12 results about "NODAL" patented technology

Updating an artificial neural network is disclosed. A node characteristic is represented using a fixed point node characteristic parameter. A network characteristic is represented using a fixed point network characteristic parameter. The fixed point node characteristic parameter and the fixed point network characteristic parameter are processed to determine a fixed point intermediate parameter having a larger size than either the fixed point node characteristic parameter or the fixed point network characteristic parameter. A value associated with the fixed point intermediate parameter is truncated according to a system truncation schema. The artificial neural network is updated according to the truncated value.

Provided is a high-fidelity quantum network coding method based on controlled teleportation. The method comprises eight steps and is characterized in that based on a pre-shared entanglement pair quantum network coding protocol, two control terminals are added to control decoding of two destination nodes respectively, and perfect transmission with the fidelity of 1 is achieved; a quantum network coding model is established by using the controlled teleportation, one control terminal is added based on source nodes and the destination nodes, the fact that the source nodes and the control terminals share the GHZ state of three particles is ensured, the entanglement state is used as a quantum channel for transmitting a one-particle state, and the fact that the destination nodes will not carry out the decoding on received quantum states till the condition that measuring information of the control terminals is received is met is achieved. The high-fidelity quantum network coding method based on the controlled teleportation takes advantage of the special characteristics of teleportation of quanta, improves safety of the quantum network coding, and has wide application prospect in the technical field of communication networks.

The invention relates to an AOA-based two-dimensional wireless sensor network semi-definite programming positioning method comprising the following steps: assuming that there is no obstruction between a reference node and a target node to be positioned in a wireless sensor network, and measuring the angle of arrival of each signal; converting the angles of arrival of the signals into distance information between the nodes; converting a problem of wireless sensor network target node positioning into a mathematical optimization problem of maximum likelihood estimate MLE and solving the problem, and providing an optimized objective function for subsequent steps; introducing redundant variables to convert the mathematical optimization problem of maximum likelihood estimate MLE into a constrained optimization problem; converting the obtained constrained optimization problem into a semi-definite programming SDP convex optimization problem using a mini-max criterion and a semi-definite relaxation SDR method, and getting an optimal solution, thus completing positioning of the target node. Through the method, the precision of target node positioning is improved.

The invention provides an abnormality detection method and device based on log graph modeling. The method and the device are applied to a nonsocial network. The method particularly includes: buildinga bipartite graph according to a key field of abnormal data annotated in advance in the nonsocial network, wherein left-side nodes of the bipartite graph correspond to multiple user accounts while right-side nodes of the same correspond to parameter combinations during business interface requesting; extracting features from the bipartite graph, and splicing the extracted features to form a featurevector; performing k-means cluster processing on the basis of the abnormal data and the feature vector to acquire optimal cluster number; according to the optimal cluster number, using a Gaussian mixing model to fit dark industry feature probability distribution; when receiving incoming data, calculating dark industry probability of the data according to feature vector of the incoming data and the Gaussian mixing model, and judging whether the data are abnormal or not according to the dark industry probability and the dark industry feature probability distribution. When the data are judged abnormal, access behaviors of users can be intervened timely, so that network attacks by hackers can be avoided.

The invention discloses a method for realizing cerebrovascular image identification by using fast boundary tracking. The method comprises the following steps: obtaining an original cerebrovascular image needed to be identified and performing pretreatment on the original cerebrovascular image; performing gradient treatment on the original cerebrovascular image; using a pixel point with the biggest gradient value in the image after the gradient treatment as a starting pixel point to perform tracing judgment on boundary points; and repainting all the obtained boundary points into an image which is the needed processed cerebrovascular image. The method is characterized in that three adjacent points of the prior point which is far from the last point are selected as candidate nodes, and the next adjacent point is determined by distributing different weights according to the importance differences of position information of the nodes so as to obtain a boundary tracing image of the cerebrovascular image. The boundary image extracted by the method has distinct outline, low noise, high precision, and smooth boundary, improves the quality of boundary extraction, and reduces the computation time greatly; besides, the method is applied to a system for computer-aided diagnosis of cerebrovascular diseases, and improves the accuracy of a computer in judging cerebrovascular pathological change parts and types.

The invention relates to a voltage stability assessment method based on thevenin equivalence and branch transmission power limits. The method comprises the steps of obtaining relative power margins of all load nodes of an electric system, and selecting the minimum value of the relative power margins as a voltage stability index of the electric system, wherein the larger the voltage stability index is, the higher the voltage stability of the electric system is; according to the step of obtaining the relative power margins, 1, thevenin equivalence parameters of load nodes to be assessed are obtained; 2, a two-node system containing the load nodes to be assessed is constructed by means of the thevenin equivalence parameters, a power balance equation is analyzed, and a quiescent voltage stability discriminant is obtained; 3, branch transmission power limits of the two-node system are obtained through calculation of the quiescent voltage stability discriminant, and the relative power margins of the load nodes to be assessed are obtained. Compared with the prior art, no local measurement data is needed, calculation is rapid and convenient, assessment indexes containing the limit transmission power are given, and online voltage stability assessment of a power grid is achieved.

The invention relates to a method for determining a mapping of the contacts and/or distances between the teeth of the maxillary arch and the mandibular arch of a patient, characterised in that it comprises the following steps:

• • obtaining the mandibular kinematics recorded on the patient;
  • obtaining surface meshes of the maxillary arch and the mandibular arch and registering said meshes relative to one another;
  • creating a reduced mesh of at least one of said arches, comprising the selection, in the mesh of said arch, of the cells in which the nodes are located at a distance of less than 1 cm from the mesh of the opposite arch;
  • creating, for each cell of said reduced mesh, a bounding box comprising a plurality of voxels surrounding said cell;
  • using the mandibular kinematics, calculating a network of contacts comprising, for each voxel of the bounding box, information on the existence of a contact between said voxel and a node of the mesh of the opposite arch during a relative movement of the mandibular arch in relation to the maxillary arch.

The invention discloses a method for building a patent document model tree. The contents of components of a patent document, multi-generation quotations and IPC (International Patent Classification) are combined to build the patent document model tree. The method comprises the following steps: the whole patent document is used as a root node; an abstract, a specification, claims, the IPC and the quotations are used as intermediate nodes; independent claims, the section, big group, small group and big class of the IPC and generation one to m-1 of quotations in the quotations are used as intermediate knots; and the name of the invention, the technical scheme and the application field of the abstract, the subordinate right claim of the claims, the prior art, the to-be-solved problem, the technical scheme, the beneficial effects and the embodiment of the specification, small classes in the IPC and the generation m0 of quotation of the quotations are used as leaf knots. m is the mth generation of quotation of the patent document, and m is greater than or equal to 1 and is less than or equal to m0. The patent document is more comprehensively, deeply and accurately expressed according to the method, and the method can be applied to similar or related calculation of the patent document.