The current quantum search algorithms do not consider the difference of improtance among the targets.To solve this problem,this paper proposed a quantum search algorithm based on distributed weight target.It analyzed the iteration results by changing the amplitude of initial states in superposition state,and demonstrated the conditions in order to keep the quantum searching algorithm still effective with endowing the weight coefficient to target states.Based on this condition,it constructed a new quantum superposition state containing the weight information of target states,and the new algorithm kept the characters of original Grover algorithm as the same time.The simulation results show that this algorithm can search each target state successfully with the probability of endowed weight coefficient.
The current quantum search algorithms do not consider the difference of improtance among the targets.To solve this problem,this paper proposed a quantum search algorithm based on distributed weight target.It analyzed the iteration results by changing the amplitude of initial states in superposition state,and demonstrated the conditions in order to keep the quantum searching algorithm still effective with endowing the weight coefficient to target states.Based on this condition,it constructed a new quantum superposition state containing the weight information of target states,and the new algorithm kept the characters of original Grover algorithm as the same time.The simulation results show that this algorithm can search each target state successfully with the probability of endowed weight coefficient.
The current Grover quantum searching algorithm cannot identify the difference in importance of the search targets when it is applied to an unsorted quantum database, and the probability for each search target is equal. To solve this problem, a Grover searching algorithm based on weighted targets is proposed. First, each target is endowed a weight coefficient according to its importance. Applying these different weight coefficients, the targets are represented as quantum superposition states. Second, the novel Grover searching algorithm based on the quantum superposition of the weighted targets is constructed. Using this algorithm, the probability of getting each target can be approximated to the corresponding weight coefficient, which shows the flexibility of this algorithm. Finally, the validity of the algorithm is proved by a simple searching example.
As searching targets in an unordered database with Grover's algorithm,difference in marked items is not taken into consideration.If fraction of marked items is greater than 1/4,success probability of search decreases rapidly with increase of marked items.When the fraction of marked items is greater than 1/2,the algorithm is disabled.Aiming at above problems,an improved Grover's algorithm with weighted targets is proposed in which every target is assigned a weight coefficient according to its significance.With these weight coefficients,targets are represented as quantum superpositions which make probability getting target equal to its weight coefficient.An adaptive phase matching method is proposed based on weighted targets.The directions of phase rotations are contrary,and amplitudes of the two phase rotations are determined by inner-product of target quantum superposition and initial state of the system.As the inner-product is greater than(((3-5)/8)~(1/2),) success probability is equal to 1 with two steps of Grover iteration at most.The improved quantum searching algorithm and the new phase matching are verified by an example.
The current Grover quantum searching algorithm can not identify the difference in importance of the marked items when it is applied to an unsorted quantum database, and the probability for each marked item is equal. To solve the problem, a Grover searching algorithm based on the weighted marked items is proposed. Firstly, each marked item is endowed a weight coefficient according to its importance. Applying these different weight coefficients, the marked items are represented as a quantum superposition state. Secondly, the novel Grover searching algorithm based on the quantum superposition of the weighting marked items is constructed. Using this algorithm the probability of getting each marked item can be approximated to the corresponding weight coefficient, which the flexibility of this algorithm is shown. Finally, the validity of the algorithm is validated by a simple searching example.
Grover search algorithm is applied to the current number of objectives in quantum unsorted database, the meaning of each goal difference is not taken into account. When the target is more than a quarter of the total items, the probability of the target falls quickly identify and target increases, and when the target is more than half of the total items, the algorithm will fail. To solve these problems, first of all, based on a weighted index improvement is proposed in which each target in accordance with the right given to the importance of the weight coefficient. These different weight coefficients, we give all the target state, which may make the probability of each target equal weights of the quantum superposition. Secondly, an improved phase matching is proposed the phase rotation in both directions is the same, and the inner product of two states by the target amplitude superposition and initial state of the system is determined. When this inner product is 1/2 or more, when using the improved phase matching, the target may be 100% probability and by only one Grover iteration. Finally, the effectiveness of these measures by a simple search instance validation.
By quantum Grover search algorithm,people can find a target item in a database faster than any classical algorithm.In this search algorithm the initial state is the equally weighted superposition of all basis states,which is operated by the unitary operator repeatedly.Consequently,non-target states transfer their amplitude to the target state,so we can find the target item with larger probability.However,sometimes we just want to know partial information,so we only need to find the subdatabase(a block) containing the target item,which is partial search.A partial search algorithm was suggested by Grover and Radhakrishnan,which is denoted as Grover-Radhakrishnan Algorithm of Partial Search(GRK),in which they considered the database contains one target item.In our paper,we study the database which contains several target items,and they are randomly distributed in two target blocks.Our algorithm is based on GRK.We minimize the number of queries to the oracle for blocks of large size,and compare this case with the one that each target block contains the same number of target items.
As a new quantum computing model, quantum walk has been widely studied in recent years. It consists of discrete time quantum walk and continuous time quantum walk. Discrete time quantum walk includes coin quantum walk and scattering quantum walk. Meanwhile as a quantum search algorithm, Grover algorithm can search an unsorted database in time complexity of O(√N) . Recent years quantum walk has been used to solve search problems and some of them have been proved to be as efficient as Grover algorithm. Making full use of the novel properties of quantum walk, the quantum walk search algorithms on the 2D grid and hypercube have been proposed. Inspired by phase matching condition of Grover algorithm, we propose a new quantum walk search algorithm which is based on coin quantum walk. Firstly we give the graph to the quantum walk, and then describe the algorithm in detail. Algorithm uses different coin operators and shift operators for two different cases and draws the corresponding iteration operators. Then we prove that iteration operators used in the algorithm are unitary operators. Then we analyze the time complexity and probability of success of the algorithm. Analysis indicates that the time complexity of the algorithm is the same as that of Grover algorithm, however when the targets to be searched are more than 1/3 of the total targets, the algorithm probability of success is greater than that of Grover algorithm. Finally we give the circuit implementation of the algorithm.