Abstract
We study the communication protocol known as a quantum random access code (QRAC) which encodes n classical bits into m qubits (\(m<n\)) with a probability of recovering any of the initial n bits of at least \(p>\tfrac{1}{2}\). Such a code is denoted by (n, m, p)-QRAC. If cooperation is allowed through a shared random string, we call it a QRAC with shared randomness. We prove that for any (n, m, p)-QRAC with shared randomness the parameter p is upper bounded by \( \tfrac{1}{2}+\tfrac{1}{2}\sqrt{\tfrac{2^{m-1}}{n}}\). For \(m=2\), this gives a new bound of \(p\le \tfrac{1}{2}+\tfrac{1}{\sqrt{2n}}\) confirming a conjecture by Imamichi and Raymond (AQIS’18). Our bound implies that the previously known analytical constructions of \((3,2,\tfrac{1}{2}+\tfrac{1}{\sqrt{6}})\)- , \((4,2,\tfrac{1}{2}+\tfrac{1}{2\sqrt{2}})\)- and \((6,2,\tfrac{1}{2}+\tfrac{1}{2\sqrt{3}})\)-QRACs are optimal. To obtain our bound, we investigate the geometry of quantum states in the Bloch vector representation and make use of a geometric interpretation of the fact that any two quantum states have a nonnegative overlap.
Similar content being viewed by others
Data availability
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.
References
Ambainis, A., Leung, D., Mancinska, L., Ozols, M.: ‘Quantum random access codes with shared randomness’, (2008), arXiv:0810.2937
Ambainis, A., Nayak, A., Ta-Shma, A., Vazirani, U.: ‘Dense quantum coding and a lower bound for 1-way quantum automata’. In Proceedings of the Thirty-First Annual ACM Symposium on Theory of Computing, STOC ’99, pp 376–383, New York, NY, USA, 1999. Association for Computing Machinery. https://doi.org/10.1145/301250.301347
Ambainis, A., Nayak, A., Ta-Shma, A., Vazirani, U.: Dense quantum coding and quantum finite automata. J. ACM 49(4), 496–511 (2002). https://doi.org/10.1145/581771.581773
Bengtsson, I., Weis, S., Życzkowski, K.: Geometry of the set of mixed quantum States: An apophatic approach. In: Kielanowski, P., Ali, S.T., Odzijewicz, A., Schlichenmaier, M., Voronov, T. (eds.) Geometric Methods in Physics, pp. 175–197. Springer Basel, Basel (2013)
Brassard, G.: ‘Brief history of quantum cryptography: A personal perspective’, 2006, arXiv:quant-ph/0604072
Hayashi, M., Iwama, K., Nishimura, H., Raymond, R., Yamashita, S.: ‘(4,1)-Quantum random access coding does not exist’. In 2006 IEEE International Symposium on Information Theory, pp 446–450, 2006. https://doi.org/10.1109/ISIT.2006.261708
Holevo, A.S.: Bounds for the quantity of information transmitted by a quantum communication channel. Probl. Peredachi Inf. 9(4), 177–183 (1973)
Imamichi, T., Raymond, R.: ‘Constructions of quantum random access codes’. Asian Quantum Information Symposium (AQIS), 2018. http://www.ngc.is.ritsumei.ac.jp/~ger/static/AQIS18/OnlineBooklet/122
Iwama, K., Nishimura, H., Raymond, R., Yamashita, S.: ‘Unbounded-error one-way classical and quantum communication complexity. In L. Arge, C. Cachin, T. Jurdziński, and A. Tarlecki, editors, Automata, Languages and Programming, pp 110–121, Berlin, Heidelberg, 2007. Springer Berlin Heidelberg
Kimura, G., Kossakowski, A.: ‘The Bloch-Vector Space for N-Level systems - the spherical-coordinate point of view. Open Syst Inf Dyn 12, 207–229 (2005). arXiv:quant-ph/0408014
Liabøtrø, O.: Improved classical and quantum random access codes. Phys. Rev. A 95(5), 052315 (2017). https://doi.org/10.1103/PhysRevA.95.052315
Nayak, A.: Optimal lower bounds for quantum automata and random access codes. In 40th Annual Symposium on Foundations of Computer Science (Cat. No.99CB37039), pp 369–376, 1999. 10.1109/SFFCS.1999.814608
Wiesner, S.: Conjugate coding. SIGACT News 15(1), 78–88 (1983). https://doi.org/10.1145/1008908.1008920
Acknowledgements
This paper is based on S. Storgaard’s bachelor’s thesis. L. Mančinska acknowledges support by Villum Fonden via the QMATH Centre of Excellence (Grant No. 10059) and Villum Young Investigator grant (No. 37532).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix A: Proof of observation 6
Observation 6 Let \(\beta \in \varvec{\beta }(Q_N) \backslash \{ 0\}\). Then,
where \(\lambda _{\min }\) and \(\lambda _{\max }\) is the smallest and largest eigenvalue of \(\varvec{\rho }(\beta )\), respectively. In particular, \(-\beta \in \partial \varvec{\beta }(Q_N)\) if and only \(\lambda _{\max }= r_N R_N\).
Proof
We list the eigenvalues of \(\varvec{\rho }(\beta )\) as \(\lambda _{\text {max}} \ge \cdots \ge \lambda _{\text {min}} \ge 0\). Then, the eigenvalues of \(\tfrac{1}{2} \sum _{i=1}^{N^2-1} \beta _i \varvec{\sigma }_i\) can be listed as
Let \(\beta '=\gamma \beta \) for some \(\gamma \in {\mathbb {R}}\). If \(\gamma \ge 0\) The eigenvalues of \(\varvec{\rho }(\gamma \beta )\) can be listed as
and hence \(\beta ' \in \partial \varvec{\beta }(Q_N)\) is in the boundary of Bloch space if \(\gamma =\tfrac{1}{1-N\lambda _{\min }}\). Similarly if \(\gamma <0\) then \(\beta ' \in \partial \varvec{\beta }(Q_N)\) if \(\gamma =\tfrac{1}{1-N\lambda _{\max }}\).
\(\square \)
Appendix B: Proof of lemma 9
Lemma 9 For any set of vectors, \(\{\mu _i \}_{i=1}^n\subseteq \mathbb {R}^{N}\), the inequality
holds with equality if and only if the \(\mu _i\)’s are orthogonal.
Proof
The inequality in (32) holds if all the \(\mu _i\)’s are 0, so we assume they are not all 0. We can interpret the sum on the left hand side of (32) as the inner product of
and a vector \(y_2 \in {\mathbb {R}}^{\{0,1 \}^n}\), whose entry corresponding to bit string \(x\in {\mathbb {R}}^{\{0,1 \}^n}\) is given by
Applying the Cauchy–Schwarz inequality, we get that the left hand side of (32) is upper bounded by
We claim now that
This can be proved by induction on n. First, Eq. (47) holds for \(n=1\) since
Assume that Eq. (47) holds for \(n=k\) and consider the case when \(n=k+1\). By explicitly carrying out the sum over \(x_{k+1} \in \{0,1 \}\) on the left hand side of Eq. (47) we get
Applying the parallelogram identity, i.e.,
the expression in (49) equals
Finally, applying the induction hypothesis, we complete the inductive step as follows:
Now, by inserting (47) in (46) we can conclude (32).
Recall that \(\left\langle y_1 ,y_2\right\rangle = \left| \left| y_1\right| \right| \left| \left| y_2\right| \right| \) if and only if \(y_2 =k y_1\) for some \(k\in {\mathbb {R}}\). Hence, the bound in (32) holds with equality if and only if the quantity \((y_2)_x\) in Eq. (45) is equal to some constant c independent of \(x\in \{0,1\}^n\). In other words, for all \(x \in \{0,1 \}^n\) we must have that
where the right hand side is constant. Now, fix \(m\in [n]\). The left hand side of (53) can be rewritten as
This must be invariant upon the interchange \(x_m \rightarrow \overline{x_m}\). Since \((-1)^{x_m}=-(-1)^{\overline{x_m}}\), we have that
Now fix \(m'\in [n]\backslash \{m \}\) and rewrite (55) as
This must be invariant upon the interchange \(x_{m'}\rightarrow \overline{x_{m'}}\) so we get \(\left\langle \mu _m,\mu _{m'}\right\rangle =0\). This completes the proof. \(\square \)
Appendix C: Proof of corollary 12
Corollary 12 For \(m>1\) there exists \(n_{\max }(m) \le 4^m-1\) such that any \((n\ge n_{\max }(m),m,p)\)-QRAC fulfills
Proof
The average success probability of an (n, m)-QRAC reaches the bound in Theorem 10 if and only if it is given by the following Bloch vector configuration:
-
1.
A set \(\{\nu _i\}_{i=1}^n\) of orthogonal unit vectors such that \(\pm \sqrt{r_{2^m}R_{2^m}}\nu _i \in \partial \varvec{\beta }(Q_{2^m})\). The POVM for measuring the jth bit is then associated with the pair \(\pm \sqrt{r_{2^m}R_{2^m}}\nu _j\) in the sense of (31).
-
2.
For the encodings, \(2^{n-1}\) pairs of orthogonal pure state Bloch vectors, \(\{\beta _x, \beta _{{{\overline{x}}}} \}\), where
$$\begin{aligned} \tfrac{1}{2}(\beta _x-\beta _{{{\overline{x}}}})=\tfrac{1}{\sqrt{n}}\sum _{i\in [n]} (-1)^{x_i} \nu _i=: V_x, \end{aligned}$$(57)with \(\{\nu _i\}_{i=1}^n\) given as in 1.
Let \(V:=\text {span} \big \{V_x \mid x\in \{ 0,1\}^n \big \}\). Assume, for some (n, m)-QRAC, that 1 and 2 above are fulfilled. This implies that its average success probability is \(\tfrac{1}{2}+\tfrac{1}{2}\sqrt{\tfrac{2^{m-1}}{n}}\). In view of Eqs. (17), (31) and (57), one can use the decomposition
to calculate the probability of correctly decoding the ith bit of x to be \(x_i\) as
As noted, the average of this is \(\tfrac{1}{2}+\tfrac{1}{2}\sqrt{\tfrac{2^{m-1}}{n}}\). It follows that the worst case success probability reaches \(\tfrac{1}{2}+\tfrac{1}{2}\sqrt{\tfrac{2^{m-1}}{n}}\) only if, we additionally have that for all \(x\in \{0,1 \}^n\), \(\tfrac{1}{2}(\beta _x+\beta _{{{\overline{x}}}})\in V^{\perp }\). Unless \(m=1\), we have that \(\tfrac{1}{2}(\beta _x+\beta _{{{\overline{x}}}})\ne 0\), i.e., for \(m>1\), \(V^{\perp }\) must be of non-vanishing dimension. From this, we conclude the desired. \(\square \)
Rights and permissions
About this article
Cite this article
Mančinska, L., Storgaard, S.A.L. The geometry of Bloch space in the context of quantum random access codes. Quantum Inf Process 21, 143 (2022). https://doi.org/10.1007/s11128-022-03470-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11128-022-03470-4