70 results about "Decoding throughput" patented technology

Modern communication systems must cope with varying channel conditions and differing throughput constraints. Polar codes despite being the first error-correcting codes with an explicit construction to achieve the symmetric capacity of memoryless channels are not currently employed against other older coding protocols such as low-density parity check (LDPC) codes as their performance at short/moderate lengths has been inferior and their decoding algorithm is serial leading to low decoding throughput. Accordingly techniques to address these issues are identified and disclosed including decoders that decode constituent codes without recursion and/or recognize classes of constituent directly decodable codes thereby increasing the decoder throughput. Flexible encoders and decoders supporting polar codes of any length up to a design maximum allow adaptive polar code systems responsive to communication link characteristics, performance, etc. whilst maximizing throughput. Further, designers are provided flexibility in implementing either hardware or software implementations.

Embodiments of the present invention provide a decoding method and a decoding apparatus for a polar code. The method comprises: grouping a first polar code having a length of N into s second polar codes, a length of each second polar code being N/s, N and s being integer powers of 2, and N>s; performing path splitting of list decoding on s second polar codes concurrently, and combining split paths of s second polar codes after path splitting, so as to obtain multiple combined paths having a length of N; choosing a first combined path from multiple combined paths having a length of N bits, the first combined path being a path having the greatest path metric value among the multiple combined paths having a length of N bits, or a path passing cyclic redundancy check (CRC) among the multiple combined paths having a length of N bits; and obtaining a decoding result of the first polar code according to the first combined path. In this case, a decoding throughput of the polar code can be improved, and a decoding delay can be decreased.

In recent years, researchers have found that some XOR erasure codes lead to higher performance and better throughput in fault-tolerant distributed data storage applications. However, little consideration has been given to the advantages of parallel processing or hardware implementations taking advantage of the emergence of multi-core processors. An efficient horizontal MDS-like (Maximum Distance Separable) RAID-6 scheme, called EEO, is provided which significantly improves the performance of the decoding procedure in parallel implementations with little storage overhead. EEO is the fastest and most efficient double disk failure recovering algorithm in RAID-6, at the cost of only two more parity symbols. In practice, it is very useful for application where high decoding throughput is desired.

Systems, methods, and apparatus are provided for increasing decoding throughput in an LDPC decoder, such as in a wireless communications receiver or in a data retrieval unit. A checker-board parity check matrix and edge-based LDPC decoder structure are provided in which both vertical and horizontal processors are used simultaneously. Horizontal processors may be grouped into type-A and type-B horizontal processors, and similarly, vertical processors may be grouped into type-A and type-B vertical processors. Type-A processors may be used in different clock cycles than type-B processors to update memory locations in a decoding matrix without causing memory access conflicts.

The invention provides coding and decoding methods, equipment and a system of intra-frame chroma prediction modes. According to the invention, intra-frame prediction modes of an intra-frame prediction mode set which is suitable for the chroma of a current block are divided into a first subgroup and a second subgroup, the amount of the intra-frame predication modes in correspondence with the first subgroup is at least two, and the amount of the intra-frame predication modes in correspondence with the second subgroup is 2<N>, wherein N is a positive integer. Due to the fact that the most required coded bit number is lower than that in the prior art, decoding throughput can be improved.

A symbol mapping method for repetition coding is disclosed. The symbol mapping method comprises performing repetition coding on codeword to output repeated codeword symbols, and mapping the repeated codeword symbols with subcarriers located in different localized resource blocks. According to the embodiments of the present invention, it is possible to obtain maximum reliability in a receiving side by mapping codeword bits with subcarriers to reduce the number of bits having low reliability when a transmitting side uses repetition coding. Also, it is possible to improve decoding throughput and obtain channel diversity.

The invention discloses a continuous elimination list decoding method of a software polarization code. Through establishment of a reference matrix and combination with the position of a current decoding bit, an upper bound and a lower bound of a phase needing to be copied are positioned; and the upper bound and the lower bound are mapped to a memory, only values between the two bounds are copied, and thus the quantity of copy operation is reduced. Compared to a conventional decoding technology, the method has the following advantages: under the condition that the decoding performance is not affected, the same data is prevented from being copied when a path is updated, the path updating time delay is substantially reduced, and the decoding throughput is improved; after the current bit is decoded, the reference matrix is updated; and the decoding method provided by the invention has universality and can be applied to different polarization code lengths, code rates and path numbers.

The invention discloses a low-storage capacity high-speed QC-LDPC (quasi-cyclic low-density parity check) code decoder based on an FPGA (field-programmable gate array) and a decoding method, which are mainly used for solving the problem of low utilization efficiency of memory resources of a node update processing unit and an RAM (random access memory) of the decoder in the prior art. The decoder can simultaneously process two frames of decoding data, the decoder is used for setting an external information value of the first frame of data as all-zero and setting the second frame of data as a channel for receiving likelihood ratio information in the data initialization phase, so that a variable node processing unit and a check node processing unit can completely alternately process the two data frames in parallel in the whole decoding process, effectively shorten the work clock cycle required for processing the two frames of data and enable the decoding throughput to be about two times as that of a traditional design method. According to the decoder disclosed by the invention, a dynamic address access management method is adopted in external information access, and the parallel access of the two frames of decoding data can be realized in the single RAM; and compared with the existing decoder, the utilization efficiency of BRAM (broadcast recognition access method) resources is doubled in comparison with the existing decoders, and the decoder can be used for error correction in information transmission of a physical layer based on LDPC codes.

The invention discloses a parallel multicode-rate convolutional code decoding method and a realization device thereof. The parallel multicode-rate convolutional code decoding method comprises the following steps of: S1. receiving Nin paths of parallel input signals, and shifting the input signals under the effect of a shifting control signal and outputting the input signals, wherein output signals are also Nin paths of parallel signals, and the Nin is positive integers; S2. partially eliminating the phase ambiguity of the output signals; S3. taking the signals processed in the S2 as effectivedata, adding guard intervals for the effective data to assemble into a data frame, and taking the data frame as the input of a parallel convolutional code; S4. carrying out the convolutional decodingon the assembled data frame after being processed in the step S3; and S5. merging multipath outputs of the parallel convolutional decoding into one path and outputting in parallel. The invention provides the parallel multicode-rate convolutional code decoding method aiming at the defect that the throughput rate of the traditional convolutional decoding can not satisfy high-speed communication, improves the decoding throughput rate and the effective payload velocity by adopting a parallel partitioning processing technology, and can satisfy requirements of high-velocity information transmissioncommunication systems, such as satellite communication, and the like.

The invention discloses a multi-element LDPC code high-speed parallel decoder based on a GPU, and a decoding method thereof, and belongs to the technical field of the communication. The decoding method is divided into three stages: an initialization stage, a decoding stage and a judgment stage; once iterative decoding of the decoder is accomplished by only using a kernel function, the synchronization expenditure between the thread blocks and the starting expenditure of the kernel function are greatly reduced, and the access mount of the whole memory is lowered; the use rate of a hardware resource is maximized by designing different parallelism degrees under different systems, thereby improving the decoding throughput to 100M level. Through a memory-access way provided by the invention, theshared memory on a GPU chip is sufficiently utilized, the access to the whole memory is greatly reduced, and the memory-access efficiency and the memory-access bandwidth are improved.

The invention relates to a high-throughput-rate LDPC decoding algorithm and architecture for a 5G terminal, and belongs to the technical field of wireless communication. According to the scheme, the requirement of high throughput rate of data in 5G communication can be met. The architecture mainly comprises a VCN module, a control logic module and a message RAM module. The algorithm comprises thesteps: layering a check matrix according to a base graph and an expansion factor of a 5G QC-LDPC code, forming a layer block structure on each layer, enabling a computing unit of a VCN module to adopta minimum sum decoding algorithm to update nodes, and updating all check nodes in the layer block structures in parallel, thereby increasing the decoding throughput rate; storing the updated posterior information in an interlayer shift register and transmitting the updated posterior information to the next layer to assist subsequent node updating, wherein the symbol information and amplitude information of the node update message are separately processed and stored, so the complexity is reduced. The scheme is suitable for a 5G communication standard, has the characteristics of low complexityand high throughput rate, and is suitable for the field of channel decoding of a wireless communication technology.

Certain aspects of the present disclosure generally relate to methods and apparatus for decoding low density parity check (LDPC) codes, and more particularly to a deeply-pipelined layered LDPC decoder architecture for high decoding throughputs.

The invention relates to an overflow control method for a state metric of a Turbo code decoder. A correction value delta of a branch metric of a current beat is determined according to logarithm metric sizes of all states before n times that are obtained by recursions from a current time k; a branch metric of the current time k is calculated and is corrected; and finally, a state metric of the current time k is calculated according to the corrected branch metric and amplitude limiting is carried out; when the calculated result is larger than a maximum max, the value of the state metric is max; when the calculated result is less than a minimum min, the value of the state metric is min; and the state metric that has been processed by the amplitude limiting is used for a next recursion. According to the invention, it can be ensured that state metrics of forward and backward recursions during a MAP decoding algorithm process of a Turbo code will not overflow; determination operations can be completed in advance; it is allowed to realize anti-overflow control by a pipelining method. Therefore, an influence of an anti-overflow operation on a recursion speed of a state metric can be reduced, so that a recursion speed during hardware realization can be improved and a decoding throughput capacity can be enhanced.

The invention discloses a polarization code belief propagation decoding method and device, and a storage medium. The method comprises the following steps: carrying out belief propagation decoding on information received by a decoder; judging whether a decoding result of the belief propagation decoding meets a judgment condition or not, if so, stopping executing the belief propagation decoding, andotherwise, generating an overturning set based on the decoding result; performing flip-based decoding according to the flip set. According to the method disclosed in the invention, the error correction performance of the list decoding method and equipment can be continuously eliminated while the high decoding throughput rate is achieved, and soft information can be iteratively output; thus co-architecture design of joint detection decoding, LDPC and polarization codes becomes possible.

The invention belongs to the field of error control coding in channel coding, and aims to shorten the delay of a critical path of a decoder on the premise of ensuring the decoding performance so as toimprove the decoding throughput rate. Therefore, the technical scheme adopted by the invention is as follows. The invention discloses a forward error correction decoding decoder based on burst errordetection which comprises a syndrome calculation SC module, a key equation solving KES module and a money search and error estimation CSEE module. A syndrome calculated by the SC module is output to the KES module, and an error position polynomial lambda (X) and an error estimation polynomial omega (X) calculated by the KES module are output to the CSEE module. The decoder is mainly applied to decoder design and manufacturing occasions.

The invention relates to an LDPC-BCH decoding method based on a graphics processor and belongs to the communication technology field. The method is characterized by converting check matrixes with different code lengths and code rates into a quasi-cyclic structure; carrying out compression, inter-row interweaving, and inline interweaving on the check matrix of the quasi-cyclic structure; and in iterative decoding, using whether BCH code decoding is correct as an iterative termination condition to eliminate an error code platform and improve the error correction performance of a decoder, whereinthe above step especially includes: carrying out matrix interweaving on the check portion of the outside information of an input codeword variable node; operating resource distribution; updating theposterior logarithmic likelihood value of a variable node and a logarithmic likelihood values transmitted by the check node to the variable node; carrying out hard decision; and when the BCH code decoding is correct or reaches the maximum number of iterations, terminating the iterative decoding, and outputting an information bit and a decoding success identifier. Decoder delay is in the order of milliseconds, and a decoding throughput is in the order of hundreds of megabytes. Error correction performance is comparable to the error correction performance recommended by a second generation digital satellite broadcasting standard.

The invention discloses a BCH (Bose, Ray-Chaudhuri, Hocquenghem) decoding system with pipeline structures, and belongs to the technical field of computer memory error correction. The BCH decoding system comprises a parallel correction factor computing module, a key equation solving-parallel Chien search module, an FIFO (first in, first out) memory module and a BCH decoding controller module. The parallel correction factor computing module is used for carrying out parallel computing according to received data to obtain correction factors; the key equation solving-parallel Chien search module isused for acquiring key equations by means of computing according to the correction factors and finding out solution of the key equations; the FIFO memory module is used for aching data read from NAND(not and) Flash chips, and data in the FIFO memory module can be outputted step by step when the solution of the key equations is computed; the BCH decoding controller module is used for implementingparallel execution on BCH decoding two-stage pipelines. The BCH decoding system has the advantages that hardware resources in different modules in BCH decoders are reused, BCH decoding can be carriedout by the aid of the parallel pipeline structures, accordingly, BCH decoding throughput can be effectively improved, and the hardware expense can be reduced.

The embodiment of the invention discloses a sliding window classification method and device in Turbo decoding. The method includes the steps of firstly, determining the sizes of Turbo decoding blocks; secondly, determining the size of sliding windows of the first classification according to the sizes of the Turbo decoding blocks; thirdly, determining the sizes and the corresponding numbers of the sliding windows of the first classification, sliding windows of the second classification and sliding windows of the third classification according to the size of the sliding windows of the first classification and allowable deviation between lengths of the sliding windows. The sizes and the numbers of the sliding windows can be set according to the lengths of the decoding blocks and the allowable deviation between the lengths of the sliding windows, and the aims of saving hardware resources, shortening decoding time delay and increasing the decoding throughput rate are achieved.

Memory systems may include a memory storage, a pre-processing checksum unit suitable for, during a first decoding iteration, receiving hard read data including channel input (Lch) sign values, and computing a checksum of the Lch sign values as a checksum_pre value, and a low-density parity-check (LDPC) decoder including an Lch memory and a checksum update unit, the LDPC decoder suitable for, during the first decoding iteration, storing the Lch sign values in the Lch memory of the LDPC decoder, receiving, with the checksum update unit, the checksum_pre value, and decoding a codeword in at least a second decoding iteration based at least in part on the checksum_pre value computed and received being a parity check on the hard read performed in the first decoding iteration.

The invention discloses an LDPC (Low Density Parity Check) code dynamic flipping decoding method based on grouping parallel processing. The method detects whether flip sequences of two adjacent iterations in an iterative flip decoding process cause decoding to be in a state loop or not, and after the state loop is detected, a packet bit flip threshold and a flip bit number are dynamically changed,so that the invalid iteration of bit flip decoding is effectively avoided and system performance is improved. Whether a state cycle condition exists in grouped bit flipping decoding or not is judgedthrough simple check formula calculation, that is, when two adjacent check formula sequences are the same, a judgment is made that iterative decoding falls into an error cycle, and bits cannot be flipped effectively. After the cyclic state is judged, the state cycle is effectively broken by adopting a method of dynamically changing the packet bit flipping threshold and the flipping bit number, andperformance improvement is obtained. The improved method is established on the basis of hard decision flipping decoding, and is fast in iterative convergence, high in decoding throughput rate and lowin implementation complexity.

The invention discloses a decoding method, a decoder, equipment and a medium, and the method comprises the steps: grouping an input bit stream, and obtaining input bit groups; inputting the input bitgroups into different Huffman decoders respectively, so that the Huffman decoders decode the input bit groups in parallel to obtain preliminary decoding results corresponding to the input bit groups,wherein the preliminary decoding results comprise code word lengths, invalid flags and decoding characters; determining a length vector based on the codeword lengths and the invalid flags, and determining a target position vector based on the length vector; and determining a target decoding result corresponding to the input bit stream based on the target position vector and the character vector. Therefore, the bit streams subjected to Huffman coding can be decoded in parallel, the decoding time is saved, the decoding efficiency is improved, the decoding throughput rate is also improved, and the equipment performance is improved.