39 results about "Connected dominating set" patented technology

A method is disclosed for determining a minimum connected dominating set in a graph, the method comprising obtaining an indication of an input graph, the input graph comprising a plurality of nodes and a plurality of edges; generating a distance table comprising for each node of the input graph, an indication of a distance between the given node and each of the other node of the plurality of nodes; generating a corresponding constrained binary quadratic programming problem using the generated distance table; providing the corresponding constrained binary quadratic programming problem to a quantum annealing solver; obtaining at least one approximate solution from the quantum annealing solver; post-processing the at least one approximate solution and providing the post-processed at least one approximate solution.

The present invention provides a wireless sensor network topology self-healing algorithm based on the neighbor relationship of nodes, finds the control set DS of the network by analyzing the neighbor relationship of the nodes, connects the nodes in the DS, and obtains the connected control set CDS; through heuristic clipping rules Cut out the CDS obtained above to complete the entire construction of the minimum connected control set MCDS; query the failed nodes on the entire network, repeat the MCDS construction in the local range of the failed nodes, and finally obtain the current network minimum connected control set MCDS after partial repair '. Simulation experiments show that in most cases, the invention can quickly generate a near-optimal MCDS of the network. The invention only relies on the neighbor information of the node to quickly construct a new MCDS', which solves the problem of network topology change caused by node failure, not only restores the connectivity of the network topology, but also effectively solves the problem of prolonging the life of WSN issue of period.

The invention belongs to a route optimization technique in wireless network communication, relates to the problem of saving energy and bandwidth resource consumption of the wireless broadcast transmission in a wireless network, and provides a construction method for a distributive multicast tree with minimized network communication resource consumption based on the problem of MCDS (minimum connected dominating set) in a graph theory. Based on the established multicast tree, the MCDS approximation algorithm-based multicast route method with minimized resource consumption finds a maximum independent set MIS, and then finds all MIS nodes connected with of a dominating node set so as to obtain a dominating multicast tree. The method optimizes the resource consumption of the multicast communication in the communication network. Due to the broadcasting nature of a wireless medium, the bandwidth and energy consumption in the multicast communication mainly depends on the message relay times inthe multicast tree. Therefore, the method achieves the aim of reducing the network resource consumption through the reduction of the number of the nodes undertaking the relay mission in the multicasttree.

The invention discloses a method for optimizing and generating a backbone network of a wireless ad hoc network. Non-critical nodes, and delete the non-critical nodes from the set Z, and update the pattern set X; Step 3, do transformation and adjustment to the pattern set X; Step 4, if the set in the pattern set X after the transformation and adjustment is an empty set, then Record the set Z of the transformed and adjusted pattern set X as a connected dominating set, and go back to step 2; otherwise, the last recorded connected dominating set is the output. The invention finds a minimum connected dominating set in the undirected graph as the backbone network of the wireless ad hoc network, with less computation time and smaller target value.

The invention provides a wireless sensing network topological construction method based on mixed integer programming. The wireless sensing network topological construction method comprises the following steps of: 1) firstly, abstractly converting a problem of searching an optimal virtual backbone network of a wireless sensing network in a random network environment into a problem of a minimum connected dominating set of the wireless sensing network; 2) defining a sign stream, and determining an optimizing function using a product of a sign distribution number and node energy consumption as a target, so as to obtain a minimum generation tree for assigning all the other nodes in a current wireless sensing network; 3) constructing an assignment constraint set: ruling that each node has at least one assignment component as a neighbor node; and 4) constructing a generation tree constraint set: defining the generation tree constraint set, so that the connectivity of a final topological structure can be ensured, and constructing the generation tree constraint set by means of meeting all rules of the sign stream. The wireless sensing network topological construction method provided by the invention does not need any random pre-treatment process and has the advantages of relatively small structural scale, relatively small energy consumption and relatively long network service life.

The invention discloses an image type smoke fire disaster detection method and system. The image type smoke fire disaster detection method comprises the following steps: filtering an obtained video image; subjecting the video image to differential operation to obtain an differential image, and generating a static image according to the video image and the differential image; dividing the differential image and the static image into a plurality of connected domains according to a preset color model, and processing the connected domains to obtain a potential smoke zone; dividing the potential smoke zone into a plurality of clusters, and obtaining the area information of each cluster, the average value of the HU monuments of the connected domains, the number of the connected domains, the fluttering feature information of each cluster and the particle position information of each cluster at the same time; and judging whether smoke exists in the potential smoke zone or not by using a Bayesian decision making device according to the area information of each cluster, the average value of the HU monuments of the connected domains, the number of the connected domains, the fluttering feature information of each cluster and the particle position information of each cluster.

The present invention provides a network node election method and node equipment, which solves the problem that more redundant nodes and a large amount of redundant system message overhead will be generated when the existing distributed clustering or connected dominance set algorithm is used for election. . The method of the present invention includes: receiving the first election information sent by the target neighbor node of the network node, the probability factor of the target neighbor node is smaller than the network probability factor; according to the first election information, determining whether the network node is a node other than the preset node , the preset node includes a DS node and a node directly communicating with the DS node; when the network node is a node other than the preset node, the second election information is sent, and the second election information includes the DS node elected by the network node . In the embodiment of the present invention, the election process is executed step by step, the information generated in the election process is fully utilized and transmitted, and the number of redundant DS nodes elected and the system overhead generated by the DS election are reduced.

The invention provides a vehicle video analysis method and a system thereof. The method comprises the following steps: (1) extraction of a target area: in a V channel of HSV space, carrying out normalization on a ratio graph of a present frame and a background graph and mapping the ratio graph to a 0-255 range, obtaining an average value u and a variance delta of the ratio graph through histogramstatistics and Gaussian fitting, and taking a part except for u+ / - Kdelta in the ratio graph as the target area, wherein, K is a coefficient related to image noise; (2) removal of a shadow: dividing directions of the shadow into four directions; with regards to removal of the shadow at left and right directions, carrying out breaking and segmentation according to width of the shadow and a vehiclebody in a same connected domain, and with regards to removal of the shadow at front and back directions, comparing gradient sum of each row and a gradient row average value of the whole connected domain in a same connected domain and carrying out segmentation. According to the video analysis method, a complete body area can be extracted, obstruction of realizing subsequent video analysis functions of analyzing behaviors of pressing a line, changing a lane, running a red light and the like of a vehicle is eliminated, and misjudgement of the functions is avoided.