Implementation method of TST digital switching network matrix supporting multicast

Abstract

The invention provides an implementation method of a TST digital switching network matrix supporting multicast, and relates to the technical field of communication processing. According to the method, a T1 matrix and a multicast space division cross selector value are obtained through derivation of a mathematical model, and digital multicast switching network support of a multicast TST switching network is achieved. The method comprises the steps of converting a TST switched network into a mathematical model, judging whether the TST switched network is a multicast network or not, comparing input and output matrixes to determine multicast elements, replacing the multicast elements, sequencing the output matrixes to obtain an S matrix, and deducing a T1 matrix and a space division cross selector value; according to the invention, unicast and multicast TST switching networks are supported, network flow prediction and resource allocation are optimized through a column priority arrangement algorithm and a machine learning algorithm, and the network efficiency and flexibility are improved. In addition, through a network function virtualization technology, a mathematical model is operated on a universal server, dependence of a dedicated chip is reduced, the circuit size is reduced, and the system flexibility is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of communication processing, and particularly to a method for implementing a TST digital switching network matrix supporting multicast. Background Art

[0002] In the application of SDH technology, a TST switching network is usually adopted to implement the switching of service time slots (the network structure is as Figure 2 shown). The TST switching network is a three-level network, which is composed of two levels of T connectors and one level of S connector; the function of the T connector is to complete the switching between different time slots on the same multiplexing line; the function of the S connector is to complete the switching of time slots between different multiplexing lines.

[0003] The first-level T connector completes the switching of the user-sent time slot to the common time slot inside the switching network; the middle S connector is used to complete the switching of the time slot carried inside the switching network from one input multiplexing line to the specified output multiplexing line; the second-level T connector completes the switching of the time slot on the output multiplexing line to the user's receiving time slot.

[0004] The purpose of the TST switching network is to be able to switch the time slot on the input multiplexing line to any time slot on any output multiplexing line. Generally, a dedicated chip is used to implement it, and the TST switching network constructed by using T connectors and S connectors is only used for a switching matrix with a small switching scale; for a switching matrix with a large scale or in the case where the scale of the switching matrix needs to be flexibly added or deleted, using a dedicated chip to build has the disadvantages of a large circuit volume and poor flexibility.

[0005] Therefore, a method for implementing a TST digital switching network matrix supporting multicast is needed to solve the problems that the existing technology needs a dedicated chip to build and has a large circuit volume and poor flexibility. Summary of the Invention

[0006] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide an implementation method of a TST digital switching network matrix supporting multicast with simple operation, so as to solve the problems that the existing technology needs a dedicated chip to build and has a large circuit volume and poor flexibility.

[0007] To achieve the above purpose, the present application proposes an implementation method of a TST digital switching network matrix supporting multicast. By derivation, a T1 matrix and a multicast space-division cross selector value are obtained and used to support the digital multicast switching network for the multicast TST switching network; wherein, the matrix derivation of the multicast TST switching network is carried out through the following steps:

[0008] Step 1: Convert the current TST switching network into a mathematical model; wherein, it includes an input matrix, a T1 matrix, an S matrix, and an output matrix;

[0009] Step 2: Determine whether the TST switching network is a multicast TST switching network; wherein, if there are multicast elements in the output matrix, the TST switching network is a multicast TST switching network, and proceed to Step A3;

[0010] Step A3: Compare the input matrix and the output matrix to determine the extra multicast elements and missing elements in the output matrix;

[0011] Step A4: Replace the extra multicast elements in the output matrix with the missing elements and record the replacement information;

[0012] Step A5: Sort the replaced output matrix through the column-first permutation algorithm, and obtain the first derived S matrix after sorting;

[0013] Step A6: Perform a secondary sorting on the first derived S matrix to obtain the second derived S matrix;

[0014] Step A7: Derive the T1 matrix and the multicast space-division cross selector values through the second derived S matrix.

[0015] As a further solution, derive the derived T1 matrix and the unicast space-division cross selector values, and use them to support the digital unicast switching network for the unicast TST switching network; wherein, the matrix derivation of the unicast TST switching network is performed through the following steps:

[0016] The matrix derivation of the unicast TST switching network is performed through the following steps:

[0017] Step 1: Convert the current TST switching network into a mathematical model; wherein, it includes an input matrix, a T1 matrix, an S matrix, and an output matrix;

[0018] Step 2: Determine whether the TST switching network is a multicast TST switching network; wherein, if there are no multicast elements in the output matrix, the TST switching network is a unicast TST switching network, and proceed to Step B3;

[0019] Step B3: Set the column-first permutation algorithm and derive the S matrix through the output matrix;

[0020] Step B4: Derive the T1 matrix through the S matrix;

[0021] Step B5: Derive the unicast space-division cross selector values according to the S matrix; wherein, the unicast space-division cross selector values are determined according to the row numbers of the S matrix elements.

[0022] As a further solution, the column-first permutation algorithm is performed through the following steps:

[0023] Traverse each column of the output matrix to find duplicate elements and missing elements;

[0024] Replace the duplicate elements with the missing elements according to the set swapping rules;

[0025] Among them, the swapping rules include direct swapping, double swapping, and triple swapping.

[0026] As a further solution, the replacement rules for replacing multicast elements with missing elements include:

[0027] Replacement rule 1: Multicast elements cannot be replaced by their own similar elements;

[0028] Replacement rule 2: On the premise of satisfying replacement rule 1, preferentially use the missing elements of the multicast similar elements for replacement;

[0029] Replacement rule 3: When the same multicast element needs to be replaced multiple times, it can only be replaced by similar elements at most once and cannot be replaced by similar elements multiple times;

[0030] Replacement rule 4: Among the available missing elements, preferentially use the type of element with the largest number of missing elements in the same category for replacement.

[0031] As a further solution, when performing a secondary sorting on the first derived S matrix, the following sorting steps are carried out:

[0032] Traverse the columns of the first derived S matrix to determine whether there are multicast elements in the current column;

[0033] If there are multicast elements, swap the replacement elements of the current multicast element into the current column according to the swapping rules;

[0034] Traverse all the replacement elements of the multicast element and perform the swapping according to the swapping rules;

[0035] Among them, the swapping rule: find the replacement element. If the current replacement element is already in the same column as the multicast element, do not swap; if the current replacement element is in a different column from the multicast element, swap the current replacement element with the similar element in the same column as the multicast element.

[0036] As a further solution, derive the T1 matrix from the second derived S matrix / S matrix: Traverse the second derived S matrix / S matrix column by column and put each element into the corresponding row of the T1 matrix. After completion of filling, the corresponding T1 matrix is obtained.

[0037] As a further solution, derive the unicast space-division cross selector value through the following steps:

[0038] Traverse the S matrix column by column: Start from the first column of the S matrix and process column by column;

[0039] Determine the unicast space-division crossbar selector values: For each column in the S matrix, the row number of each element is the corresponding unicast space-division crossbar selector value;

[0040] Create a unicast space-division crossbar selector value matrix: Create an empty matrix corresponding to the S matrix, and fill the empty matrix with the unicast space-division crossbar selector values corresponding to the elements of the S matrix to obtain the unicast space-division crossbar selector value matrix.

[0041] As a further solution, derive the multicast space-division crossbar selector values through the following steps:

[0042] Traverse the second-derived S matrix column by column: Start from the first column of the second-derived S matrix and process each column one by one;

[0043] Determine the multicast space-division crossbar selector values: For each column in the second-derived S matrix, the row number of each element is the corresponding multicast space-division crossbar selector value; among them, if the element is a replacement element, the row number of the replaced multicast element needs to be used to set the corresponding multicast space-division crossbar selector value;

[0044] Create a multicast space-division crossbar selector value matrix: Create an empty matrix corresponding to the second-derived S matrix, and fill the empty matrix with the multicast space-division crossbar selector values corresponding to the elements of the second-derived S matrix to obtain the multicast space-division crossbar selector value matrix.

[0045] As a further solution, through network function virtualization, run the mathematical model of the TST switching network on a general server; among them, each mathematical model is set in the form of a container, and digital multicast / unicast switching network support is provided through cloud services.

[0046] As a further solution, also predict the network traffic of the TST switching network through machine learning algorithms, and adjust the number of containers, expand / contract the server, and set the priority processing queue according to the network traffic.

[0047] Compared with the related technologies, an implementation method of a TST digital switching network matrix supporting multicast in the present invention obtains the T1 matrix and multicast space-division crossbar selector values through mathematical model derivation, and realizes digital multicast switching network support for the multicast TST switching network. The method includes converting the TST switching network into a mathematical model, determining whether it is a multicast network, comparing the input and output matrices to determine multicast elements, replacing multicast elements, sorting the output matrix to obtain the S matrix, and deriving the T1 matrix and space-division crossbar selector values.

[0048] The present invention supports unicast and multicast TST switching networks, optimizes network traffic prediction and resource allocation through column-priority sorting algorithms and machine learning algorithms, and improves network efficiency and flexibility. In addition, through network function virtualization technology, the mathematical model is run on general-purpose servers, reducing the dependence on dedicated chips, shrinking the circuit volume, and enhancing system flexibility. The present invention is applicable not only to the current embodiments, but also to equivalent structural transformations in other related technical fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0050] To more clearly illustrate the technical solutions in the embodiments of the present application or in related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0051] Figure 1 Schematic diagram of the implementation steps of a multicast-supported TST digital switching network matrix provided by the present invention;

[0052] Figure 2 Schematic diagram of the TST switching network provided by the present invention;

[0053] Figure 3 Schematic diagram of the mathematical model of the unicast TST switching network provided by the present invention;

[0054] Figure 4 Schematic diagram of direct switching provided by the present invention;

[0055] Figure 5 Schematic diagram of two-stage switching provided by the present invention;

[0056] Figure 6 Schematic diagram of three-stage switching provided by the present invention;

[0057] Figure 7 Schematic diagram of the unicast S matrix provided by the present invention;

[0058] Figure 8 Schematic diagram of the unicast T1 matrix provided by the present invention;

[0059] Figure 9 Schematic diagram of the matrix of the unicast space-division selector values provided by the present invention;

[0060] Figure 10 Schematic diagram of the multicast TST switching network provided by the present invention;

[0061] Figure 11 Schematic diagram of the multicast output matrix provided by the present invention Figure 1 ;

[0062] Figure 12 Schematic diagram of the multicast output matrix provided by the present invention Figure 2 ;

[0063] Figure 13 Schematic diagram of the multicast output matrix provided by the present invention Figure 3 ;

[0064] Figure 14 Schematic diagram of the first derived S matrix provided by the present invention Figure 1 ;

[0065] Figure 15 Schematic diagram of the first derived S matrix provided by the present invention Figure 2 ;

[0066] Figure 16 Schematic diagram of the first derived S matrix provided by the present invention Figure 3 ;

[0067] Figure 17 Schematic diagram of the second derived S matrix provided by the present invention;

[0068] Figure 18 Schematic diagram of the multicast T1 matrix provided by the present invention;

[0069] Figure 19 Schematic diagram of the matrix of the multicast space-division selector values provided by the present invention.

[0070] The realization, functional features and advantages of the purpose of this application will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0071] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations.

[0072] Please refer to Figure 1 , an implementation method of a TST digital switching network matrix supporting multicast is provided in the embodiments of this application. By derivation, a derived T1 matrix and multicast space-division cross selector values are obtained and used to support digital multicast switching networks for multicast TST switching networks; among them, the matrix derivation of the multicast TST switching network is performed through the following steps:

[0073] Step 1: Convert the current TST switching network into a mathematical model, including an input matrix, a T1 matrix, an S matrix, and an output matrix.

[0074] Step 2: Determine whether the TST switching network is a multicast TST switching network. If there are multicast elements in the output matrix, the TST switching network is a multicast TST switching network, and proceed to Step A3.

[0075] Step A3: Compare the input matrix and the output matrix to determine the extra multicast elements and missing elements in the output matrix.

[0076] Step A4: Replace the extra multicast elements in the output matrix with the missing elements and record the replacement information.

[0077] Step A5: Sort the replaced output matrix using the column-first permutation algorithm. After sorting, obtain the first derived S matrix.

[0078] Step A6: Perform a secondary sort on the first derived S matrix to obtain the second derived S matrix.

[0079] Step A7: Derive the T1 matrix and the multicast space-division crossbar selector values from the second derived S matrix.

[0080] Derive the derived T1 matrix and the unicast space-division crossbar selector values, which are used to support the digital unicast switching network for the unicast TST switching network. The matrix derivation of the unicast TST switching network is performed through the following steps:

[0081] It should be noted that according to the working principle of the TST switching network, we can convert it into a mathematical model as shown in Figure 3 and Figure 10 where

[0082] Figure 3 is the mathematical model of the unicast TST switching network. The matrix obtained after the first time-slot exchange of the input matrix is called the T1 matrix. The T1 matrix is obtained by transforming the elements of the input matrix within the row. The matrix obtained after the space-division exchange of the T1 matrix is called the S matrix. The S matrix is obtained by exchanging the elements of the T1 matrix in the column. The matrix obtained after the second time-slot exchange of the S matrix is called the output matrix. The output matrix is obtained by transforming the elements of the S matrix within the row.

[0083] Therefore, to support the digital multicast switching network for the multicast TST switching network, it is necessary to reverse-derive the S matrix and the T1 matrix from the output matrix.

[0084] Next, we analyze the characteristics of the S matrix: 1. In each column, elements from the same row in the input matrix cannot appear simultaneously; 2. The number of elements in each column is the same as the number of rows in the input matrix.

[0085] Based on the characteristics of the S matrix, we can implement the conversion from the output matrix to the S matrix by setting a column - first permutation algorithm; the following concepts are agreed upon in this algorithm:

[0086] Logical row number: The row number of the element in the input matrix is called the logical row number.

[0087] Physical row number: The row number of the element in the output matrix is called the physical row number.

[0088] Repeated elements: The repeated elements in each column of the output matrix are the elements with the same logical row number in that column.

[0089] Missing elements: The logical row numbers that are missing in that column.

[0090] Other elements: The elements in that column that are neither repeated elements nor missing elements.

[0091] The column - first permutation algorithm traverses each column of the output matrix to find repeated elements and missing elements; and replaces the repeated elements with the missing elements according to the set exchange rules; among them, the exchange rules include direct exchange, two - time exchange, and three - time exchange; the following takes a 4X4 matrix as an example to illustrate the direct exchange, two - time exchange, and three - time exchange:

[0092] Direct exchange: As Figure 4 shown, if there is a missing element B in the physical row where the repeated element A is located, then a direct exchange is sufficient.

[0093] Two - time exchange: As Figure 5 shown, if there is no missing element B in the physical row where the repeated element A is located, then find another element C in the physical row where A is located; if there is a missing element B in the physical row corresponding to the other element C in the physical row where A is located, then after two exchanges, the replacement is achieved.

[0094] Three - time exchange: As Figure 6 shown, if there is still no missing element B in the physical row of the other element C, then find another element D in the physical row of C; at this time, for the 4 - order case, there must be a B in the physical row of D, and the requirement can be met after three exchanges.

[0095] For an N - order matrix, because N is a finite value, after a finite number of exchanges, columns that meet the conditions can always be permuted; the S matrix obtained through the column - first permutation algorithm is as Figure 7 shown.

[0096] When deriving the T1 matrix from the S matrix: The T1 matrix is derived from the S matrix. Traverse the S matrix column by column and place the elements into the corresponding rows of the T1 matrix. For example, when traversing Figure 3 the first column of the S matrix in

[0097] If the first element is C3, then place this element in the first column of the 3rd row of the T1 matrix;

[0098] If the second element is D2, then place this element in the first column of the 4th row of the T1 matrix;

[0099] If the third element is B1, then place this element in the first column of the 2nd row of the T1 matrix;

[0100] If the fourth element is A4, then place this element in the first column of the 1st row of the T1 matrix;

[0101] Traverse all columns of the S matrix in this way, and the T1 matrix can be obtained.

[0102] Deriving the unicast space-division cross selector value: The unicast space-division cross selector can be derived from the S matrix. When the value of the unicast space-division cross selector is 1, select the 1st row as the input; when the value of the unicast space-division cross selector is 2, select the 2nd row as the input; when the value of the unicast space-division cross selector is 3, select the 3rd row as the input; when the value of the unicast space-division cross selector is 4, select the 4th row as the input. Traverse the S matrix column by column, as in Figure 3 the first column of the S matrix in

[0103] If the first element is C3, then the value of the space-division selector 1 is 3, and select the 3rd row as the input;

[0104] If the second element is D2, then the value of the space-division selector 2 is 4, and select the 4th row as the input;

[0105] If the third element is B1, then the value of the space-division selector 3 is 2, and select the 2nd row as the input;

[0106] If the fourth element is A4, then the value of the space-division selector 4 is 1, and select the 1st row as the input;

[0107] Traverse all columns of the S matrix in this way, and the matrix of the unicast space-division cross selector value can be obtained (as shown in Figure 9 ).

[0108] Therefore, based on the above content, we perform matrix derivation on the unicast TST switching network through the following steps:

[0109] Step 1: Convert the current TST switching network into a mathematical model; among them, it includes an input matrix, a T1 matrix, an S matrix, and an output matrix;

[0110] Step 2: Determine whether the TST switching network is a multicast TST switching network; if there is no multicast element in the output matrix, the TST switching network is a unicast TST switching network, and proceed to step B3;

[0111] Step B3: Set the column-first arrangement algorithm and derive the S matrix through the output matrix;

[0112] Step B4: derive the T1 matrix through the S matrix;

[0113] Step B5: deriving a unicast space division cross selector value according to the S matrix; wherein the unicast space division cross selector value is determined according to the row number of the S matrix element.

[0114] The above analysis is about the unicast TST switching network. The following analysis is about the multicast TST switching network:

[0115] exist Figure 10 In the multicast TST switching network, the C1 element of the output matrix appears 4 times and the A2 element appears 3 times. Therefore, the C1 element and the A2 element are multicast elements. By observation, if multicast elements appear in the output matrix, then some elements in the input matrix must be missing in the output matrix. For example Figure 10 In the matrix, there are 4 extra multicast elements C1 and 2 extra multicast elements A2; correspondingly, the 5 elements A3, B1, C4, D1, and D4 in the input matrix are missing. If the extra multicast elements are replaced with the missing elements, the first derived S matrix can be derived using the column-first permutation algorithm.

[0116] The specific implementation steps are as follows: Find the missing elements in the output matrix; for example Figure 11 The five elements A3, B1, C4, D1, and D4 are missing in the output matrix; replace the redundant multicast elements with the missing elements and keep a record of the replacement.

[0117] The replacement rules are as follows:

[0118] Replacement rule 1: A multicast element cannot be replaced by an element of the same type.

[0119] Replacement rule 2: Under the premise of satisfying "Replacement rule 1", the missing elements of the same type as multicast elements are replaced first.

[0120] Replacement rule 3: When the same multicast element needs to be replaced multiple times, it can only be replaced once by elements of the same type and cannot be replaced multiple times by elements of the same type.

[0121] Replacement rule 4: Among the available missing elements, the elements with the largest number of missing elements in the same class are given priority for replacement.

[0122] Taking an example according to the above replacement principle: If the multicast element A2 is found in the 3rd row and 4th column of the output matrix and needs to be replaced by a missing element. Currently, there are 5 missing elements, namely A3, B1, C4, D1, and D4. According to "Replacement Rule 1", A2 cannot be replaced by the same type of element A3. Therefore, only one element can be selected from B1, C4, D1, and D4 to replace A2. According to "Replacement Rule 2", C4 and C1 are the same type of elements, and C1 is a multicast element. Therefore, C4 is preferentially used to replace A2 in the 3rd row and 4th column and mark it. After replacement, as Figure 12 shown.

[0123] Continue to find that the multicast element A2 in the 4th row and 4th column needs to be replaced by a missing element. Currently, there are 4 missing elements, namely A3, B1, D1, and D4. According to "Replacement Rule 1", A2 cannot be replaced by the same type of element A3. Therefore, only one element can be selected from B1, D1, and D4 to replace A2. According to "Replacement Rule 2", B1, D1, and D4 do not meet the requirements. According to "Replacement Rule 3", A2 has been replaced multiple times and has been replaced by C4 once. Therefore, C-type elements cannot be used to replace it again. At this time, there are no C-type elements among the elements available for replacement, which meets the conditions. According to "Replacement Rule 4", among the missing elements available for replacement, there is one B-type element (B1) and two D-type elements (D1, D4). Therefore, D-type elements are preferentially used to replace A2 in the 4th row and 4th column and mark it. After replacement, as Figure 13 shown.

[0124] According to the above rules, the output matrix after replacing the multicast element C1 is as Figure 14 shown.

[0125] The column-first permutation algorithm is used to sort the replaced output matrix; the first derived S matrix after sorting is as Figure 15 shown.

[0126] The first derived S matrix is sorted again. The sorting steps are as follows:

[0127] Traverse the columns to determine whether there are multicast elements in the column.

[0128] If there are, swap the replacement element of the multicast element to this column. The swap rule is as follows:

[0129] Swap rule: Search for the replacement element. If the replacement element is already in the same column as the multicast element, do not swap. If the replacement element is in a different column from the multicast element, swap the replacement element with the element of the same type (the element of the same type as the replacement element) in the same column as the multicast element.

[0130] After traversing all the replacement elements of the multicast element, swap them according to the swap rule.

[0131] TakingFigure 15 An example of sorting the first derived S matrix shown is as follows:

[0132] a) No multicast element is found by traversing the first column;

[0133] b) No multicast element is found by traversing the second column;

[0134] c) When traversing the third column, the first row is a multicast element C1. At this time, the replacement element of C1 needs to be found. The steps are as follows:

[0135] i. It is found that A3 in the second row and third column is the replacement element of the multicast element C1. This replacement element is in the same column as C1 and no exchange is performed;

[0136] ii. It is found that B1 in the third column and first row is the replacement element of the multicast element C1. This replacement element is not in the same column as C1 and an exchange is required; According to the exchange rule, B1 is exchanged with B4.

[0137] The first derived S matrix after exchange is as Figure 16 shown.

[0138] iii. Continuing, it is found that D4 in the fourth row and second column is the replacement element of the multicast element C1. This replacement element is not in the same column as C1 and an exchange is required; According to the exchange rule, D4 is exchanged with D3. The first derived S matrix after exchange is as Figure 17 shown.

[0139] When traversing the fourth column, the first row is a multicast element A2. At this time, the replacement element of A2 needs to be found. The steps are as follows:

[0140] i. It is found that C4 in the second row and fourth column is the replacement element of the multicast element A2. This element is in the same column as A2 and no exchange is needed;

[0141] ii. Continuing, it is found that D1 in the fourth row and fourth column is the replacement element of the multicast element A2. This element is in the same column as A2 and no exchange is needed;

[0142] After the second sorting of the first derived S matrix, the second derived S matrix obtained is as Figure 17 shown.

[0143] Deriving the T1 matrix from the second derived S matrix

[0144] The T1 matrix can be derived from the S matrix. Traverse the S matrix column by column and place the elements in the corresponding rows of the T1 matrix. For example, when traversing Figure 8 in the first column of the S matrix:

[0145] The first element is C3, then this element is placed in the first column of the third row of the T1 matrix;

[0146] If the second element is D2, then place this element in the position of the first column of the fourth row of the T1 matrix;

[0147] If the third element is B4, then place this element in the position of the first column of the second row of the T1 matrix;

[0148] If the fourth element is A4, then place this element in the position of the first column of the first row of the T1 matrix;

[0149] In this way, traverse all columns of the second derived S matrix in sequence, and the T1 matrix as shown in Figure 18 can be obtained.

[0150] Derive the multicast space-division cross selector value: The multicast space-division cross selector value can be derived from the second derived S matrix. When the value of the space-division selector is 1, select the first row as the input; when the value of the space-division selector is 2, select the second row as the input; when the value of the space-division selector is 3, select the third row as the input; when the value of the space-division selector is 4, select the fourth row as the input;

[0151] When traversing the S matrix column by column, for unicast, the value of the space-division selector is derived according to the row number of the element; when multicast is supported, if the element is a replacement element, the row number of the replaced multicast element needs to be used to derive the value of the space-division selector.

[0152] As in Figure 17 when traversing the third column of the S matrix:

[0153] The element in the first row is C1, and C1 is not a replacement element, so the value of space-division selector 1 is the row number 3 of C1, and the third row is selected as the input;

[0154] The element in the second row is A3, and A3 is a replacement element, so the value of space-division selector 2 is the row number 3 of the replaced multicast element C1, and the third row is selected as the input;

[0155] The element in the third row is D4, and D4 is a replacement element, so the value of space-division selector 3 is the row number 3 of the replaced multicast element C1, and the third row is selected as the input;

[0156] The element in the fourth row is B1, and B1 is a replacement element, so the value of space-division selector 4 is the row number 3 of the replaced multicast element C1, and the third row is selected as the input;

[0157] In this way, traverse all columns of the S matrix in sequence, and the matrix of space-division selector values as shown in Figure 19 can be obtained.

[0158] After deriving the above mathematical model, we also run the mathematical model of the TST switching network on a general server through network function virtualization. Among them, each mathematical model is set in the form of a container, and digital multicast / unicast switching network support is provided through cloud services. In this way, we can get rid of the current dedicated chip, reduce the circuit volume, and enhance the flexibility.

[0159] In addition, we also predict the network traffic of the TST switching network through machine learning algorithms, and adjust the number of containers, expand / contract the server, and set the priority processing queue according to the network traffic. In this way, we can dynamically coordinate the processing capabilities of the cloud, making the TST switching network work more smoothly.

[0160] The above are only some embodiments of this application, and do not limit the patent scope of this application. Any equivalent structural transformation made under the technical concept of this application by using the content of the specification and drawings of this application, or any direct / indirect application in other related technical fields, is included in the patent protection scope of this application.

Claims

1. A method for implementing a TST digital switching network matrix supporting multicast, characterized in that: The derived T1 matrix and the multicast space division cross selector value are obtained by derivation and used to support the multicast TST switching network for digital multicast switching network; wherein the matrix of the multicast TST switching network is derived by the following steps: Step 1: Convert the current TST switching network into a mathematical model, which includes an input matrix, a T1 matrix, an S matrix, and an output matrix; Step 2: Determine whether the TST switching network is a multicast TST switching network; if there is a multicast element in the output matrix, the TST switching network is a multicast TST switching network, and proceed to step A3; Step A3: By comparing the input matrix with the output matrix, determining the extra multicast elements and the missing elements in the output matrix; Step A4: replace the extra multicast elements in the output matrix with the missing elements, and record the replacement information; Step A5: sorting the replaced output matrix by a column-first sorting algorithm, and obtaining a first derived S matrix after the sorting is completed; Step A6: performing secondary sorting on the first derived S matrix to obtain a second derived S matrix; Step A7: derive the T1 matrix and the multicast space division cross selector value through the second derived S matrix.

2. The method for implementing a TST digital switching network matrix supporting multicast according to claim 1, characterized in that: The derived T1 matrix and the unicast space division cross selector value are obtained by derivation and used to support the digital unicast switching network for the unicast TST switching network; wherein the matrix of the unicast TST switching network is derived by the following steps: The matrix of the unicast TST switching network is derived by the following steps: Step 1: Convert the current TST switching network into a mathematical model, which includes an input matrix, a T1 matrix, an S matrix, and an output matrix; Step 2: Determine whether the TST switching network is a multicast TST switching network; if there is no multicast element in the output matrix, the TST switching network is a unicast TST switching network, and proceed to step B3; Step B3: Set the column-first arrangement algorithm and derive the S matrix through the output matrix; Step B4: derive the T1 matrix through the S matrix; Step B5: deriving a unicast space division cross selector value according to the S matrix; wherein the unicast space division cross selector value is determined according to the row number of the S matrix element.

3. The method for implementing a TST digital switching network matrix supporting multicast according to claim 1, characterized in that: The column-first sorting algorithm proceeds through the following steps: Traverse each column of the output matrix and find duplicate elements and missing elements; Replace the repeated elements with the missing elements according to the set exchange rules; The exchange rules include direct exchange, double exchange and triple exchange.

4. The method for implementing a TST digital switching network matrix supporting multicast according to claim 1, characterized in that: The replacement rules for multicast elements by missing elements include: Replacement rule 1: A multicast element cannot be replaced by its own element of the same type; Replacement rule 2: Under the premise of satisfying replacement rule 1, the missing elements of the same type as multicast elements are replaced first; Replacement rule 3: When a multicast element needs to be replaced multiple times, it can only be replaced once by the same element at most and cannot be replaced multiple times by the same element. Replacement rule 4: Among the available missing elements, the elements with the largest number of missing elements of the same type are given priority for replacement.

5. The method for implementing a TST digital switching network matrix supporting multicast according to claim 1, characterized in that: When the first derived S matrix is ​​sorted again, the following sorting steps are performed: Traverse the columns of the first derived S matrix to determine whether there is a multicast element in the current column; If a multicast element exists, the replacement element of the current multicast element is exchanged into the current column according to the exchange rule; Traverse all replacement elements of the multicast element and exchange them according to the exchange rule; The exchange rule is as follows: search for a replacement element. If the current replacement element is already in the same column as the multicast element, do not exchange it. If the current replacement element and the multicast element are in different columns, exchange the current replacement element with the same type of element in the same column as the multicast element.

6. The method for implementing a TST digital switching network matrix supporting multicast according to claim 2, characterized in that: Derive the T1 matrix through the second derived S matrix / S matrix: traverse the second derived S matrix / S matrix column by column, and put each element into the corresponding row of the T1 matrix, and obtain the corresponding T1 matrix after filling.

7. The method for implementing a TST digital switching network matrix supporting multicast according to claim 2, characterized in that: The unicast space division crossbar selector value is derived by the following steps: Traverse the S matrix column by column: start from the first column of the S matrix and process column by column; Determine the value of each unicast space division cross selector: for each column in the S matrix, the row number of each element is the value of the corresponding unicast space division cross selector; Create a unicast space division cross selector value matrix: create an empty matrix corresponding to the S matrix, and fill the value of the unicast space division cross selector corresponding to each element of the S matrix into the empty matrix to obtain the unicast space division cross selector value matrix.

8. The method for implementing a TST digital switching network matrix supporting multicast according to claim 1, characterized in that: The multicast space division cross-link selector value is derived through the following steps: Traverse the second derived S matrix by column: start from the first column of the second derived S matrix and process column by column; Determine the value of each multicast space division cross selector: for each column in the second derived S matrix, the row number of each element is the value of the corresponding multicast space division cross selector; wherein, if the element is a replacement element, the row number of the replaced multicast element needs to be used to set the value of the corresponding multicast space division cross selector; Create a multicast space division cross selector value matrix: create an empty matrix corresponding to the second derived S matrix, and fill the value of the multicast space division cross selector corresponding to each element of the second derived S matrix into the empty matrix to obtain the multicast space division cross selector value matrix.

9. The method for implementing a TST digital switching network matrix supporting multicast according to claim 1, characterized in that: Through network function virtualization, the mathematical model of the TST switching network is run on a general-purpose server; wherein each mathematical model is set up in the form of a container, and digital multicast / unicast switching network support is provided through cloud services.

10. The method for implementing a TST digital switching network matrix supporting multicast according to claim 9, characterized in that: The network traffic of the TST switching network is also predicted through machine learning algorithms, and the number of containers, server expansion / reduction adjustments and priority processing queue settings are adjusted based on the network traffic.

Images