32 results about "Heapsort" patented technology

The invention provides a path optimization method and a path optimization system based on Dijkstra. The method comprises the following steps: getting a shortest path directed graph according to multiple nodes and the path weights between the nodes; initializing an adjacency point set, a marked vertex set of acquired shortest paths, and sorting nodes to be stored; getting a min heap through heap sorting, taking the heap top node of the min heap as an intermediate node, correcting the length of the shortest path from the start node to each node in the complementary set of the marked vertex set via the intermediate node through the intermediate node, performing repeated iteration until all the nodes in the adjacency point set are added to the marked vertex set, and getting the length of the shortest path from the start node to any other node in the nodes. Storage space is saved through data storage in an adjacency table, the running performance of the algorithm is improved through heap sorting, and the processing efficiency is improved.

The invention relates to a column-storage-oriented B+ tree index method for a DWMS (data warehouse management system). The column-storage-oriented B+ tree index method is characterized by comprising a first step, generating column data; a second step, turning to a fourth step for building if a B+ tree keyword is a row number, and turning to a third step for sorting if the B+ tree keyword is not the row number; the third step, sorting column value data by means of using multi-line merging with heapsort; the fourth step, initiating a B+ tree index; a fifth step, creating leaf nodes; and a sixth step, generating a middle nodes an a bottom-up manner. The column-storage-oriented B+ tree index method for the DWMS is used for column storage, and has the advantages that 1), the number of layers of a B+ tree is the smallest, and the number of searching is reduced; and 2), a traditional plug-in method for building the B+ tree is abandoned, and the bottom-up B+ tree creating method is utilized. When the method is used, division operation is omitted, and a lot of expense is reduced.

The invention discloses an A star path finding method and system based on binary heap node sorting. The A star path finding method includes: conducting initialization; judging whether a node n in an OPEN graph is a target solution A, if not, unfolding all follow-up nodes of the node n to form directly correlated sub nodes, putting the judged node n in a CLOSE graph, calculating the appraisal value of each follow-up node of the node n, conducting heat sorting on the OPEN graph according to the appraisal value, putting the node with the smallest appraisal value at the graph head of the OPEN graph, and repeating the steps until all the nodes are sorted. The A star path finding method introduces heat sorting in the normal A star algorithm so that the execution efficiency of the A star algorithm is greatly improved. The A star path finding method overcomes the defects of the traditional A star algorithm of frequently maintaining the OPEN graph and the CLOSE graph, causing long calculating time as the nodes in the OPEN graph have to be sorted each time, and being incapable of meeting high game timeliness requirements.

The invention discloses a quality map phase unwrapping method based on heap sort. Firstly, a phase image to be unwrapped is obtained through flat-earth phase removal and filtering; then, phase quality maps are extracted from the phase image to be unwrapped and sorted through the heap sort method, so that the unwrapping sequence of points to be unwrapped in the phase image to be unwrapped is determined; finally, according to the unwrapping sequence, the points to be unwrapped are unwrapped till the whole phase image to be unwrapped is unwrapped. Quality information of the phase image to be unwrapped is utilized to guide integral unwrapping of the points to be unwrapped, and meanwhile the heap sort method is adopted to sort the phase quality maps, so that time needed for phase unwrapping is shortened, finally, phase unwrapping is rapidly completed with high precision, and a phase-unwrapped image is generated.

The invention provides a receiver detection method assisted by the lattice reduction algorithm and applied to a wireless MIMO system. The receiver detection method assisted by the lattice reduction algorithm and applied to the wireless MIMO system comprises the steps that horizontal moving and zooming are conducted on a receiving vector; a new channel matrix is obtained through the lattice reduction algorithm; in the layer where detection is started, integers within a certain value range are selected to serve as K optimum nodes of the layer; in the other layers, K optimum nodes of each layer are selected according to the method for expanding child nodes by means of father nodes and the candidate least heap sorting algorithm; for the K optimum nodes of the first layer, proper nodes are selected according to specific judgment conditions, and an original sending symbol is obtained by means of a left multiplication transformation matrix through horizontal moving and zooming. According to the receiver detection method assisted by the lattice reduction algorithm and applied to the wireless MIMO system, during multiple-input-multiple-output detection, the lattice reduction assisting algorithm is adopted, the influence of channel correlation on a multiple-input-multiple-output detector is reduced, and the detection performance is improved compared with a detector not based on the lattice reduction assisting algorithm.

The invention provides a Kalman filtering phase unwrapping method having a heapsort function in an embedded manner. An interferometric phase quality map used for guiding a phase unwrapping path can be generated. A heap array based on a complete binary tree is introduced, and the interferometric phase quality value can be used as the ordering key word. The non-boundary image element having the highest phase quality in the interferometric image can be selected, and the adjacent coiled image element can be embedded in the heap array, and the maximum heap can be formed after being adjusted. The interferometric phase value of the optimal to-be-unwrapped image element X of the maximum heap root knot part can be estimated by the Kalman filtering method, and then the unwrapped image element X can be eliminated from the heap array, and maximum heap can be formed after being adjusted. The coiled image element of the adjacent image element x in the interferometric image can be embedded in the heap array, which can be adjusted to be the maximum heap. The previous step can be repeated, till all of the image elements of the heap array are unwrapped, and the unwrapping can be completed. The rapid inserting the rapid ordering of the to-be-unwrapped image elements can be realized, and the coiled phases can be unwrapped along the path from the high quality to the low quality of the interferometric image by adopting the Kalman phase unwrapping method, and the searching consumption time can be reduced, the efficiency and the precision are high, and the stability is strong.

The invention discloses a data processing method based on hardware sorting MapReduce. The method comprises the following steps that CPU-based quick sorting is replaced by GPU-based quick sorting; CPU-based merge sorting is replaced by GPU-based merge sorting; CPU-based heap sort is replaced by GPU-based merge sorting. The GPU-based sort algorithm replaces the CPU-based sort algorithm, the powerful computing capability of a GPU is made full use of, the middle data processing speed is improved, the MapReduce performance is improved, and the method is especially suitable for the large data field.

The invention belongs to the field of heapsort, and especially relates to a sorting method suitable for FPGA implementation. According to the sorting method suitable for FPGA implementation, through designing a new data structure, a binary tree structure of an existing heapsort algorithm is modified to be a 2k-ary tree, k>=2, the degree of parallelism of data comparison is increased, the number of layers of the tree is reduced, and therefore the number of times accessing the data is decreased. By utilization of the function of outputting multiple data in one clock of an RAM of an FPGA, 2k child nodes to be compared can be read out in one clock, and the data reading time is reduced. The 2k-ary tree structure is used, addresses of the child nodes are simply obtained by shifting and adding non-leaf node addresses, thereby facilitating data access in the implementation and reduction of complexity of the implementation. Sorting time delay can be largely reduced by the method, and real-time performance of a sorting module is improved.

The invention relates to the field of image recognition, in particular to a method for identifying shielded fruits in a natural environment. The method comprises: preprocessing, solving an optimal shielding threshold value by utilizing a distance fluctuation curve of adjacent manual marking points on a region edge and a mass center, judging a fruit shielding region and an unshielded edge, and carrying out random three-point circle determination on non-adjacent marking points on the unshielded edge; and taking the obtained circle center point set as the input of an improved cohesion hierarchical clustering classifier based on heap sorting, solving the circle center mean value of each class in the result, and performing circle fitting on the circle center mean value and the farthest markingpoint on the edge of the fruit to realize the identification of the shielded fruit. Compared with the prior art, the recognition method is good in real-time performance, high in recognition rate and small in fluctuation range.

The present invention relates to a system and a method for providing sorted data. The system comprises: a client, configured to generate a query request with a sorting requirement, a plurality of data sources, a link part and a sorting mechanism, wherein each data source generates a sorting data record according to a corresponding sub query request and transmits the data record to a corresponding link; a link part, configured to manage and assign links to the data sources; a sorting mechanism, configured to process the data record of the partial sorting from the data sources via the corresponding links to provide the data record for overall sorting to the client. The sorting mechanism uses a limit mark of a buffer region corresponding to each link to control the use of a memory, and repeatedly performs heapsort on each link to complete the overall sorting of the data.

Embodiments of the invention disclose a task processing method and device and a computer system. The method comprises the steps of constructing at least one high-priority thread and at least one low-priority thread according to high-priority and low-priority task calculation capability quota information; marking priorities for the obtained tasks according to the task priority configuration information; constructing a priority blocking task queue based on heap sorting, and sequentially placing the tasks marked with the priorities in the task queue from high to low according to the priorities; concurrently starting the high-priority thread and the low-priority thread; when any high-priority thread monitors that the priority of the task at the head of the queue in the task queue is not less than a preset first threshold value, taking out the task at the head of the queue and processing the task at the high-priority thread; and when any low-priority thread monitors that the priority of thetask at the head of the queue in the task queue is not less than a second threshold, taking out the task at the head of the queue and processing the task at the low-priority thread. Timely priority processing of the high-priority tasks is ensured.

A computer readable storage medium includes executable instructions to order a binary tree using primary values and selected secondary values required to resolve a position in the binary tree. The remaining secondary values are in a do not care state. A new primary value is compared to the primary value at the root node of the binary tree. If necessary, a new secondary value is compared to the secondary value at the root node to determine whether the new primary value and the new secondary value or the value at the root node should be placed in a sorted list.

A modem chip includes a processor configured to generate instructions, a timing controller configured to respectively generate control signals corresponding to the instructions at the execution times of the instructions, and a plurality of intellectual property blocks, each configured to operate in response to a corresponding control signal of the control signals. The timing controller includes a heap sorting circuit configured to sort the instructions according to execution orders of the instructions based on heap sorting using the execution times, a reference counter configured to generate a reference time, and a signal generator configured to generate a control signal corresponding to a current instruction when the reference time matches the execution time of the current instruction having a highest execution order among the instructions.

The invention discloses a phase unwrapping algorithm based on UIF (unscented information filtering). The phase unwrapping algorithm comprises the following steps of firstly, establishing an UIF phase unwrapping algorithm recursive state estimation model, utilizing a phase gradient estimation method based on an AMPM (amended matrix pencil model) to obtain the phase gradient information required by the recursive state estimation model, and utilizing an LEVENBERG-MARQUARDT method to optimize the recursive state estimation model, so as to improve the algorithm convergence; introducing a heapsort quick mass map guide strategy, using the adjacent wrapping pixel of the unwrapped pixel as a heap array of a to-be-unwrapped pixel, and adjusting the heap array as the maximum heap according to the quality value of the to-be-unwrapped pixel; in each unwrapping step, utilizing the UIF phase unwrapping algorithm recursive state estimation model to unwrap the optimum to-be-unwrapped pixel at the root node of the heap array, deleting the pixel from the heap array, and adjusting the heap array as the maximum heap, until the phases of all wrapped pixels are unwrapped. The phase unwrapping algorithm has the characteristics that the accuracy is higher, and the stability is higher.

The invention discloses a parallel acceleration method for sequencing big data genome comparison files. The method comprises steps of reading and decompressing a target BAM file, and storing the target BAM file into a continuous first buffer area B; performing multi-thread sorting after the first buffer area B is full, performing merging through heap sorting to form intermediate files; reading theintermediate files in sequence, putting the intermediate files into an associated second buffer area MB, and merging the data of each second buffer area MB through heap sorting; compressing the merged data through a plurality of threads, and writing the compressed data into a result file. The method is advantaged in that the threads are independently distributed for reading and decompressing, thread pools are constructed for decompressing and compressing respectively, the number of developed threads is reduced, multi-thread resources are fully utilized, file reading and writing efficiency isimproved, the number of intermediate files is reduced, the number of times of memory copying operation is reduced, and the processing time is shortened.

The invention relates to a column-storage-oriented B+ tree index method for a DWMS (data warehouse management system). The column-storage-oriented B+ tree index method is characterized by comprising a first step, generating column data; a second step, turning to a fourth step for building if a B+ tree keyword is a row number, and turning to a third step for sorting if the B+ tree keyword is not the row number; the third step, sorting column value data by means of using multi-line merging with heapsort; the fourth step, initiating a B+ tree index; a fifth step, creating leaf nodes; and a sixth step, generating a middle nodes an a bottom-up manner. The column-storage-oriented B+ tree index method for the DWMS is used for column storage, and has the advantages that 1), the number of layers of a B+ tree is the smallest, and the number of searching is reduced; and 2), a traditional plug-in method for building the B+ tree is abandoned, and the bottom-up B+ tree creating method is utilized. When the method is used, division operation is omitted, and a lot of expense is reduced.

The invention provides a face recognition method based on wavelet dimension reduction under homomorphic encryption, relates to the technical field of face recognition, and solves the problems that anexisting face recognition method based on homomorphic encryption is large in occupied space, long in face recognition time and poor in real-time performance in the face recognition process. The methodcomprises the steps of preprocessing a face picture input by a client and a face picture of a face database, then carrying out discrete wavelet transform in a homomorphic encryption domain on the encrypted face picture transmitted by the client, carrying out wavelet transform on the face picture in the face database in a plaintext state, then carrying out dimension reduction processing transform,thus reducing resource space occupancy rate and time consumed in the face recognition process. In addition, based on the homomorphic encryption property, the encrypted face picture can only be decrypted by an image owner, the requirement of both a cloud server and a client for privacy is guaranteed, and the comparison frequency can be reduced by comparing the Euclidean distance based on heap sorting, thus improving face recognition speed.

The invention discloses a phase unwrapping method based on rank information filtering, and the method combines a local phase gradient estimation technology based on AMPM with a rank information filter, builds a phase unwrapping program based on rank information filtering, and converts an interferogram phase unwrapping problem into a state estimation problem under a rank information filtering framework. An H-infinity operator is introduced to optimize a state variable information matrix, and the state variable estimation precision is improved; a fast path tracking strategy based on heap ordering is utilized to guide a phase unwrapping path, and it is ensured that a phase unwrapping program based on rank information filtering unwraps an interferogram along a path from a high-quality pixel to a low-quality pixel. Simulation data and actual measurement data experiment results show that the algorithm is effective, and a stable result can be obtained from a noise winding interference pattern.

The invention discloses a method for phase unwrapping of a quality map based on heap sorting. First, the phase image to be unwrapped is obtained by removing the flat-ground effect and filtering, and then the phase quality map is extracted from the phase image to be unwrapped, and the heap sorting method is used to unwrap the phase image. The phase quality map is sorted, that is, it is determined that the unwrapping points in the phase image to be unwrapped are smooth, and finally, the unwrapped points are unwrapped according to the unwrapping sequence until the entire phase image to be unwrapped is unwrapped. The present invention utilizes the quality information of the phase image to be unwrapped to guide the points to be unwrapped to carry out integral unwrapping, and at the same time adds the heap sorting method to sort the phase quality map, thereby shortening the time required for phase unwrapping, and finally achieving fast and high-precision The phase unwrapping is completed and the phase unwrapped image is generated.

The present invention relates to a system and a method for providing sorted data. The system comprises: a client, configured to generate a query request with a sorting requirement, a plurality of data sources, a link part and a sorting mechanism, wherein each data source generates a sorting data record according to a corresponding sub query request and transmits the data record to a corresponding link; a link part, configured to manage and assign links to the data sources; a sorting mechanism, configured to process the data record of the partial sorting from the data sources via the corresponding links to provide the data record for overall sorting to the client. The sorting mechanism uses a limit mark of a buffer region corresponding to each link to control the use of a memory, and repeatedly performs heapsort on each link to complete the overall sorting of the data.

This invention relates to a method for heapsort in SDH including: after a network control protocol decompresses received compressed messages, it adds an ardinal number of a preset length of byte at the end of each record used in marking the storage sequence of said record to be generated by an accumulator, said network control protocol puts the decompressed record together with the corresponding sequence in the storage and puts the data in good order in the library first when packaging and sets up initial heaps to data according to key fields, if it finds that the key fields recorded in the two messages are complete the same when setting up the initial heaps, it compares the successive relation of their corresponding ordinal numbers, if finding the key fields are the same when the control protocol re-sets up the heaps, then it compares the order relation of the corresponding ordinal numbers.

A method for heap sorting in SDH of the present invention, which comprises: after the network control protocol decompresses the received compressed message, add a sequence number of a predetermined length byte at the end of each record to identify the record The order of storage, and generated by an accumulator; the network control protocol puts the decompressed record into storage together with the sequence number corresponding to the record; Sort the data in the library, and build the initial heap according to the key fields of the data in the library. When creating the initial heap, if the key fields of the two message records are found to be identical, compare the sequence of the corresponding ordinal numbers; During the rebuilding process of the network control protocol, if it is found that the key fields of the two message records are completely the same, the sequence relationship of the corresponding ordinal numbers is compared. The method of the invention realizes a simple and effective method for improving heap sorting into stable sorting.

A method for determining a pair of compliance events is provided. A set of event data collected by a PET detector may be divided into N sub-sets, and each of the N sub-sets of event data may be sorted according to an occurrence time of each event included in the event data, wherein N is an integer greater than or equal to 2. A K stage heapsort may be performed on the sorted N sub-sets of event data according to the occurrence time of each event, so as to sequentially output data of each event in the sorting order, wherein K is an integer greater than or equal to 1. A time compliance determination and a space compliance determination may be performed on the sequentially-outputted event data, and a pair of compliance events may be determined if both the time compliance determination and the space compliance determination are passed.

The invention discloses a phase unwrapping algorithm based on UIF (unscented information filtering). The phase unwrapping algorithm comprises the following steps of firstly, establishing an UIF phase unwrapping algorithm recursive state estimation model, utilizing a phase gradient estimation method based on an AMPM (amended matrix pencil model) to obtain the phase gradient information required by the recursive state estimation model, and utilizing an LEVENBERG-MARQUARDT method to optimize the recursive state estimation model, so as to improve the algorithm convergence; introducing a heapsort quick mass map guide strategy, using the adjacent wrapping pixel of the unwrapped pixel as a heap array of a to-be-unwrapped pixel, and adjusting the heap array as the maximum heap according to the quality value of the to-be-unwrapped pixel; in each unwrapping step, utilizing the UIF phase unwrapping algorithm recursive state estimation model to unwrap the optimum to-be-unwrapped pixel at the root node of the heap array, deleting the pixel from the heap array, and adjusting the heap array as the maximum heap, until the phases of all wrapped pixels are unwrapped. The phase unwrapping algorithm has the characteristics that the accuracy is higher, and the stability is higher.

The invention provides a receiver detection method assisted by the lattice reduction algorithm and applied to a wireless MIMO system. The receiver detection method assisted by the lattice reduction algorithm and applied to the wireless MIMO system comprises the steps that horizontal moving and zooming are conducted on a receiving vector; a new channel matrix is obtained through the lattice reduction algorithm; in the layer where detection is started, integers within a certain value range are selected to serve as K optimum nodes of the layer; in the other layers, K optimum nodes of each layer are selected according to the method for expanding child nodes by means of father nodes and the candidate least heap sorting algorithm; for the K optimum nodes of the first layer, proper nodes are selected according to specific judgment conditions, and an original sending symbol is obtained by means of a left multiplication transformation matrix through horizontal moving and zooming. According to the receiver detection method assisted by the lattice reduction algorithm and applied to the wireless MIMO system, during multiple-input-multiple-output detection, the lattice reduction assisting algorithm is adopted, the influence of channel correlation on a multiple-input-multiple-output detector is reduced, and the detection performance is improved compared with a detector not based on the lattice reduction assisting algorithm.

The invention discloses an efficient codebook selection method for polar code fast serial cancellation list decoding, relates to the technical field of polar code decoding, and is used for solving the problems that an existing polar code fast serial cancellation list decoding method is redundant in operation and high in decoding delay. The method comprises the following steps: for two outer codes: Rate-1 and SPC outer codes, respectively calculating according to a code word error pattern to obtain a plurality of path reliability measures corresponding to L paths and code words to be selected; selecting a path reliability metric minimum value from each group of multiple path reliability metrics to obtain a minimum path metric set PM and a to-be-selected optimal code word set; comparing in the set PM by using a small root heap sorting method, selecting the path reliability metric PM1 with the minimum numerical value in the set, and selecting the codeword corresponding to the path reliability metric PM1 from the optimal codeword set to be selected as the optimal codeword; and popping the PM1 out of a set PM, repeating the steps, and selecting the path reliability measurement PM2,..., PMn with the minimum numerical value and the corresponding optimal code word every time until L optimal code words are obtained.

The invention relates to a 5G signaling analysis method and system, and belongs to the field of communication network signaling monitoring. The method comprises the following steps: sorting signalingkey messages Key by utilizing a large top heap according to the Value occurrence probability of the signaling key messages Key; mapping the Key to a hash table, determining the position of the Key inthe linked list during conflict processing according to the probability, and placing the Key value with the larger probability at the position with the smaller conflict in the hash table; accessing ahash table record, and associating and synthesizing the same signaling process information of the same user. According to the invention, the efficiency of processing hash conflicts in signaling synthesis and multi-protocol association is improved, a signaling accurate monitoring function in a signaling analysis system is realized, and the accurate positioning of a communication network user and acorresponding signaling analysis file is realized.

The present invention proposes a MapReduce optimization method based on GPU sorting, wherein MapReduce includes a Map stage, a Shuffle stage, and a Reduce stage, the Map stage includes a Spill process and a Merge process, and the Reduce stage includes a Merge process, wherein the Spill process in the Map stage The GPU-based quick sorting process is adopted in the process, and the GPU-based merge sorting process is adopted in the Merge process of the Map stage and the Merge process of the Reduce stage. By replacing the traditional CPU-based quick sort, merge sort, and heap sort algorithms with GPU-based quick sort and merge sort algorithms, the intermediate data processing speed is improved, thereby improving the performance of MapReduce.