30results about How to "Reduce computing time overhead" patented technology

The invention, which belongs to the technical field of car networking communinication, relates to an adaptive rate control method based on mobility and DSRC / WAVE network relevance feedback. The method comprises establishment of a traffic flow density prediction module, a t+1 time communication interference calculation module, an SINR calculation module, a t+1 time available link bandwidth calculation module, a channel congestion cost calculating module and an adaptive message generation rate calculation module. A traffic flow density value at a next time is predicted; according to the density value of the next time, a transmitting power, and a rate, an interference module of a communication process is established, a signal to noise ratio is calculated, and an available link bandwidth of a node at the next time is predicted; on the basis of mismatching of a transmission rate and mismatching of a transmission queue length, a channel congestion cost module is established, so that a message generation rate at the next time is adjusted adaptively. According to the method, adaptive rate adjustment is carried out in advance by using the prediction technology, so that the channel congestion is avoided; and the low communication delay and the high data packet transmission rate are guaranteed with the low calculation time and cost.

The invention discloses a transient frequency acquisition method based on a stack noise reduction automatic encoder, which can be used for comprehensively evaluating the transient frequency stabilityof a power system by predicting a multi-dimensional frequency index value after a disturbance accident, and is an important basis for formulating emergency control strategies such as a high cycle cutting machine and low frequency load reduction. Firstly, important characteristic variables are screened out through a random forest algorithm to serve as input data, and input data dimension reductionis achieved; then, a plurality of noise reduction automatic encoders are stacked, and a deep learning network structure is constructed; a pre-training- parameter fine tuning method is used for training network parameters; in the pre-training process, a Dropout technology is introduced to improve the algorithm generalization ability, overfitting is prevented, fine tuning is conducted on network parameters based on an RMSProp optimization method, the algorithm rate is increased, the probability of falling into local optimum is reduced, and the stack noise reduction automatic encoder can effectively represent a complex function and improve the prediction precision and generalization ability.

The invention relates to an indoor WLAN fingerprint database multipoint self-adaptive updating method based on a radial basis interpolation. According to the method, based on the radial basis neural network interpolation, received signal strength (RSS), actually measured, of a few feedback points is used for building a mathematical interpolation model and constructing an RSS estimation curved surface, and RSS estimated values of reference points close to the feedback points are obtained by means of correlation of spatial positions. The RSS values of the multiple reference points can be updated in each time of calculation by using the method, on the premise of not repeatedly updating the reference points, a physical environment is divided into P updating regions, different cut-off region radiuses are adopted in the different updating regions according to the difference of the density and the number of the feedback points in the corresponding cut-off regions where the updating regions are located, the radiuses are obtained by calculating a partial least squares regression model obtained through fingerprint information of an old database, and a whole database can be updated after P times of calculation. The problem that as the fingerprint information of the database changes too much along with time change, positioning accuracy is reduced is solved, time expenditure of database updating is effectively reduced, and the WLAN positioning accuracy is greatly improved.

The invention discloses an indoor WLAN (wireless local area network) positioning networking method established on the basis of SimBank similar combined adjacent graphs. The method comprises the following steps: firstly, constructing a physical position adjacent graph and a received signal strength (RSS) adjacent graph among different position reference points in a positioning target environment, and combining the two adjacent graphs according to a certain rule to obtain a reference point adjacent graph; secondly, establishing a SimBank model from a graph theory to calculate the similarity among the reference points, and establishing an objective function for WLAN access point (AP) position optimization on the basis of increasing the differences among the reference points (or known as reducing the similarity among the reference points); and finally, solving the objective function by using a simulated annealing algorithm to realize rapid optimization of AP positions. By adopting the method, while relatively high positioning accuracy is ensured, the computing time cost of a positioning system under a multi-AP condition (when the number of the APs is more than 2) is significantly reduced.

The invention discloses a response matrix parallel processing implementation method based on an integral variable block method, which comprises the following steps of: constructing different response matrix sets by using a plurality of computing cores in a parallel manner by using the integral variable block method and having natural decoupling property by using the construction among the different response matrix sets; secondly, dividing a reactor core into a corresponding number of non-overlapped sub-regions according to a given calculation core number for a transport equation solving part in an integral variable segmentation method, and performing Gauss-Seidel method solution according to a neutron balance relationship in the block and a neutron flow continuity relationship on the surface of the block to obtain a converged neutron flux density expansion moment and an incident neutron flow density expansion moment. According to the invention, by using an information transfer interface protocol (MPI) parallel method, a parallel computing function of several computing cores or even dozens of computing cores in a Windows or Linux system can be realized, and the computing time cost of an original computing case is shortened by several times or even dozens of times.

The invention provides a group buying information transmission method and a group buying system. The method includes that after a group buying website receives a group buying request, a group buying website authentication request is generated, wherein the group buying website authentication request carries a website identification of the group buying website, a commodity identification of a group-bought commodity and corresponding prestored signiture data; a media exchange center generates a random number and a transaction serial number according to the group buying website authentication request; the group buying website takes an invited user number group carried in the group buying request as a corresponding group buying invited user terminal and generates a group buying notice; the group buying website combines the group buying notice, the random number and the transaction serial number into group buying information; the media exchange center checks the corresponding random number according to the transaction serial number, and verification is passed if checking is successful. As the prestored signiture data are obtained through calculation of the group buying website specific to the commodities, cost of time calculation during a private key signiture generation is reduced and work efficiency is improved.

The invention discloses a multi-temperature target set value switching multi-parameter predictive control algorithm for a process medium of a circulating cooling water system. The multi-parameter predictive control algorithm comprises the following specific steps: acquiring historical data of the circulating cooling water system; screening characteristic variables from the historical data, and dividing the characteristic variables into a training sample set and a test sample set after data processing; carrying out greedy unsupervised pre-training on the training sample set layer by layer to obtain a weight matrix W of an input layer and a hidden layer of deep neural network initialization, and a threshold matrix B of the input layer and the hidden layer; performing parameter fine tuning toobtain an initial process medium temperature prediction model; testing and evaluating again; and obtaining a cooling water supply and return pressure difference set value and a process medium temperature predicted value by obtaining current data. The method has the beneficial effects of accurate cooling and high precision, and can save energy to the maximum extent.

The invention discloses a dichotomy-based multi-level simplified space voltage pulse width modulation (SVPWM) adjustment strategy. The adjustment strategy comprises the steps of firstly, performing Alpha Beta 2gh coordination conversion on a reference voltage vector under a rectangular coordinate system; performing vertical integer operation on reference vector coordinate under a gh coordinate system, judging the nearest triangular position where the reference vector is located on the basis of dichotomy, and calculating acting time of neatest three basic vectors under the gh coordinate system on the basis of a volt-second balancing principle; and finally, arranging an acting sequence of the basic vectors. According to the adjustment strategy, simple arithmetic is only needed in calculation of the basic vector acting time, a large amount of trigonometric function calculation is omitted, and the operation time of a processor is reduced; the dichotomy is used for judging the reference vector position, the position where the reference vector is located can be rapidly searched, so that a switch vector sequence can be directly generated according to control requirement, and complicated logic judgment is omitted; and with the increase of level number, no influence is generated on simplification of complexity of a space vector modulation algorithm, and the adjustment strategy has relatively good adaptability.

The invention discloses an MIMO system transmitting terminal pre-encoding method based on mercury water injection. The method comprises the following steps: S1, using an SVD technology to decompose a channel matrix to obtain N parallel sub-channels; S2, carrying out power allocation for the signal-to-noise ratio gamma i of the ith sub-channel according to a mercury water injection method to obtain a power allocation matrix sigma MWF; S3, constituting a modulation diversity unitary matrix VM according to a modulation diversity planisphere rotation method to reduce overlapped constellation points; and S4, using a product of an original pre-encoding matrix VH in S1, the matrix sigma MWF in S2 and the matrix VM in S3 as a transmitting terminal pre-encoding matrix, namely, G = VH sigma MWF VM, to obtain the pre-encoding method disclosed by the invention, and using the matrix to transmit signals. By adopting the MIMO system transmitting terminal pre-encoding method disclosed by the invention, a pre-encoding computation time cost is reduced, mutual information of the channel is improved, and the requirements of practical application are satisfied.

The invention discloses an indoor WLAN access point arranging method based on a fast water-filling algorithm. The method has the advantages that a water-filling optimization problem model is built on the basis of a transmitting power self-adaptation distribution method according to channel conditions of a simulation wireless channel, the AP weight which should be distributed to each AP candidate position is acquired by analyzing the influence of different AP placing positions on reference point fingerprint difference and using the fast water-filling algorithm, fast optimization of the AP positions is achieved according to the influence of the fingerprint difference on the precision of a fingerprint positioning method, and system calculation time expense is lowered evidently.

The invention discloses a cloud computing task scheduling method based on a genetic algorithm. The method comprises the steps of receiving a resource allocation request from a client, randomly generating a probability matrix Lambda for all the sub-tasks according to the resource allocation request, processing the task dependency graph in the step (1) by adopting a breadth-first search algorithm; therefore, acquiring a plurality of sub-task sequences; sequentially allocating each sub-task in each sub-task sequence to a physical machine corresponding to the maximum random probability value of the row where the task unit to which the sub-task belongs in the probability matrix is located; further forming a parent population, calculating the time cost and the energy cost of each physical machine for processing the corresponding task unit, and obtaining the total time cost and the total energy cost of all the physical machines corresponding to each element in the parent population accordingto the time cost and the energy cost so as to form a vector set. The technical problem that an effective mechanism capable of reducing the cloud computing time cost and the cloud data center energy consumption at the same time is lacked in the existing cloud computing field can be solved.

The invention discloses a quick point searching method for partition packaging, a storage medium and a terminal, and belongs to the technical field of computational fluid mechanics and grids, and the method comprises the steps: building a bounding box of each grid block in a structured grid system, and dividing the bounding box into a plurality of subspaces; obtaining attribute information of grid points in each subspace in each bounding box, and packaging and storing the attribute information; determining a subspace of a bounding box corresponding to a given grid point; and calculating the distances between the given grid point and all grid points in the subspace to which the given grid point belongs, and determining the nearest grid point of the given grid point by comparing the distances. According to the method, after the attribute information of the grid points in each subspace in each bounding box is obtained, packaged and stored, only the distances between the given grid point and all the grid points in the subspace to which the given grid point belongs need to be calculated, the nearest grid point of the given grid point is determined by comparing the distances, the calculation amount is small, the calculation time cost can be greatly saved, and the calculation efficiency is improved.

The invention discloses an incremental clustering algorithm based on community detection. The algorithm is characterized in that the algorithm comprises the steps: employing a Commonity community concept and an Online-Offline two-stage framework, and introducing an IMC concept; obtaining a similarity graph of the target corpus through a similarity calculation method according to the document representation word vector, the representation keyword of the document and the named entity identification prediction word of the document; and finally, processing the similarity graph through a Louvain algorithm to obtain an initialized community result, and employing an incremental clustering algorithm on the basis of the initialized community result to obtain a final clustering result. Compared withthe prior art, the method has the advantages that the calculation time overhead is reduced under the same hardware condition, and the clustering result is quickly generated, so that upstream and downstream services of an application scene can be better served, timely response is achieved, the function of distinguishing hot events from continuously reported events is achieved, and effective clustering and event-level filtering are conducted on news events.

The invention discloses a self-adapting peak search method for a spectrum of a semiconductor laser. The self-adapting peak search method comprises the following steps of: establishing a local maximum matrix of a spectrum data sequence, correcting the matrix with a row vector feature, positioning a peak value according to a column vector feature of the corrected matrix, and performing further correction and compensation with a certain method. The method has relatively good noise resistance, and the whole search process is free of human intervention, so that the adaptivity and the robustness are high and real-time calculation requirements are met. An actually measured data calculation result shows that compared to a direct comparison method, an optimized derivative method and a genetic algorithm, the self-adapting peak search method has the characteristics that the advantages in check accuracy and calculation time are obvious, the average check accuracy can reach 98%, the average calculation time is only 0.12 seconds, and the self-adapting peak search method can be applied to real-time analysis of actually measured spectrums of semiconductor lasers.

The invention discloses a transient frequency acquisition method based on a stack noise-reduction autoencoder, which can be used to comprehensively judge the transient frequency stability of a power system by predicting the multi-dimensional frequency index value after a disturbance accident. An important basis for emergency control strategies such as machine and low-frequency load shedding. The present invention first adopts the random forest algorithm to screen out important characteristic variables as input data to realize dimensionality reduction of the input data; then, stacks multiple noise-reduction autoencoders to build a deep learning network structure; adopts the method of "pre-training-parameter fine-tuning" for training Network parameters. In the pre-training process, dropout technology is introduced to improve the generalization ability of the algorithm and prevent overfitting. The network parameters are fine-tuned based on the RMSProp optimization method to speed up the algorithm speed and reduce the probability of falling into local optimum. The stack noise reduction is automatic Encoders can effectively represent complex functions, improve prediction accuracy and generalization ability.

The invention discloses a homomorphic encryption neural network framework of a PS and PL collaborative architecture and a reasoning method. The framework comprises a PL side and a PS side, the PL side comprises a structure parameter analysis unit, a plaintext * ciphertext unit and a structure parameter scheduling unit; the structure parameter analysis unit is used for receiving and analyzing the DNN model structure parameters sent by the PS side; the data parameter scheduling unit is used for caching the received weight parameters of the PS side and the order of a polynomial in a ciphertext domain, splicing and outputting to the plaintext * ciphertext unit; and the plaintext * ciphertext unit is used for performing polynomial multiplication operation on the received data in the ciphertext domain and sending a result to the PS. The PS side comprises a convolution summation unit, a partial sum accumulation unit, a BN unit, a data updating unit, a global average pooling unit and a full connection unit. According to the method, the PS side and the PL side work cooperatively, the execution efficiency of the picture classification tasks is improved, and the reasoning time is shortened.

The invention relates to an indoor WLAN fingerprint database multipoint self-adaptive updating method based on a radial basis interpolation. According to the method, based on the radial basis neural network interpolation, received signal strength (RSS), actually measured, of a few feedback points is used for building a mathematical interpolation model and constructing an RSS estimation curved surface, and RSS estimated values of reference points close to the feedback points are obtained by means of correlation of spatial positions. The RSS values of the multiple reference points can be updated in each time of calculation by using the method, on the premise of not repeatedly updating the reference points, a physical environment is divided into P updating regions, different cut-off region radiuses are adopted in the different updating regions according to the difference of the density and the number of the feedback points in the corresponding cut-off regions where the updating regions are located, the radiuses are obtained by calculating a partial least squares regression model obtained through fingerprint information of an old database, and a whole database can be updated after P times of calculation. The problem that as the fingerprint information of the database changes too much along with time change, positioning accuracy is reduced is solved, time expenditure of database updating is effectively reduced, and the WLAN positioning accuracy is greatly improved.

The invention, which belongs to the technical field of car networking communinication, relates to an adaptive rate control method based on mobility and DSRC / WAVE network relevance feedback. The method comprises establishment of a traffic flow density prediction module, a t+1 time communication interference calculation module, an SINR calculation module, a t+1 time available link bandwidth calculation module, a channel congestion cost calculating module and an adaptive message generation rate calculation module. A traffic flow density value at a next time is predicted; according to the density value of the next time, a transmitting power, and a rate, an interference module of a communication process is established, a signal to noise ratio is calculated, and an available link bandwidth of a node at the next time is predicted; on the basis of mismatching of a transmission rate and mismatching of a transmission queue length, a channel congestion cost module is established, so that a message generation rate at the next time is adjusted adaptively. According to the method, adaptive rate adjustment is carried out in advance by using the prediction technology, so that the channel congestion is avoided; and the low communication delay and the high data packet transmission rate are guaranteed with the low calculation time and cost.

The invention discloses a fast MIMO system transmitting terminal precoding method, comprising the following steps: S1, using SVD technology to decompose a channel matrix to obtain N parallel sub-channels; S2, carrying out power allocation in view of the signal-to-noise ratio gamma i of the ith sub-channel according to a classic injection method to obtain a power allocation matrix sigma CWF; S3, constituting a modulation diversity unitary matrix VM according to a modulation diversity constellation rotation method to reduce the overlap of constellation points; S4, using a product of an original precoding matrix VH in S1, the matrix sigma CWF in S2 and the matrix VM in S3 as a transmitting terminal precoding matrix, namely, G=VH sigma CWFVM, obtaining the precoding method disclosed by the invention, and using the matrix to transmit signals. The fast MIMO system transmitting terminal precoding method disclosed by the invention is used for reducing the precoding computation time cost, improving the mutual information of the channels and meeting the demands of practical application.

The invention discloses a fast point-finding method, a storage medium and a terminal for partition encapsulation, and belongs to the field of computational fluid dynamics and grid technology. Divide into multiple subspaces; obtain the attribute information of grid points in each subspace in each bounding box and package and store them; determine the subspace of the bounding box corresponding to a given grid point; calculate all The distance between grid points, by comparing the size of the distance to determine the nearest grid point to a given grid point. After the present invention obtains the grid point attribute information in each subspace in each bounding box and encapsulates and stores it, it only needs to calculate the distance between a given grid point and all grid points in the subspace to which it belongs, and then determine the distance by comparing the distance Given the nearest grid point of a grid point, the calculation amount is small, which can greatly save the calculation time and improve the calculation efficiency.

The invention discloses a CRISPR induced RNA library design method. The method comprises the following steps: 1, generating a kmer set according to a reference genome; 2, segmenting the kmer into a kmer1 part and a kmer2 part, taking the corresponding kmer2 with the same kmer1 as one category, and constructing the kmer2 of the same category into the same retrieval tree, wherein the key sequence ofeach retrieval tree is kmer1 corresponding to the kmer2; 3, acquiring an induced RNA and an off-target sequence thereof in parallel; in the step, when the kmer is compared with a kmer formed by connecting a key sequence of a retrieval tree and kmer2 in the key sequence, the kmer1 of the kmer is compared with the key sequence of the retrieval tree, so as to check whether a set condition is met ornot is judged: if yes, the kmer2 of the kmer is continuously compared with the kmer2 in the retrieval tree. According to the invention, the calculation efficiency is improved.

The invention discloses an indoor WLAN (wireless local area network) positioning networking method established on the basis of SimBank similar combined adjacent graphs. The method comprises the following steps: firstly, constructing a physical position adjacent graph and a received signal strength (RSS) adjacent graph among different position reference points in a positioning target environment, and combining the two adjacent graphs according to a certain rule to obtain a reference point adjacent graph; secondly, establishing a SimBank model from a graph theory to calculate the similarity among the reference points, and establishing an objective function for WLAN access point (AP) position optimization on the basis of increasing the differences among the reference points (or known as reducing the similarity among the reference points); and finally, solving the objective function by using a simulated annealing algorithm to realize rapid optimization of AP positions. By adopting the method, while relatively high positioning accuracy is ensured, the computing time cost of a positioning system under a multi-AP condition (when the number of the APs is more than 2) is significantly reduced.

The invention relates to a Wi-Fi (wireless fidelity) positioning fingerprint database construction method. The major problem for the positioning method based on signal fingerprints is indoor environment surveying and position fingerprint acquisition with the heavy workload. Besides, with the change of the indoor environment temperature, the layout and the like, the positioning accuracy is substantially reduced. The method comprises steps as follows: firstly, selecting the total number of reference points at equal intervals and acquiring RSS (received signal strength) at the selected reference points; then mapping physical positions of the reference points and corresponding signal fingerprints into small images, and enlarging the images with an improved bi-cubic image interpolation method; finally, performing image denoising with an adaptive median filter method, mapping the images into fingerprint information, filtering out singular points, obtaining RSS estimated values of all the reference points in the environment, and constructing a Wi-Fi positioning fingerprint database. The method can effectively reduce data acquisition manpower and time and can be applied to indoor wireless communication network environments.

The invention discloses an MIMO system transmitting terminal pre-encoding method based on mercury water injection. The method comprises the following steps: S1, using an SVD technology to decompose a channel matrix to obtain N parallel sub-channels; S2, carrying out power allocation for the signal-to-noise ratio gamma i of the ith sub-channel according to a mercury water injection method to obtain a power allocation matrix sigma MWF; S3, constituting a modulation diversity unitary matrix VM according to a modulation diversity planisphere rotation method to reduce overlapped constellation points; and S4, using a product of an original pre-encoding matrix VH in S1, the matrix sigma MWF in S2 and the matrix VM in S3 as a transmitting terminal pre-encoding matrix, namely, G = VH sigma MWF VM, to obtain the pre-encoding method disclosed by the invention, and using the matrix to transmit signals. By adopting the MIMO system transmitting terminal pre-encoding method disclosed by the invention, a pre-encoding computation time cost is reduced, mutual information of the channel is improved, and the requirements of practical application are satisfied.

The invention discloses an indoor WLAN access point arranging method based on a fast water-filling algorithm. The method has the advantages that a water-filling optimization problem model is built on the basis of a transmitting power self-adaptation distribution method according to channel conditions of a simulation wireless channel, the AP weight which should be distributed to each AP candidate position is acquired by analyzing the influence of different AP placing positions on reference point fingerprint difference and using the fast water-filling algorithm, fast optimization of the AP positions is achieved according to the influence of the fingerprint difference on the precision of a fingerprint positioning method, and system calculation time expense is lowered evidently.

The invention provides a group buying information transmission method and a group buying system. The method includes that after a group buying website receives a group buying request, a group buying website authentication request is generated, wherein the group buying website authentication request carries a website identification of the group buying website, a commodity identification of a group-bought commodity and corresponding prestored signiture data; a media exchange center generates a random number and a transaction serial number according to the group buying website authentication request; the group buying website takes an invited user number group carried in the group buying request as a corresponding group buying invited user terminal and generates a group buying notice; the group buying website combines the group buying notice, the random number and the transaction serial number into group buying information; the media exchange center checks the corresponding random number according to the transaction serial number, and verification is passed if checking is successful. As the prestored signiture data are obtained through calculation of the group buying website specific to the commodities, cost of time calculation during a private key signiture generation is reduced and work efficiency is improved.