Synthesis of Quantum Logic Circuits

Vivek V. Shende,Stephen S. Bullock,Igor L. Markov

DOI: https://doi.org/10.1109/TCAD.2005.855930

2006-04-18

Abstract: We discuss efficient quantum logic circuits which perform two tasks: (i) implementing generic quantum computations and (ii) initializing quantum registers. In contrast to conventional computing, the latter task is nontrivial because the state-space of an n-qubit register is not finite and contains exponential superpositions of classical bit strings. Our proposed circuits are asymptotically optimal for respective tasks and improve published results by at least a factor of two.

Quantum Physics

What problem does this paper attempt to address?

### Problems the paper attempts to solve

This paper aims to solve the problem of efficient synthesis of quantum logic circuits, specifically including two main tasks:

1. **Achieving universal quantum computing**: That is, designing quantum logic circuits capable of performing any quantum computing task.
2. **Initializing quantum registers**: That is, preparing a specific quantum state to initiate the quantum computing process.

Unlike classical computing, the state space of a quantum register is continuous and contains exponentially many superpositions of classical bit strings. Therefore, initializing a quantum register is a nontrivial task. The paper proposes asymptotically optimal quantum circuits to accomplish these two tasks and improves on previously published results, with at least a two - fold increase in efficiency.

#### Specific problems and challenges

- **Exponentially growing state space**: Since the state space of a qubit grows exponentially with the number of bits, designing efficient quantum circuits poses a huge challenge.
- **Optimization of the number of CNOT gates**: In quantum computing, the CNOT gate is one of the most commonly used two - qubit gates, but its implementation is more complex and more error - prone than that of single - qubit gates. Therefore, reducing the number of CNOT gates is crucial for improving the overall performance of the circuit.
- **Limitations of physical implementation**: Current quantum computing technologies usually only allow interactions between nearest - neighbor qubits, which further increases the difficulty of circuit design.

### Main contributions of the paper

1. **Preparation of any n - qubit quantum state**: Proposed a circuit containing no more than \(2^{n + 1}-2^{n}\) CNOT gates for preparing any n - qubit quantum state. This result is only four times away from the theoretical lower bound.
2. **Implementation of any n - qubit operator**: Proposed a circuit containing no more than \(\frac{23}{48}\times4^{n}-\frac{3}{2}\times2^{n}+\frac{4}{3}\) CNOT gates for implementing any n - qubit operator. This result is two times better than the previous best work and is less than two times away from the theoretical lower bound.
3. **Optimization of three - qubit circuits**: In the case of three - qubit, the proposed circuit requires only 20 CNOT gates, while the previous best result required 40 CNOT gates.
4. **Adaptation to physical implementation constraints**: Even in the case where only nearest - neighbor interactions are allowed, the proposed technique does not change its asymptotic behavior.

### Methods and innovation points

- **Quantum Multiplexor**: Defined a new circuit module - the quantum multiplexor, which can generalize gates such as CNOT, Toffoli, and Fredkin, and simplifies the discussion of many quantum logic synthesis results.
- **Quantum Shannon decomposition**: Introduced the quantum version of Shannon decomposition for analyzing and designing quantum circuits.
- **Recursive decomposition**: Through the recursive decomposition method, complex quantum multiplexors are decomposed into basic gates, thereby reducing the number of CNOT gates.

In summary, this paper significantly improves the efficiency of quantum logic circuits and solves key problems in quantum computing by introducing new circuit modules and decomposition methods.