Kuantum devre sentezi

This thesis presents a new approach for the synthesis of quantum circuits. Quantum computers, more specifically quantum algorithims, take on the eyes with their computational promises. They paved the way for calculation of the complex problems that can't be solved in polynomial time by traditional counterparts. Exploiting quantum mechanical phenomena and its features is the main power source of the quantum computing idea. This is also leaning on the concept that accepts information as a physical item. In addition, quantum mechanical unitarity brings reversibility for quantum algorithms and quantum circuits. This creates ideal application area for reversible computing and reversible circuit design. Reversible computing is motivating scientist and researchers for years to achieve energy efficient computation. With the help of advancing technology, quantum computation become applicable. At this point, reversible quantum circuit design is coming forward that is the core of this computation. When compared with classical computation methods, quantum systems are very sensitive. This is one of the main reasons to synthesize these circuits minimally as possible. Depending on quantum computation method, each gate in quantum circuits corresponding to one or more pulse operations. Therefore, optimized circuits will improve both the security and the run time of the computation. Still, optimal circuit synthesis for higher bit count is hard to achieve, it can take very long time which is not practical at all. To compete with classical computation in a realistic manner, this is one of the main obstacles to overcome and it constitutes the main motivation why we aim at a fast synthesis algorithm in this thesis. To fulfill the needs of this upcoming technology, we perform synthesis and optimization of quantum circuits in two main parts. In the first part, we propose a fast synthesis algorithm that implements any given reversible Boolean function with quantum gates. Instead of an exhaustive search on every given function, our algorithm creates a library of essential functions and performs sorting to obtain desired function. In the second part, we optimize our circuits by using templates. The proposed templates mainly consist of identical neighbor gates and Toffoli gates, realized with V, V† and CNOT gates. We also improve the CNT library by taking negative control lines into account which provides important circuit cost reduction. We call this new library as PN-CNT that stands for "Positively and Negatively Controlled CNOT, Not, Toffoli". Quantum computers implemented in various ways so far. Each implementation has its own physical cost. For the calculation of quantum circuit cost, we preferred to use widely accepted NCV-111 cost metric in this thesis. Finally we compared our results with studies in the literature.