Synthesis And Optimization Of Quantum Circuits Based On Reversible Logic

Reversible logic has good application prospects in emerging fields such as quantum computing,optical computing,and low-power circuits.Many quantum algorithms need to calculate some classical logic functions.Directly using reversible logic circuits to implement logic functions often consumes a lot of resources.Therefore,the synthesis and optimization of reversible logic circuits have become a research hotspot in recent years.The research content of this paper is mainly divided into the following aspects:1.Oracle route optimization based on minimum weight and template matchingFor the Oracle circuit constructed by the set of Boolean functions f(x),an optimization algorithm based on minimum weight and template matching is proposed.The whole process can be divided into two stages.The first stage reorders the MCT gates of the same controlled point of the Oracle line based on the minimum weight matching algorithm to minimize the number of gates of the generated line;the second stage uses the template matching method,Further reduce the gate count and cost of the line.Experiments show that,compared with the optimization tool RCViewer+,the number of Oracle circuit gates under the Deutsch-Jozsa(DJ)algorithm is reduced by about 48.3%and the cost is reduced by about 64.5%under the 4-10 qubits.The Oracle gate count decreased by about 25.0%,and the cost decreased by about 18.2%.2.ESOP circuit optimization based on shared nodesAiming at the problem that the number of shared nodes in the ESOP line is equal,which leads to the randomness of node selection and the inability to maximize the cost reduction,an ESOP optimization based on shared nodes is proposed.In the first stage,the maximum cost is used as the measure,and the shared nodes are searched by dynamic programming,and the best match can be found.The second stage utilizes the conversion templates of various gate pair combinations,and adopts a greedy strategy to optimize the shared line,which can further optimize the line.Compared with the shared node method,the two-stage quantum cost is reduced by about 6.8%and 18.4%on average;compared with the template matching method,the total line cost is reduced by about 40.58%.3.The realization of the 0-1 knapsack problem circuit based on Grover quantum algorithmCombined with Grover quantum search algorithm,the DKP problem is realized by circuit.The Oracle circuit of the knapsack problem is constructed by the comprehensive algorithm of the quantum logic circuit,and the NCV gate library is introduced at the same time,which reduces the total cost of the circuit and simplifies the circuit structure.The experimental results show that after optimization,the line cost is reduced by about 30%,and the more qubits,the more obvious the optimization effect.