficient simulation of quantum system by quantum computers. Fortschr Phys. 1998;46:877–9. https://doi.org/10.1002/(SICI)15213978(199811)46:6/8<877::AIDPROP877>3.0.CO;2A.
Strini G. Error sensitivity of a quantum simulator I: a first example. Fortschr Phys. 2002;50:171–83. https://doi.org/10.1002/15213978(200203)50:2<171::AIDPROP171>3.0.CO;2B.
Benenti G, Strini G. Quantum simulation of the singleparticle Schrödinger equation. http://xxx.lanl.gov/arXiv:0709.1704v2.
Ostrowski M. Quantum simulation of two interacting Schrödinger particles. Open Syst Inf Dyn. 2016;23(4):1650020. https://doi.org/10.1142/S1230161216500207.
Ostrowski M. Simulation of diffusion of a single Schrödinger particle in the quantum register. Acta Phys Pol A. 2020;137(6):1182–6. https://doi.org/10.12693/APhysPolA.137.1182.
Garg N, Parthasarathy H, Upadhyay DK. Realtime simulation of HP noisy Schrödinger equation and Belavkin filter. Quantum Inf Process. 2017;16:121.
Yepez J, Boghosian B. An efficient and accurate quantum latticegas model for the manybody Schrödinger wave equation. Comput Phys Commun. 2002;146:280–94. https://doi.org/10.1016/S00104655(02)004198.
Yepez J, Vahala G, Vahala L. Quantum Inf Process. 2006;4(6):457–69. https://doi.org/10.1007/s1112800500088.
Yin H, Lu D, Zhang R. Quantum windowed Fourier transform and its application to quantum signal processing. Int J Theor Phys. 2021;60:3896–918. https://doi.org/10.1007/s10773021049337.
Zhu H, Zhang Y, Li Z. Efficient quantum blind signature scheme based on quantum Fourier transform. Int J Theor Phys. 2021;60:2311–21. https://doi.org/10.1007/s10773021048545.
Klappenecker A, Rötteler M. Discrete cosine transforms on quantum computers. 2001. arXiv:quantph/0111038v1.
Pang CY, Zhou RG, Hu BQ, Hu WW, ElRafei A. Signal and image compression using quantum dicrete cosine transform. Inf Sci. 2019;473:121–41.
Jiang N, Lu X, Hu H, Dang Y, Cai Y. A novel quantum image compression method based on JPEG. Int J Theor Phys. 2018;57:611–36.
Fijany A, Williams CP. Quantum wavelet transform: fast algorithms and complete circuits. 1998. arXiv:quantph/9809004.
Li HS, Fan P, Xia Hy, Song S. Quantum multilevel wavelet transforms. Inf Sci. 2019;504:113–35.
Wang J, Geng YC, Liu JQ. Adaptive quantum image encryption method based on wavelet transform. 2019. arXiv:1901.07762.
Chakraborty S, Shaikh SH, Chakrabarti A et al.. An image denoising technique using quantum wavelet transform. Int J Theor Phys. 2020;59:3348–71. https://doi.org/10.1007/s10773020045902.
Hoyer P. Efficient quantum transforms. 1997. arXiv:quantph/9702028.
Childs AM, Su Y, Tran MC, Wiebe N, Zhu S. Theory of Trotter error with commutator scaling. Phys Rev X. 2021;11:011020. https://doi.org/10.1103/PhysRevX.11.011020.
Ostrowski M. Loading initial data into the quantum register. J Appl Comput Sci. 2020;28(1):39–49.
Funding
The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
Author information
Authors and Affiliations
Contributions
This paper is written entirely by Marcin Ostrowski
Corresponding author
Ethics declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix A: The Discrete Sine Transform (DST1)

formal definition of the DST1 transform:
$$ X_{k}=\sqrt{\frac {2}{N+1}}\sum _{n=0}^{N1}x_{n}\sin \biggl( \frac {\pi}{N+1}(n+1) (k+1) \biggr) \quad \text{for } k=0,\dots ,N1 $$(A.1) 
inverse transformation is given by the same formula:
$$ x_{n}=\sqrt{\frac {2}{N+1}}\sum _{k=0}^{N1}X_{k}\sin \biggl( \frac {\pi}{N+1}(n+1) (k+1) \biggr) \quad \text{for } n=0,\dots ,N1 $$(A.2) 
inserting \(X_{k}=(1,0\dots 0)\) to Eq. (A.2) we obtain:
$$ x_{n}=\sqrt{\frac{2}{N+1}}\sin \biggl(\frac {\pi}{N+1}(n+1) \biggr). $$(A.3)Samples \(x_{1}=x_{N}=0\). Therefore N is number of nonzero samples. Function period is equal to \(N+1\).

kth sample \(X_{k}\) from Eq. (A.1) corresponds to the stationary state \(\psi _{n}(x)\) from Eq. (C.4) for \(k=n1\).
Appendix B: The QDCT and the QDST algorithms
In our simulation the QDST1 algorithm has been used. The algorithm (shown in Fig. 2 from [32]) implements both the DCT1 and the DST1 transforms. In the first \(2^{n_{q}1}\) samples the DCT1 transform of the input state is computed, while in the last \(2^{n_{q}1}\) samples the DST1 transform is obtained (with multiplication by i factor). Hence, in our simulation the state of the last (auxiliary) qubit is set to \(\vert 1\rangle \).
Important note: in the DST1 transform from Appendix Anth sample codes \(\psi _{n+1}\) state. For the QDST algorithm nth sample stores \(\psi _{n}\) state.
Appendix C: The Schrödinger particle in the infinite potential well
Let us consider the Schrödinger particle with the mass m inside the infinite potential well with a length 2a (as showni in Fig. 1). The stationary states of the particle can be predicted in the following form:
where \(k=\sqrt{2Em}/\hbar \).
By imposing boundary conditions in the form \(\psi (a)=\psi (a)=0\) we get:
where
The normalization condition allows to calculate \(A=\frac {1}{2\sqrt{a}}\).
Finally, we obtain the stationary states in the well in the following form:
Appendix D: The Gaussian state
The localized Gaussian state of the Schrödinger particle takes the form:
where \(\langle x\rangle \) is the expected value of the position, \(\langle p\rangle \) is the expected value of the momentum, dx is standard deviation of the position while C is a normalization constant. In our work, instead of explicit value of \(\langle p\rangle \), kinetic energy of the particle \(T=\langle p\rangle ^{2}/(2m)\) is used.
Appendix E: The simulation code in Qiskit
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Ostrowski, M. Application of the QDST algorithm for the Schrödinger particle simulation in the infinite potential well. EPJ Quantum Technol. 11, 12 (2024). https://doi.org/10.1140/epjqt/s40507024002233
Received:
Accepted:
Published:
DOI: https://doi.org/10.1140/epjqt/s40507024002233