Abstract
We propose an intermediate walk continuously connecting an open quantum random walk and a quantum walk with parameters \(M\in {{\mathbb {N}}}\) controlling a decoherence effect; if \(M=1\), the walk coincides with an open quantum random walk, while \(M=\infty \), the walk coincides with a quantum walk. We define a measure which recovers usual probability measures on \({{\mathbb {Z}}}\) for \(M=\infty \) and \(M=1\) and we observe intermediate behavior through numerical simulations for varied positive values M. In the case for \(M=2\), we analytically show that a typical behavior of quantum walks appears even in a small gap of the parameter from the open quantum random walk. More precisely, we observe both the ballistically moving towards left and right sides and localization of this walker simultaneously. The analysis is based on Kato’s perturbation theory for linear operator. We further analyze this limit theorem in more detail and show that the above three modes are described by Gaussian distributions.
Similar content being viewed by others
References
Ambainis, A.: Quantum walks and their algorithmic applications. Int. J. Quant. Inf. 1(2003), 507–518 (2003)
Attal, S., Guillotin-Plantard, N., Sabot, C.: Central limit theorems for open quantum random walks and quantum measurement records. Ann. Henri Poincaŕe 16, 15–43 (2015)
Attal, S., Petruccione, F., Sabot, C., Sinayskiy, I.: Open Quantum Random Walks. J. Stat. Phys. 147, 832–852 (2012)
Brun, T.A., Brun, Carteret, H.A., Ambainis, A.: Quantum random walks with decoherent coins. Phys. Rev. A 67, 032304 (2003)
Dhahri, A., Mukhamedov, F.: Open quantum random walks and quantum Markov chains. Funct. Anal. Appl. 53, 137–142 (2019)
Higuchi, Yu., Segawa, E.: Dynamical system induced by quantum walks. J. Phys. A 52, 39 (2019)
Kato, T.: A Short Introduction to Perturbation Theory for Linear Operators. Springer, New York (1982)
Kendon, V.: Decoherence in quantum walks—a review. Math. Struct. Comp. Sci. 17, 1169–1220 (2007)
Konno, N., Komatsu, T., Sato, I.: Walk/Zeta Correspondence. J. Stat. Phys. 190, 36 (2023)
Konno, N.: Limit theorems and absorption problems for one-dimensional correlated random walks. Stoch. Models 25, 28–49 (2009)
Konno, N.: Quantum walks. In: Lecture Notes in Mathematics, Vol. 1954, pp. 309–452 Springer, Heidelberg (2008)
Ko, C.K., Konno, N., Segawa, E., Yoo, H.: Limit theorems for open quantum random walks. J. Stat. Phys. 176, 710–735 (2019)
Portugal, R.: Quantum Walk and Search Algorithms, 2nd edn. Springer, Cham (2018)
Segawa, E., Konno, N.: Limit theorems for quantum walks driven by many coins. Int. J. Quant. Inf. 6, 1231–1243 (2008)
Sunada, T., Tate, T.: Asymptotic behavior of quantum walks on the line. J., Funct. Anal. 262, 2608–2645 (2012)
Suzuki, A.: Asymptotic velocity of a position-dependent quantum walk. Quant. Inf. Process. 15, 103–119 (2016)
Whitfield, J.D., Rodríguez-Rosario, C.A., Aspuru-Guzik, A.: Quantum stochastic walks: a generalization of classical random walks and quantum walks. Phys. Rev. A 81, 022323 (2010)
Acknowledgements
The authors thank to Jir̆í Marys̆ka, Stanisla Skoupý for fruitful discussion. The authors also thank to the referee for carefully reading and giving constructive suggestions to this paper. K.M. was supported by World Premier International Research Center Initiative (WPI), Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan and the grant-in-aid for young scientists No. JP17K14235, Japan Society for the Promotion of Science, Grant-in-Aid of Scientific Research (B) No. JP21H01001. E.S. acknowledges financial supports from the Grant-in-Aid of Scientific Research (C) No. JP19K03616, Japan Society for the Promotion of Science and Research Origin for Dressed Photon.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Hal Tasaki.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Propositions A.1–A.3
Propositions A.1–A.3
The matrix representation for \({\hat{W}}_{s,t}(0)=:{\hat{W}}_{s,t}\) is explicitly written as
for \(s=-1\) and \(t=0\). Using the formula for \(\lambda _1(k)\) and \(\lambda _2(k)\), we know that eigenvalues of \({\hat{W}}_{s,t}\) are
which shows that the matrix \({\hat{W}}_{s,t}\) possesses eigenvalues with multiplicity greater than 1. By the Cayley-Hamilton theorem, we know that \(F_{{\hat{W}}_{s,t}}({\hat{W}}_{s,t})= O\), where
Interestingly, we observe the following result.
Lemma A.1
The minimal polynomial \(f_{{\hat{W}}_{s,t}}(\lambda )\) of the matrix \({\hat{W}}_{s,t}\) is
Proof
Direct calculations yield that
\(\square \)
A direct consequence of this lemma is the following.
Proposition A.1
All eigenvalues of \({\hat{W}}_{s,t}\) are semisimple. In particular, the matrix \({\hat{W}}_{s,t}\) is diagonalizable.
Thanks to Proposition A.1, the eigenvalue \(\lambda = 1\) of \({\hat{W}}_{s,t}\) is semisimple and hence the associating eigenspace, which is a three-dimensional linear space, is generated by three linearly independent eigenvectors of \(\lambda = 1\). We have the following observation for eigenstructures at \(\lambda =1\) by a direct computation of the non-perturbed matrix \({\hat{W}}_{s,t}\).
Lemma A.2
The eigenspace of \({\hat{W}}_{s,t}\) associated with \(\lambda = 1\) is generated by the following vectors:
One of their orthonormal choice is
Next consider the perturbation of these eigenstructures. Namely, consider eigenpairs of \({\hat{W}}_{s,t}(\delta )\) with sufficiently small \(\delta \). Thus to apply the reduction process [7], the next task is to obtain an explicit expression for the matrix \({\tilde{T}}^{(1)}\) because its eigenvalue \(\lambda _j^{(1)}\) gives the coefficient of the second order of the eigenvalue of \({\hat{W}}_{s,t}(\delta )\). To this end, we need to prepare an explicit expression for the eigenprojection \(\Pi \) of the eigenvalue 1 of \({\hat{W}}_{s,t}(0)\). Our strategy here is to compute eigenvalue \(\lambda _j^{(1)}\) of \({\tilde{T}}^{(1)}\). Since \(\Pi \) is an eigenprojection, it holds that \(\Pi ^2=\Pi \) and \(\Pi {\hat{W}}_{s,t}={\hat{W}}_{s,t}\Pi \). In addition, it can be shown that the property \(\Pi =\Pi ^*\) holds by [6] which treats more general setting of the truncated unitary matrix.
Proposition A.2
Let the centered generalized eigenspace of \({\hat{W}}_{s,t}\) be defined by
Then we have the following properties of \({{\mathcal {H}}}_c\).
-
1.
The centered generalized eigenspace of \({\hat{W}}_{s,t}:={\hat{W}}_{s,t}(0)\) is expressed by
$$\begin{aligned} {{\mathcal {H}}}_c&= \bigoplus _{|\lambda |=1}\ker (\lambda -{\hat{W}}_{s,t}) \\&= \textrm{span}\{ \chi _{s,t}^*\psi \;|\; \textrm{supp}(\psi )\subset \{s,\dots ,t\},\;\psi \mathrm {\;is\;an\;eigenvector\;of\;} {\hat{W}}_{-\infty ,\infty }(0)={\hat{U}}(0) \} \end{aligned}$$ -
2.
The complementary invariant subspace of \({{\mathcal {H}}}_c\) can be expressed by \({{\mathcal {H}}}_c^\perp \).
Proof
This is a direct consequence of Lemma 3.3, (3.12) and Lemma 3.4 in [6]. \(\square \)
By the first part of Proposition A.2 implies that \({{\mathcal {H}}}_c=\ker (1-{\hat{W}}_{s,t})\). Moreover the second part of Proposition A.2 implies that the eigenprojection \(\Pi \) of the eigenvalue 1 can be computed by using Lemma A.2 as follows.
which is an orthogonal projection, in which case \(\Pi ^2 = \Pi \) and \(\Pi ^*= \Pi \) hold.
On the other hand,
and hence
Direct calculations yield the following result.
Proposition A.3
The matrix \(\Pi T^{(1)}\Pi \) is a skew-Hermitian matrix. Eigenvalues of the matrix \(\Pi T^{(1)}\Pi \) on the eigenspace \(\textrm{R}(\Pi )\) are 0, \(i / \sqrt{3}\) and \(-i/\sqrt{3}\). In particular, all these eigenvalues are simple. The corresponding eigenprojections are \(\Pi ^{(1)}_1:=v_1v_1^{^*}\), \(\Pi ^{(1)}_2:=v_2v_2^*\) and \(\Pi ^{(1)}_3:=v_3v_3^*\), respectively. Here
Remark A.1
-
1.
\(\Pi =\Pi _1^{(1)}+\Pi _2^{(1)}+\Pi _3^{(1)}\) and \(\Pi _i^{(1)}\Pi _j^{(1)}=\delta _{ij}\Pi _i^{(1)}\) holds.
-
2.
The eigenvalue on \(\textrm{R}(1-\Pi )\) is also 0. The eigenvector of eigenvalue \(\epsilon \) for the matrix \(\Pi (T^{(1)}-\epsilon I)\Pi \) corresponds to that of \(\textrm{R}(\Pi )\). Taking \(\epsilon \rightarrow 0\) to this eigenvector, we obtain the eigenvector \(v_1\). See [7] for more detail.
-
3.
The coefficients of the second orders of the eigenvalues of \(\lambda _1(\delta )\) for \(j=0\) in (27) and \(\lambda _2(\delta )\) for \(j=1,2\) in (30) identify with the eingenvalues of \(\Pi T^{(1)}\Pi \) in \(R(\Pi )\). This derives from nothing but the contribution of the reduction process of the perturbation theory [7]. Moreover, it is possible to obtain the coefficients of more than the third orders by repeating the reduction process. For example, the coefficient of \(\lambda _2(\delta )\) for \(j=1\) can be computed by the same procedure by replacing \({\tilde{T}}^{(1)}\) with \({\tilde{T}}^{(2)}\), where
$$\begin{aligned} {\tilde{T}}^{(2)}:=\Pi T^{(2)}\Pi -\Pi T^{(1)}\Pi T^{(1)} S- \Pi T^{(1)}S T^{(1)} \Pi -ST^{(1)}\Pi T^{(1)}\Pi . \end{aligned}$$Here S is the reduced resolvent of \({\hat{W}}_{s,t}\) at the eigenvalue 1.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Konno, N., Matsue, K. & Segawa, E. A Crossover Between Open Quantum Random Walks to Quantum Walks. J Stat Phys 190, 202 (2023). https://doi.org/10.1007/s10955-023-03211-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10955-023-03211-6