M. Jezek,J. Rehacek,J. Fiurasek
Abstract:We propose a numerical algorithm for finding optimal measurements for quantum-state discrimination. The theory of the semidefinite programming provides a simple check of the optimality of the numerically obtained results.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to develop a general numerical algorithm for finding the optimal quantum - state discrimination measurement method. Specifically, the author proposes an iterative algorithm to find the best strategy for distinguishing quantum states with the minimum error rate.
### Main problems and background
1. **Basic characteristics of non - orthogonal quantum states**:
- In quantum mechanics, the existence of non - orthogonal quantum states is a fundamental feature. This means that two non - orthogonal quantum states cannot be perfectly distinguished.
- This characteristic sets limits on information processing at the microscopic level because it determines the amount of information that can be encoded into a quantum system.
2. **Two models of quantum - state discrimination**:
- **Minimizing the Bayesian cost function**: This is a form of the generalized error rate and is applicable to the case of linearly independent pure states.
- **Unambiguous discrimination**: This scheme combines error - free discrimination and a certain proportion of uncertain results.
3. **Existing research progress**:
- The optimal measurements for pairwise, triple, quadruple states and linearly independent symmetric states are well understood, and many theoretically optimal devices have been experimentally realized, mainly using polarized light.
### Core problems of the paper
- **Objective**: Develop a general method to optimize the ambiguous discrimination between arbitrary quantum states.
- **Specific problem description**: Suppose Alice sets up M different quantum system sources, each described by its density matrix. Alice randomly selects a source and sends the generated quantum system to Bob. Bob's task is to determine which source generated this quantum system with as few errors as possible.
### Mathematical description
- **Objective function**: Maximize Bob's success probability \(P_s\), defined as:
\[
P_s=\sum_{j = 1}^{M}\xi_j\text{Tr}(\rho_j\Pi_j)
\]
where \(\xi_j\) is the prior probability that the \(j\) - th source is selected, \(\rho_j\) is the state of the \(j\) - th source, and \(\Pi_j\) is Bob's measurement operator (POVM element).
- **Constraints**:
\[
\Pi_j\geq0,\quad j = 1,\ldots,M,\quad\sum_{j = 1}^{M}\Pi_j = 1
\]
### Solution
- **Iterative algorithm**: The author proposes a simple iterative algorithm based on the variational method, incorporating the constraint conditions into the model by introducing the Lagrange multiplier method. The resulting iterative equations are:
\[
\Pi_j=\xi_j^2\lambda^{- 1}\rho_j\Pi_j\rho_j\lambda^{-1},\quad j = 1,\ldots,M
\]
and
\[
\lambda=\left(\sum_{j = 1}^{M}\xi_j\rho_j\Pi_j\rho_j\right)^{1/2}
\]
- **Optimality check**: A simple optimality check method is provided using semi - definite programming (SDP) theory to ensure the correctness of the numerical results.
### Application example
- **Three asymmetric coplanar qubit states**: The effectiveness of this algorithm is demonstrated through numerical calculations, especially the change in the average error rate under different prior information \(\xi\) when the angle \(\phi=\pi/16\). The results show that as the prior information changes, the optimal measurement strategy will change from a two - output channel to a three - output channel and then back to a two - output channel.
### Conclusion
This paper successfully develops a general iterative algorithm for finding the optimal quantum - state discrimination measurement method. By combining with semi - definite programming theory, the accuracy and optimality of the numerical solution are ensured. This method is not only of great theoretical significance but also shows its practical value in practical applications.