47 results about "Path reconstruction" patented technology

According to the big data processing method based on interactive cloud computing and the artificial intelligence server, the first data interaction track between different data service processing endscan be generated by obtaining the one-way service data and the first service demand report corresponding to the one-way service data, so that the first path distribution and the first associated dataset are determined. Therefore, service association can be carried out on the plurality of data service processing ends, and data path reconstruction is carried out to achieve restoration of a data service scene. According to the invention, by detecting the target service data, the security during service data integration can be ensured based on the consensus cloud computing key; and the unidirectional service data can be comprehensively considered to avoid the incompleteness of a data service scene, and the data security between the target service data and the unidirectional service data canbe ensured through the consensus cloud computing key. By means of the scheme, the scene missing problem and the data security problem existing when the data service scene is restored are solved.

The invention discloses a self-adaptive delay-sensing energy-saving method applied to the optical wireless mixed broadband access network. In the prior art, dormancy of an optical network unit can effectively improve the energy efficiency of the optical wireless mixed broadband access network, however, the energy efficiency of a continuous dormant gateway is not high, and delay of a system data packet may be aggravated. Thus, in the downlink direction of the optical wireless mixed broadband access network, the optical network unit can select the growth speed of the dormancy period adaptively according to the arrival rate of the data packet, thereby fully utilizing the delay tolerance of the data packet; and in the uplink direction, a wireless net-shaped network of multiple gateway balance loads can reduce delay and congestion caused of path reconstruction caused by dormancy of optical network unit, thereby reducing delay in the network. The method can enable the energy efficiency of the network to be maximal while satisfying delay of the data packet.

The invention discloses a data packet path reconstruction method in a software defined network. In a network initialization stage, a controller allocates an ID to each switch/router according to a network size. When a sending end generates a data packet, related data flow information is sent to the controller, and the controller decides a specifically used path reconstruction method and field allocation of a newly added data packet header, and feeds information back to the sending end. When the switches/routers forward the data packet, corresponding operation is executed according to the newlyadded packet header, and path-related information is recorded in the packet. Finally, an administration server in the network reconstructs a transmission path of the data packet according to the path-related information in the received data packet. The method provided by the invention has no special requirements for the category of the software defined network topology, has the characteristic oflow calculation overheads of the switches/routers, can still precisely reconstruct the transmission path of the data packet under the conditions of network failure and congestion, and greatly facilitates administration and failure checking of the software defined network.

The invention relates to the technical field of network communication, in particular to a fast alternative route distribution method based on overlay network environment. Aiming at the problem that the traditional overlay network alternative route distribution method has low recovery time efficiency, the method disclosed in the invention introduces the delay constraint during route reconstructionas the selection condition of alternative routes, which is beneficial to the selection of an alternative route with fast recovery speed; the process of selecting a unique alternative route is converted into the selection of a set of alternative routes by expanding a failure area, and the success probability of selecting a high quality alternative route can be increased. The fast alternative routedistribution method based on overlay network environment can be used for reducing the delay time of failure route recovery and increase the success rate of route recovery.

The invention belongs to the field of traffic planning and management and specifically relates to a vehicle travel path reconstruction macro/micro integrated new method based on automatic vehicle identification (AVI) data. The method constructs a macro/micro integrated reconstructed frame only by utilizing the automatic vehicle identification data and is characterized by, in micro level, reconstructing vehicle paths based on individual path selection principle and with particle filter theory being integrated, and increasing a reconstruction path traffic and path set constraint for a path traffic estimator in macro level; and in the macro level, realizing path traffic estimation based on random user equalization principle and with the path traffic estimator being integrated, and updating travel time consistency of particle filter and a path gravitation observation equation. Through the method, high-precision and complete vehicle travel paths and path traffic and road section traffic can be obtained under the conditions that only the AVI data is available and the AVI coverage rate is low.

The invention provides an HR-WPAN Mesh (high-rate wireless personal area networks mesh) network efficient route repair method based on self-adaptation two-way path rebuilding. Through adopting four new mechanisms of self-adaptation two-way path re-building, Route Error message simplification, Route Formation message simplification and inter-layer auditing, the method solves the four problems that the existing HR-WPAN Mesh network route repair method based on a topology server adopts a reverse path rebuilding mode, so that the redundancy control expenditure and data grouping waiting delay are caused, when the end points of two damaged paths are mutually used as source and target nodes, the path rebuilding is repeatedly carried out, and when the Route Error message and the Route Formation message have redundancy fields and link failure, data groups cached by the upstream node can be easily lost due to route information limitation. The method has the advantages that the speed of putting the repaired route into use is accelerated, the data grouping transmission success rate is improved, and the control expenditure and the data grouping average time delay are reduced. The fast and efficient route repair method provided by the invention comprises three steps including fault reporting, path building notification and path rebuilding during the operation.

A system for plant leaf identification includes: a plant image capturing unit which captures an image of a target plant to generate a plant image; a plant area image extraction unit which separates a background area and a plant area in the plant image to generate a plant area image including the plant area; a plant area image skeletonization unit which skeletonizes the plant area image to generate a skeletonized plant area image; a candidate leaf path generation unit which identifies a root vertex, a junction vertex and a leaf tip vertex in the skeletonized plant area image, and generates a plurality of candidate leaf paths by calculating all possible paths from the root vertex to the leaf tip vertex; and a final leaf path reconstruction unit which reconstructs a final leaf path matching the plant image by selecting the plurality of candidate leaf paths.

The invention relates to a method for reconstructing a roundabout path during interruption of wide-area protection communication under a heavy load. The method takes a wide-area protection communication network under the heavy load as an object, and constructs an appropriate shunt roundabout communication path quickly based on the left undamaged network when a wide-area protection communication optical fiber is interrupted so as to ensure that a wide-area protection communication service is carried out normally. A single light-load channel roundabout path reconstruction algorithm is designed for a GOOSE message with the highest delay requirement and importance degree by taking satisfaction for requirements of wide-area protection for information as a foundation and taking network load balance after heavy-load communication roundabout as an objective. The algorithm comprehensively considers factors such as the bandwidth, the distance, the channel utilization rate and the like which affect the channel delay, reconstructs a channel weight index, and optimizes a roundabout path with the shortest delay by the aid of a Dijkstra optimization algorithm.

The invention discloses an attack path reconstruction method based on similarity calculation, the invention forms an attack event by integrating multiple attacks in an attack event, Then by forming amultiple attack steps into an attack path, the originally complex network attack process is clarified and organized effectively, which helps to accurately detect the attack behavior, track the attacksource and find the weak links of the network system.

The invention relates to the technical field of network space security, in particular to a network attack path reconstruction method based on topological information, which comprises the following steps of: firstly, obtaining a network topological structure by utilizing topological measurement probes distributed in each area of a network through a traceroute technology, then extracting router information through which an attack flow passes through through a NetFlow/NetStream technology, and finally reconstructing a network attack path according to the extracted router information, and then, according to the extracted routers and related network topology data, adopting a Dijkstra shortest path selection algorithm to gradually reconstruct an attack path of the whole attack flow. According to the method, active and passive network measurement methods are comprehensively utilized, the path information of the network attack can be accurately and completely drawn, and the correctness of a reconstruction structure is ensured.

The invention belongs to the field of network security, and particularly relates to a universal IP traceability system and method based on a block chain. The system is divided into a data acquisition layer, a traceability service layer, a traceability supervision layer and a traceability coordination layer, and a credible service-oriented universal IP traceability system is provided by referring to a business mode of immediate collection and immediate payment in a block chain so as to promote large-scale deployment of the traceability system; the traceability supervision layer comprises a data module, a consensus module and a verification module; the data module simplifies block data, the consensus module improves the working efficiency and performance of the block chain and reduces the deployment cost, and the verification module prevents malicious nodes from tampering transaction information and ensures the data integrity. The traceability method comprises a traceability trace establishment stage and an attack path reconstruction stage. A concurrent backtracking mechanism is adopted, so that the traceability efficiency is improved; the system is also suitable for a traceability method based on a data packet, is compatible with most of current IP traceability methods, and has universality.

The invention discloses an evidence obtaining method and system for a DDOS attack. The method comprises the following steps: reporting an abnormal information table; analyzing the information table; and representing a FMS data packet. The system comprises an NS2 simulation module, an anomaly reporting module, an analysis algorithm module and a fragment packet marking method module. According to the system, the network environment and the network nodes are simulated by using the NS2, so that not only can the real network environment be simulated, but also network paralysis caused by actual attacks can be prevented. All analysis algorithms can be realized in NS2 simply and efficiently without complex tools. According to the method, a botnet and DDoS attack electronic evidence analysis methodis used, fragment data packet marks are used when a server analyzes the electronic evidences so as to track the source and path reconstruction of a router, and therefore the scale recognition rate isincreased to 93%.

The invention discloses a ship area distribution power system intelligent fault diagnosis method based on an FWASSO algorithm and a ship area distribution power system intelligent fault reconstruction method based on an optimization particle swarm algorithm. According to the method, a ship area distribution power system is taken as a control object, and diagnosis problems during main switch faults and circuit breaker faults are researched. The method comprises the following steps: according to data fed back by a system sensor, firstly, carrying out preliminary processing on the data through a fuzzy neural network adaptive Kalman filtering method, and removing interference parts such as noise; and optimizing a kernel function and a penalty factor in kernel principal part analysis through a firework algorithm and a particle swarm algorithm to obtain a diagnosis model for improving the model training speed and the classification accuracy, thereby improving the diagnosis speed and reliability. And finally, an optimal path reconstruction model is constructed by using the improved particle swarm algorithm, so that the speed and precision of fault reconstruction are improved.

The invention discloses an attack and defense evolution method of an uploading channel under DoS attack in microgrid secondary control. The method comprises the following steps: S1, establishing a network attack and defense evolution game model; S2, designing a strong defense model by adopting a prediction mechanism combining an extreme learning machine and model prediction control; S3, designinga weak defense model, and designing related control parameters; and S4, respectively constructing a replication dynamic equation of the defense party and a replication dynamic equation of the attack party, carrying out simultaneous solution on equations, and carrying out evolution stability state analysis according to results. The method gives full play to the function of an information layer, proposes a passive defense strategy of adjacent prediction and path reconstruction, gives out a design method of related parameters, and can efficiently and reasonably cope with the DoS attack problem.

Disclosed herein are a user-participation data selection apparatus and method for updating a positioning infrastructure DB. The user user-participation data selection apparatus includes a data communication unit for transmitting and receiving user-participation data, a movement path reconstruction unit for reconstructing an entire information movement path corresponding to the user-participation data using all information included in the user-participation data, a reliability calculation unit for calculating reliability corresponding to the entire information movement path, and a data selection unit for selecting update-use data for updating a positioning infrastructure DB based on the reliability.