34 results about "Membrane computing" patented technology

The invention discloses a multi-objective optimization method for radar radiation source signal sorting based on a membrane system. First, the traditional clustering is transformed into a multi-objective optimization problem, and then the membrane calculation optimization theory is used as the main line, and the genetic algorithm is integrated into the membrane structure framework. Algorithms, computational intelligence and other intelligent information processing technologies are intersected, and the self-improvement ability of the algorithm and the calculation efficiency and fusion ability of the current typical intelligent algorithms are compared to verify the convergence accuracy and effectiveness of the membrane system, in order to obtain multiple data sets with different clusters. The Pareto frontier solution maintains the diversity and difference of sorting results, which is conducive to the accuracy of later sorting. Combined with the nature of radiation source signal sorting, the clustering validity index, feature evaluation index and other evaluation criteria related to sorting are constructed into corresponding problem models and multi-objective functions. For the intercepted unknown radar signals, the essence of unsupervised classification is in line with the classification requirements of radar emitter signals.

The invention relates to the field of network search, and discloses a search method based on membrane computing. The method comprises the following steps of (A) obtaining the optimum weighing factor; (B) obtaining the similarity of document themes; (3) predicting and sequencing the optimum values. The method has the beneficial effect that the accuracy of predicting the similarity of the unvisited URL (uniform resource locator) themes is improved, and the recall ratio and accuracy of theme crawlers are further improved, so the theme crawlers are instructed to collect more webpage sets with higher quality from an internet, and the webpage sets which are more interested by a user can be more effectively collected.

The invention discloses an improved DND algorithm and its realization method based on FPGA. The improved DND algorithm includes a forward phase and a reverse phase. The FPGA-based implementation method of the improved DND algorithm includes steps: using FPGA to express the cell-type membrane system; using hardware description language, according to the improved DND algorithm design rule multi-set non-deterministic selection and parallel execution of the FPGA circuit; using the FPGA circuit to execute the cell Evolution of Type Membrane Systems. The improved DND algorithm is based on the principle of maximal consumption of objects, and directly generates and executes a multiset of rules in each region. The improved DND algorithm is based on the implementation method of FPGA, which realizes the representation of basic features such as the rules, objects and membrane structure of the membrane system in FPGA, and establishes the relationship between the structural and functional characteristics of FPGA and the maximum parallelism and non-deterministic characteristics of membrane calculation. Based on the mapping relationship, a design method of parallel processing circuit architecture that can better realize the function of membrane computing is proposed, which lays the foundation for the application of membrane computing in many engineering fields.

The invention relates to a mobile robot path planning method for fusion membrane calculation and RRT. Based on the original RRT algorithm, the method uses membrane calculation to optimize it, sets the membrane calculation structure, expands the random tree in each basic membrane, and increases the number of sampling points when the random tree expands in each basic membrane; The communication rules between membranes are used to update the corresponding relationship between the sampling points output by each basic membrane and its parent node, and the updated results are returned to the corresponding basic membrane to continue to expand the random tree. The random tree in the whole membrane system continues to expand iteratively until a random tree in a basic membrane reaches the end point and stops expanding. By increasing the number of sampling points and setting the exchange rules between the basic membranes, the method makes the expansion of the random tree in the membrane more directional, thereby increasing the real-time and rationality of the mobile robot's path planning.