599results about How to "Improve error correction performance" patented technology

The invention discloses a polarization code and multi-bit even parity check code cascaded error correction coding method. The method comprises the steps: a transmitting end encoder utilizes a multi-bit even parity check code as an outer code, and utilizes a polarization code as an inner code; a receiving end decoder decodes by utilizing a modified successive cancellation list (SCL) decoding algorithm. On the aspect of error correction performance, comparing with the prior art utilizing middle-short code length non-cascaded polarization codes of the SCL decoding algorithm, the polarization code and multi-bit even parity check code cascaded error correction coding method has the advantages that frame error rate performance of a system can be remarkably improved, and a maximum likelihood bound (ML Bound), which cannot be broken through by the SCL decoding algorithm, can be remarkably broken through. On the aspect of engineering realization, according to the polarization code and multi-bit even parity check code cascaded error correction coding method, the outer code utilizes the multi-bit even parity check code, which is simple to code; the modified SCL decoding algorithm is utilized to decode, bit decision and even parity check are combined to be carried out in a decoding process, and compared with the original SCL decoding algorithm, the method provided by the invention does not increase the decoding complexity, and facilitates the engineering realization.

The invention discloses a user-oriented cloud storage data integrity protection method. The method includes the steps of 1, uploading, wherein a user side segments and numbers a file to generate intra-block redundancy check codes, the hash value of each file block is calculated and saved as a range-based 2-3 tree, the hash values are linked into one value and signed with private keys, and inter-block redundancy check codes are generated and encrypted with stream substitution secret keys; 2, verification, wherein a cloud management node sends a verification request to a corresponding cloud storage server, corresponding original data and algebraic signatures of the inter-block redundancy check codes corresponding to the original data are calculated and returned to the user side for data integrity verification, and the step 3 is executed if data destruction is found; 3, recovery, wherein the user side acquires all the file blocks on corresponding branch trees from a cloud side according to tree information and performs error recovery on the file blocks through erasure codes of the file blocks. According to the method, integrity verification and recovery are performed on the dynamic data stored in the cloud side from the user perspective.

The present invention relates to an error correction encoding method using a three-dimensional Reed-Solomon code. In the error correction encoding method, pieces of input information are arranged in a three-dimensional data block. Three-dimensional error correction encoding is performed with respect to the three-dimensional data block, thereby adding horizontal, vertical and z-axial error correction parity symbols to the three-dimensional data block in horizontal, vertical and z-axial directions, respectively.

The invention provides an error correction method and device for a search query term. After the error correction device for the search query term receives the query term in a search request, firstly, the error correction candidate sub-term list of each sub-term in the query term is obtained; then, according to a pre-established N-Gram language model, each error correction candidate term spliced by each error correction candidate sub-term list is subjected to real-time on-line grading to obtain an error correction target term list; and finally, according to the score of each error correction target term in the error correction target term list, a final error correction result is output. By use of the method and the device, on the basis of on-line analysis and calculation, the search query term input by the user is subjected to error correction in real time, off-line mass data processing and historical error correction databases are not required, and large-scale search logs and user feedback are not depended. For application scenes including mobile application markets and the like, error correction capability can be effectively improved.

The invention provides a multi-node synchronous sampling control and data transmission method, the steps are as follows:(1) an overall synchronizer that matches with the upper layer data processing equipment and the process variable acquisition control equipment (DAE) is constructed, the equipment produces an overall synchronous signal S1 by utilizing an outer sync source or a local timer; (2) the overall synchronous signal is accessed to one or more data communication equipment (DCE), which produce a synchronous signal S2 for each accessed process variable acquisition control equipment; (3) the absolute time scale in the signal S1 and the overall sampling counter are utilized to realize the secondary synchronous sampling on the process variable among the data communication equipment; (4) the acquisition control equipment (DAE) for acquiring and outputting the process variable detects the synchronous signal and order data after receiving the signal S2; (5) the acquisition control equipment dynamically adjusts the start time of the S2 signal according to Delta ti and realizes the long-distance multinode synchronous sampling.

A method of data encoding is disclosed. An encoder receives a set of information bits and performs a lifted LDPC encoding operation on the information bits to produce a codeword. The encoder then punctures all lifted bits of the codeword that correspond to one or more punctured base bits of a base LDPC code used for the LDPC encoding operation. The base LDPC code has no multiple edges, and the one or more punctured base bits are those that correspond with one or more punctured base nodes, respectively, of the base LDPC code. For some embodiments, the one or more punctured base nodes correspond to one or more degree 2 variable nodes.

The invention discloses a mobile communication service user experience evaluation method based on an artificial neural network. Firstly, a mobile communication service is taken as an object, the acceptable performance, the retentivity and the integrality of the service are started, the basic characteristic of the terminal user experience quality is collected, and then the basic characteristic is associated to the quality indicator of the corresponding end-to-end mobile communication service, thereby forming an evaluation system of layered terminal user experience quality; secondly, through a plurality of dial tests, the corresponding mobile communication service quality indicator data is collected from the network management aspect; thirdly, the mobile communication service quality indicator data collected from the network management aspect is used as the input of the artificial neural network, and the basic characteristic evaluation of the corresponding terminal user experience quality is used as the output, thereby finishing the modeling for using the artificial neural network; fourthly, the network mobile communication service quality indicator of the user calling service collected from the network management aspect is used as the input of a built model, thus the forecasting and the evaluation of the true experience of the user to the mobile communication service can be obtained.

An apparatus and method for decoding a Low Density Parity Check (LDPC) code having a maximum error correction capability and an error detection capability. In the apparatus, a decoder receives a signal, and decodes the received signal according to a second parity check matrix having parity check expressions obtained by selecting a predetermined number of dependent parity check expressions among dependent parity check expressions generated by combining, with a predetermined scheme, parity check expressions representing rows of a first parity check matrix obtained by encoding information data into a block code having an optimum minimum distance considering a predetermined coding rate, and adding the selected dependent parity check expressions to the parity check expressions of the first parity check matrix. An error detector determines if there is an error in a signal output from the decoder.

The invention belongs to the field of digital communication and digital storage, and particularly relates to a method for multi-code-rate coding based on grouped Markov superposition coding. The method is used for coding a binary information sequence shown in a specification to a codon shown in the specification, wherein the length of the binary information sequence conforms to the formula: K=kBL, the length of the codon is nB(L+m[k]), n is larger than 1, k ranges from 1 to n-1, namely, a code rate set is {1/n, 2/n,..., (n-1)/n}, L is the number of kB sequence groups of the same length, and mk is the memory span of each sub code with the code rate being k/n. The method comprises the following steps that firstly, the information sequence is divided into L groups shown in the specification with the equal length and a sequence shown in the specification with the length of nB is initiated when t is equal to -1, -2,...,-(mk-1), -mk; then, when t is equal to 0, 1, ...L-1, a sequence shown in the specification with the length being kB is divided into B groups and the B groups are sent to conversion defined by an n-dimensional matrix H for coding, so that a coding sequence shown in the specification with the length of nB is obtained and combined with elements shown in the specification, and the tth sub sequence shown in the specification of the codon shown in the specification is calculated. The method for multi-code-rate coding is simple in design, wide in code rate range and superior in performance.

The invention discloses a semantic representation model processing method, a semantic representation model processing device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence. According to the specific implementation scheme, a training corpus set comprising a plurality of training corpuses is collected; and training a semantic representationmodel based on at least one of word segments, sentences and articles by adopting the training corpus set. According to the method and the apparatus, the unsupervised or weakly supervised pre-trainingtasks of three different levels of word segments, sentences and articles are constructed, so that the semantic representation model can learn knowledge of different levels of the word segments, the sentences and the articles from mass data, the capability of general semantic representation is enhanced, and the processing effect of the NLP task is improved.

The invention provides a method for constructing a quasi-cyclic low-density parity-check code (QC-LDPC code), which comprises the following steps: determining a parameter of an encoding matrix; constructing an exponential matrix M (H) of a QC-LDPC code parity check matrix; using a progressive edge growth (PEG) algorithm of the QC-LDPC code parity check matrix to initialize the exponential matrix M (H); endowing the exponential matrix each nonnegative element to form a cyclic shift value with the longest minimum ring and the least ring number; carrying out iteration replacement on the elements in the exponential matrix; and completing the iteration replacement process of the exponential matrix by an exponential matrix iteration assignment algorithm of the QC-LDPC code parity check matrix. The method for constructing the QC-LDPC code of the invention can guarantee the girth and the average minimum girth of the code maximized so as to obtain excellent error code correcting performance.

A communication apparatus, includes unit extracting a reception key feature of a temporal change in received power in each frequency band contained in a received signal, unit storing, in correspondence with one another, a system name of a system which uses the frequency band, event information indicating that a communication apparatus belonging to the system starts or stops transmission, and a transmission key feature of a temporal waveform change shape of a transmission signal from a communication apparatus belonging to the system, a transmission key feature corresponding to the event information, unit determining whether there is any transmission key feature matching the reception key feature, by comparing the reception key feature with the transmission key feature, and unit selecting, when the determination unit determines that there is a matched transmission key feature, the event information and the system name which correspond to the transmission key feature from the storage unit.

The invention discloses a concatenated error correction coding and decoding method based on a polarization code and belongs to the technical field of error correction coding and decoding. The method comprises steps that at the transmitting end, an outer encoder first encodes an information sequence into which the fixed bit is inserted, and obtains the outer code word; the outer code word is used as an input sequence of the polarization code encoder, the polarization code is encoded, and the concatenated code word is obtained; and a receiver decoder uses a combined SCL decoding algorithm to perform decoding. Meanwhile, the invention also discloses a concatenated error correction coding and decoding system based on the polarization code. The frame error rate performance of the coding method provided by the technical scheme of the invention significantly exceeds the performance of the frame error rate of the polarization code under the SCL decoding algorithm or the CRC-aided SCL decoding algorithm; and the outer encoder is extremely low in storage complexity and operation complexity, and the coding is simple. The decoding complexity of the combined SCL decoding algorithm adopted by the invention is not increased compared with the original SCL decoding algorithm or the CRC-aided SCL decoding algorithm, so that the engineering realization is benefited.

The invention discloses a video sharing method and device, and belongs to the technical field of terminals. The method comprises the steps that when a video shooting command is received, in a specified area of a social APP interface, and a shooting panel is displayed so that a shooting device can be started, wherein the shooting panel comprises a framing area and a shooting control area; a long pressing operation corresponding to the shooting control area is conducted, and in the long pressing process, the shooting device is used for conducting shooting; if no clicking operation on any one of the framing area and the shooting control area is detected within a first preset time period after the long pressing operation is completed, the shooting process is completed, and shot videos are shared to a social APP interface specified by a user. By means of the method, video shooting is conducted conveniently, the videos can be shot and shared without multiple button type operations, the video sharing method enables operation to be more convenient and efficient, power consumption of the terminal is reduced, the error correction capability for mistaken operations is improved, and continuous video shooting is guaranteed.

This invention relates to earth digital television broadcast coding method in digital information transmission technique field, which comprises the following steps: diving transmission flow into 752 bit set and adding 261 zeros to get 1013 bit for BCH coding and removing front 261 bit to get BCH(762,752); forming one set of BCH(762,752) with certain number to generate matrix of LDPC(7493,3048),LDPC(7493,4572) or LDPC(7493,6096); deleting output 7493 bit front five correction bit to get final 7488 output bit.

Removing adverse effects in case error detection using error detection codes cannot be performed correctly, and improving the data transmission efficiency. The turbo decoder 6 compares data which is decoded this time with data which is decoded previous time, both of which are decoded in repetitive decoding processing, to obtain discord number information indicative of the number of discordant sign bits, and sends the information to a terminal I/F unit 7 and a packet flow retransmission controlling unit 10. Even though it is determined that error detection result from a CRC recalculating unit 9 shows that there exists no bit error, in case the number of discordant sign bits is larger than a predetermined threshold value, it is determined that there is generated error. And, a packet flow retransmission controlling unit 10 sends a signal requesting retransmission of data. Furthermore, data amount to be retransmitted is varied in accordance with the number of discordant sign bits to improve the data transmission efficiency.

A semiconductor memory device includes a memory cell array in which a plurality of memory cells are arranged. The semiconductor memory device includes an error correcting code (ECC) circuit configured to generate parity data based on main data, write a codeword including the main data and the parity data in the memory cell array, read the codeword from a selected memory cell row to generate syndromes, and correct errors in the read codeword on a per symbol basis based on the syndromes. The main data includes first data of a first memory cell of the selected memory cell row and second data of a second memory cell of the selected memory cell row. The first data and the second data are assigned to one symbol of a plurality of symbols, and the first memory cell and the second memory cell are adjacent to each other in the memory cell array.

The present invention discloses a polar code-repeat code cascade error correction coding method. At a transmit end, the repeat code is used as an outside code, and the polar code is used as an inside code; and a decoder of a receive end uses a correction successive cancellation list (SCL) decoding algorithm to decode. In terms of correction performance, the frame error rate performance of the coding method provided in the present invention is highly stronger than the polar code frame error rate performance of the SCL decoding algorithm or a cyclical redundancy check (CRC)-aided SCL decoding algorithm, and the frame error rate far exceeds the unsurpassed maximum likelihood limit of the SCL decoding algorithm. In terms of engineering implementation, the outside code uses the repeat code, the storage complexity of the outside code coder is very low, the coding is simple, the correction SCL decoding algorithm is adopted for decoding, bit determining of the inside code-polar code and bit determining of the outside code-repeat code are performed jointly, and the decoding complexity is not increased compared with that of the original SCL decoding algorithm and the CRC-aided SCL decoding algorithm, so that the engineering is easy to implement.

The invention discloses a hard real-time cascading multi-node synchronous sampling and data transmission method. The method comprises the following steps: 1) dividing a cascading system into four layers; 2) determining a sampling interval Ts of the cascading system, dividing a physical channel into two logical channels: a real-time logical channel and a non-real-time logical channel; 3) orderly constructing devices in a second layer, a third layer and a fourth layer for synchronous sampling, synchronous transmission and link management in the cascading system; 4) computing a channel retardation of each layer; 5) reconstructing a local synchronous signal of a data combining and transferring device of the third layer, and a local synchronous signal of a data acquisition and control terminal of the fourth layer; 6) controlling data transmission by using sampling period Ts as rhythm. The extra time delay produced by physical link length and software protocol stack among a plurality of multi-stage cascaded acquisition control nodes in different physical distribution is solved, and the hard real-time transfer capability is provided for the key service data, the synchronous precision and real time property are high, the channel construction cost and the construction workload are low, and the good application prospect is realized.

The invention belongs to the technical field of a tracing system of agricultural products and discloses a method for tracing anti-fake markings for agricultural products. The agricultural products are coded based on a supervision model to form agricultural product tracing supervision codes; then the agricultural product tracing supervision codes are compressed and encrypted to generate a one-dimensional tracing code, the result obtained through an encryption algorithm is unique, the fact that one product with one code is achieved, individual marking and tracing can be achieved, and the anti-fake function is high. The one-dimensional tracing code and an encryption initial key of the one-dimensional tracing code are stored in a two-dimensional bar code, because the two-dimensional bar code is provided with good error correcting capacity, when the one-dimensional tracing code is damaged and cannot be recognized, the one-dimensional tracing code can be obtained from the two-dimensional bar code for the tracing, and when a network is unavailable, off-line tracing of basic information can be performed directly through the two-dimensional bar code. Simultaneously, flexible tracing for individuals and batches can be achieved through associating the one-dimensional code with the two-dimensional bar code.

The present invention belongs to the technical field of wireless communication, and specifically discloses an iterative multiple-input multiple-output (MIMO) communication system and method based on IDMA. In the present invention, a non-orthogonal MIMO system is designed based on IDMA (Interleave-Division-Multiplexing-Access); a non-orthogonal access idea of IDMA is applied to the MIMO system; multi-antenna diversity can be realized by distributing different interleavers for different antennas, without the need for designing a complicate spreading code; and all FEC codes can be simply applied to the system. The method and the system of the present invention are applicable to any number of antennas, and can control an error rate of frame synchronization in an acceptable range.

The present invention relates to a method of constructing a low density Parity Check code, a method of decoding the same and a transmission system using the same. The method comprises steps of: constructing a low density Parity Check matrix of the low density Parity Check code by using a fixed pattern; blocking data sent from an information source (101), encoding the data by directly or indirectly using the constructed low density Parity Check matrix to obtain codewords of the low density Parity Check code; and outputting the codewords of the low density Parity Check code. The low density Parity Check code can be encoded by using the Parity Check matrix or a generation matrix. According to the proposed constructing method, a Parity Check matrix of a low density Parity Check code having excellent performance, a method of constructing the low density Parity Check code and a complexity-reduced decoding method can be obtained.

An encoder for encoding source information into an encoded codeword used in a communication channel includes a data input to receive source data, a processor, and a memory to store an encoder program. The encoder program makes the processor to encode the source data into a turbo product coding (TPC) structure, and the TPC structure comprises a data block corresponding to the source data, a first parity block including a first column part, a first corner part and a first bottom part, the first parity block being arranged so as to cover a right end column of the data block, a right bottom corner of the data block and a bottom row of the data block by the first column part, the first corner part and the first bottom part, and a second parity block having a row parity block, a joint parity block and a column parity block.