Search

CN-121984639-A - Encoding caching method and system based on linear error correction code check matrix

CN121984639ACN 121984639 ACN121984639 ACN 121984639ACN-121984639-A

Abstract

The invention provides a coding caching method and a system based on a linear error correction code check matrix, which relate to the technical field of digital communication and comprise the steps of carrying out coding caching of a plurality of files in a coding caching library formed by a server and a plurality of caching nodes, wherein the method comprises the steps of selecting a linear error correction code with the minimum distance meeting a preset condition, and constructing the check matrix of the linear error correction code; constructing a cross resolvable structure, partitioning each file to be cached through the structure, storing all the partitions on a server, caching part of the partitions in corresponding cache nodes, collecting file requests of all users by the server, identifying target data blocks, linearly combining the target data blocks to generate coding signals, sending the coding signals to all the users, and decoding the received coding signals by the users to recover missing sub-blocks. The invention provides more flexible user accessibility selection, ensures the transmission efficiency and reduces the realization complexity of the system.

Inventors

  • WEN JIEJING
  • ZHAO SHITAO

Assignees

  • 山东大学

Dates

Publication Date
20260505
Application Date
20260108

Claims (10)

  1. 1. A code caching method based on a linear error correction code check matrix is characterized in that in a code caching library formed by a server and a plurality of caching nodes, a plurality of files stored on the server are coded and cached, and the method comprises the following steps: According to the preset user accessibility, selecting a linear error correction code with the minimum distance meeting a preset condition, and constructing a check matrix of the linear error correction code; Constructing a cross resolvable structure by utilizing the linear independence of column vectors in a check matrix, partitioning each file on a server through the structure, caching partial blocks into corresponding cache nodes, and enabling each user to have access rights to the partial cache nodes according to the user access degree; the method comprises the steps that a server collects file requests of all users, identifies a union set of missing file sub-blocks of all request files, and linearly combines the missing file sub-blocks by utilizing cross property of cross solvable design to generate coding signals and sends the coding signals to all users; and (3) utilizing the linear independence of the vectors of the check matrix columns and the intersection property of the cross-resolvable design to combine the file blocks on the cache nodes with the access rights, and decoding the received coded signals by a user to recover the missing sub-blocks.
  2. 2. The method for encoding and buffering based on the linear error correction code check matrix as claimed in claim 1, wherein the selecting the linear error correction code with the minimum distance satisfying the preset condition is specifically as follows: For user accessibility Then select Is limited in (2) The upper parameters are Wherein, Is a finite field Is of a size of (a) and (b), For the length of the code word, For the linear subspace dimension, Is the minimum distance between any two codewords.
  3. 3. The code buffer method based on a linear error correction code check matrix as claimed in claim 1, wherein said check matrix is a finite field A kind of electronic device Row of lines Column matrix: Wherein, the For the length of the code word, Rank of , Any of (3) All in columns Linear independence and each Are all non-zero vectors.
  4. 4. The method for encoding and buffering based on the linear error correction code check matrix as claimed in claim 1, wherein the constructing the cross-resolvable structure is based on a set of column vectors of the check matrix to generate a resolvable design, specifically: The parallel classes are divided according to column vectors of the check matrix, each column vector corresponds to one parallel class, each parallel class comprises a plurality of granules, indexes of the file sub-blocks in each granule need to meet the requirement that the inner products of the indexes and the corresponding column vectors are elements on a finite field, and finally, index union sets in all the granules forming the parallel classes can cover the index intervals of the files.
  5. 5. The method for encoding and caching as claimed in claim 1, wherein the step of identifying the union of missing file sub-blocks of all requested files includes comparing the file blocks requested by all users with cached blocks in the cache node having access to each of the file blocks, and identifying file blocks that are commonly missing by all users but that can be complemented by the cached blocks in the cache node.
  6. 6. The method for encoding and buffering based on the linear error correction code check matrix as claimed in claim 1, wherein the generating the encoded signal is specifically to linearly combine the missing file sub-blocks to generate a composite encoded packet.
  7. 7. The method for encoding and caching based on the linear error correction code check matrix as claimed in claim 1, wherein the user decodes the received encoded signal to recover missing sub-blocks, and specifically comprises the step of solving all the missing sub-blocks in the requested file from the received encoded signal according to a linear constraint relation defined by the check matrix by using the sub-blocks stored in the cache nodes which are authorized to access by the user as known information.
  8. 8. A code caching system based on a linear error correction code check matrix, characterized in that a code caching method based on a linear error correction code check matrix according to any one of claims 1-7 is adopted, comprising: The construction module is configured to select a linear error correction code with a minimum distance meeting a preset condition according to a preset user access degree and construct a check matrix of the linear error correction code; The storage module is configured to construct a cross-resolvable structure by utilizing the linear independence of column vectors in the check matrix, divide each file on the server into blocks through the structure, store all the blocks on the server, cache part of the blocks in corresponding cache nodes, and each user has access rights to part of the cache nodes according to the user access degree; The server collects file requests of all users, identifies a union of missing file sub-blocks of all request files, and linearly combines the missing file sub-blocks by utilizing the intersection property of intersection solvable design to generate coding signals and sends the coding signals to all users; and the recovery module is configured to utilize the linear independence of the check matrix array vectors and the intersection property of the cross-resolvable design, combine the file partitioning on the cache nodes with the access rights, and decode the received coded signals by a user to recover the missing sub-blocks.
  9. 9. A non-transitory computer readable storage medium storing computer instructions which, when executed by a processor, implement a linear error correction code check matrix based code caching method as claimed in any one of claims 1 to 7.
  10. 10. An electronic device comprising a processor, a memory and a computer program, wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes a code caching method based on a linear error correction code check matrix according to any one of claims 1-7.

Description

Encoding caching method and system based on linear error correction code check matrix Technical Field The invention relates to the technical field of digital communication, in particular to a code caching method and system based on a linear error correction code check matrix. Background In recent years, with the popularization of audio and video streams, web content browsing and social platforms, communication traffic generated by accessing to internet equipment presents a surge situation, but because network traffic has a high time-varying property, the network traffic often causes a problem of network congestion in peak periods and a problem of network utilization deficiency in valley periods. In order to solve the above-mentioned problems, the caching technology uses the advantages of distributed storage, and by pre-distributing the content in the valley period to realize the "peak clipping and valley filling" of the bandwidth demand, unlike the conventional caching scheme which relies on the prediction of user preference and is limited by the local capacity, the coded caching scheme can realize the multicast transmission of the content of a plurality of different user demands in a coded manner, thereby further reducing the data transmission in the peak period and realizing the global caching benefit. In a multi-channel coded caching scheme, where the system allows a user to obtain data by accessing multiple cache nodes, the prior art generally utilizes a cross-resolvable design (Cross Resovable Design, CRD) in combinatorial mathematics to construct user access rules to the cache nodes, where an access rule based on Hadamard matrices and affine geometry is a classical, powerful and systematic method of constructing cross-resolvable designs, however, current access rules based on Hadamard matrices and affine geometry suffer from the following limitations: 1) Poor flexibility of access While existing solutions often require a fixed and small number of cache nodes accessible to the user (this value is typically 2), in practical configurations, the system architecture requires a more flexible setting of the user's accessibility, i.e. can support any integer number of settings. 2) The system has higher implementation complexity Along with the continuous expansion of the scale of users, the number of file blocks to be processed presents an exponential explosion situation, and the nonlinear growth scale forms a great challenge for the storage index and the calculation scheduling of the bottom layer, which directly leads to the marginal cost sharp increase of hardware resources. Therefore, the prior art has the defects in the aspects of accessibility, transmission efficiency and implementation complexity Disclosure of Invention In order to solve the problems, the invention provides a coding caching method and a system based on a linear error correction code check matrix, which provide more flexible user access degree selection, ensure transmission efficiency and reduce the realization complexity of the system. According to some embodiments, the present invention employs the following technical solutions: a coding caching method based on a linear error correction code check matrix comprises the steps of in a coding caching library formed by a server and a plurality of caching nodes, carrying out coding caching on a plurality of files stored on the server, and comprising the following steps: According to the preset user accessibility, selecting a linear error correction code with the minimum distance meeting a preset condition, and constructing a check matrix of the linear error correction code; Constructing a cross resolvable structure by utilizing the linear independence of column vectors in a check matrix, partitioning each file on a server through the structure, caching partial blocks into corresponding cache nodes, and enabling each user to have access rights to the partial cache nodes according to the user access degree; the method comprises the steps that a server collects file requests of all users, identifies a union set of missing file sub-blocks of all request files, and linearly combines the missing file sub-blocks by utilizing cross property of cross solvable design to generate coding signals and sends the coding signals to all users; and (3) utilizing the linear independence of the vectors of the check matrix columns and the intersection property of the cross-resolvable design to combine the file blocks on the cache nodes with the access rights, and decoding the received coded signals by a user to recover the missing sub-blocks. According to some embodiments, the present invention employs the following technical solutions: a code caching system based on a linear error correction code check matrix, comprising: The construction module is configured to select a linear error correction code with a minimum distance meeting a preset condition according to a preset user access degree and construct a check m