Multi-subsystem weighted interference elimination coding cache method and system in multi-relay interference network

Abstract

The present invention relates to a method and system for weighted interference elimination coding cache of multiple subsystems in a multi-relay interference network, and the method comprises: step one, the source node executes file partitioning; step two, the relay node and the user cache files; step three, the source node file transmission step; step four, the relay node transmission step; step five, the user node receives file information step. The present invention can further improve the total degree of freedom of the system on the basis of interference elimination by sensing system parameters and dynamically adjusting the partition weights, taking the effective information of a single interference-free transmission and the total degree of freedom of the system as evaluation indicators; at the same time, by using the coding cache strategy, the physical layer file partitioning and the network layer coding delivery are jointly designed, which greatly reduces the link load and transmission complexity, and makes full use of the link resources in the idle phase of the network.

Description

Technical Field

[0001] The invention belongs to the field of digital communication, and in particular relates to a multi-subsystem weighted interference elimination coding cache method and system in a multi-relay interference network. Background Art

[0002] With the explosive growth of data traffic, heavy pressure is placed on the existing network infrastructure. Since Maddah-Ali and Niesen proposed a new centralized coding cache technology from the perspective of information theory in 2014, designing coding cache strategies and improving the lower bound of link rate in different system environments have become hot research topics.

[0003] The coding cache is divided into the placement phase and the delivery phase. The coding multicast gain is maximized and the link load is reduced by jointly designing the physical layer file partitioning placement and the network layer coding delivery. The transmission process is divided into the placement phase and the delivery phase. Cache placement usually occurs during the network idle period, and the user stores the divided sub-files in the cache. Coding delivery usually occurs during the network peak period, and the server transmits the sub-files after network encoding according to the user request. Compared with the traditional caching strategy, the coding cache technology can not only obtain local cache gain, but also obtain global cache gain among multiple users, so it can greatly reduce the network transmission load.

[0004] In a multi-relay interference network scenario, the decentralized coding cache technology proposed by Maddah-Ali and Niesen is used in multiple relay nodes, which will cause the user terminal to receive unrequested subfiles, causing interference, while the traditional cache strategy does not achieve the global cache gain between users. The present invention achieves interference alignment at the transmitter and interference elimination at the receiver by designing a weighted transmission strategy, thereby improving the total degree of freedom of the system. Some of the prior arts involved are as follows:

[0005] 1. Layered coding cache

[0006] The system is divided into two subsystems. System 1 uses the cache capacity of relay nodes and users to obtain coded multicast gain between adjacent layers. System 2 ignores the cache capacity of relay nodes and directly obtains coded multicast gain between source nodes and user nodes. The link rate is minimized by optimizing the system partition parameters.

[0007] In the placement phase, relay nodes and users adopt decentralized cache placement. In system one, relay nodes store part of the sub-files according to file division and do not cache the sub-files to be transmitted in system two.

[0008] In the delivery phase, based on the decentralized delivery strategy, the requested sub-files are transmitted in System 1 and System 2 respectively. For details, see "Karamchandani N, Niesen U, Maddah-Ali MA, et al. Hierarchical Coded Caching[J]. IEEE Transactions on Information Theory, 2016, 62(6): 3212-3229."

[0009] 2. Decentralized Code Cache

[0010] During the placement phase, each user caches a portion of the sub-files independently.

[0011] In the delivery phase, users randomly request files, and the server sends the encoded subfiles to user subsets of different sizes, combining user cache to eliminate interference and obtain coded multicast gain. For details, see "Maddah-Ali MA, Niesen U. Decentralized Coded Caching Attains Order-Optimal Memory-Rate Tradeoff[J]. IEEE / ACM Transactions on Networking, 2015, 23(4): 1029-1040."

[0012] 3. Weighted Interference Elimination

[0013] In the scenario where multiple servers deliver requested subfiles that interfere with each other, based on decentralized coding cache, in the placement phase, each user randomly caches a portion of subfiles according to the number of files and user cache capacity, using decentralized placement. In the delivery phase, by designing the delivery weights of each subfile, interference alignment is achieved at the sender, and interference elimination gain is obtained at the receiver. For details, see "Naderializadeh N, Maddah-Ali MA, Avestimehr A S. Fundamental Limits of Cache-Aided Interference Management[J]. IEEE Transactions on Information Theory, 2017."

[0014] 4. Channel coding and modulation and demodulation methods

[0015] Channel coding, also known as error control coding, refers to artificially adding some protection components (supervisory code elements) to the information sequence to be transmitted, so that automatic error correction can be performed when decoding at the receiving end, thereby enhancing the signal's anti-interference ability. The decoding process usually occurs at the receiving end and is the inverse process of channel coding, restoring the encoded information to the original signal.

[0016] Modulation refers to converting various digital baseband signals into digital modulated signals (modulated signals or frequency band signals) suitable for channel transmission. Modulation is to use baseband signals to control the changes of one or several parameters of the carrier signal, and to load information on it to form a modulated signal for transmission, while demodulation is the reverse process of modulation, and the original baseband signal is restored from the parameter changes of the modulated signal through specific methods. Demodulation refers to restoring the received digital frequency band signal to a digital baseband signal at the receiving end. Modulation is the process of moving the spectrum of the baseband signal to the channel passband or a certain frequency band therein, while demodulation is the reverse process of restoring the frequency band signal from the channel to the baseband signal. For details, see "Fan Changxin, Cao Lina. Principles of Communication. 7th Edition [M]. National Defense Industry Press, 2012." Summary of the invention

[0017] Based on the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide a multi-subsystem weighted interference elimination coding caching method and system in a multi-relay interference network. Compared with traditional orthogonal unicast and single weighted interference elimination strategies, additional system total degrees of freedom can be obtained, thereby increasing the effective information of a single transmission.

[0018] In order to achieve the above-mentioned object of the invention, the present invention adopts the following technical solutions:

[0019] A multi-subsystem weighted interference elimination coding cache method in a multi-relay interference network comprises the following steps:

[0020] Step 1.1, taking the user node cache capacity, the relay node cache capacity, the number of user nodes and the number of relay nodes as input parameters, the source node executes the file partitioning method to obtain the divided sub-files;

[0021] Step 1.2, taking the divided sub-files as input parameters, the relay node and the user end divide the sub-files and store them in the corresponding cache;

[0022] Step 1.3, using the divided sub-files and the request index as input parameters, the source node performs a joint design weighted interference cancellation and orthogonal unicast method to obtain an XOR combination of the requested sub-files; wherein the request index represents the file sequence number requested by the user node;

[0023] Step 1.4, XOR combination of the requested sub-files, sequentially executing channel coding and modulation and demodulation methods, and then broadcasting them to all relay nodes in free space through the transmitting antenna;

[0024] Step 1.5, using the divided sub-files and the request index as input parameters, the relay node executes the collaborative decentralized encoding transmission method to obtain the XOR combination of the remaining requested sub-files;

[0025] Step 1.6, XOR combination of the remaining request sub-files, sequentially executing channel coding and modulation and demodulation methods, and then broadcasting them to all relay nodes in free space through the transmitting antenna;

[0026] Step 1.7, for the sub-files received in step 1.4 and step 1.6, the channel coding and modulation and demodulation methods are sequentially performed to obtain the request sub-file after XOR combination;

[0027] Step 1.8, the user node XORs the requested sub-files combined with the sub-files cached by itself to obtain all sub-files of the requested file, and combines them to obtain the requested file.

[0028] As a preferred solution, in step 1.1, the file division method includes the following steps:

[0029] Step 2.1, the system includes a source node and K1 relay nodes, denoted as S i , i is a positive integer, 0≤i≤K1, K1 is a positive integer; K2 user nodes, denoted as U i , i is a positive integer, 0≤i≤K2, K2 is a positive integer;

[0030] Assume that there are N files in the file library, N is a positive integer, denoted by W i , i is a positive integer, 0≤i≤N, each file is B bits, B is a real number greater than or equal to 0, the source node can access all files, each relay node has the same cache capacity, recorded as M1B bits, satisfying M1≤N, each user has the same cache capacity, recorded as M2B bits, satisfying M2≤M1, M1, M2 are real numbers greater than or equal to 0;

[0031] definition:

[0032]

[0033] Among them, α represents the file division ratio, and β represents the user cache capacity division ratio. k , k∈[αN], divide the files into disjoint subfiles of equal size, denoted as:

[0034]

[0035] Among them, W k,T,R Indicates that the subfile is only cached in S i and Uj In the equation, i∈T, j∈R;

[0036] Therefore, S i and U j The cached contents are:

[0037] Z 1,i = {W k,T,R},i∈T,k∈[N] (3)

[0038] Z 2,j = {W k,T,R},j∈R,k∈[N] (4)

[0039] Among them, Z 1,i and Z 2,j Represent the contents of relay node and user cache respectively;

[0040] Step 2.2, for file W k ,k∈[(1-α)N], divided into disjoint subfiles, U j ,j∈[K2]fromW k Random Cache sub-files; user nodes cache the sub-files of all files in turn, and the caches of each user node do not affect each other.

[0041] As a preferred solution, in step 1.3, the weighted interference elimination and orthogonal unicast method are jointly designed, including the following steps:

[0042] Step 3.1: When t1+t2≤K2, for the sub-files that need to be delivered:

[0043]

[0044] in, Indicates that the user U j The requested file is cached in the relay node set T and the user node set R. Indicates that element j does not belong to set R, j∈[K2] indicates that element j belongs to K2, [K2]={1,2,...,K2};

[0045] First, the sub-file to be transferred Divide into several subsets of size t1+t2, and divide each subfile into smaller files; for the requested subfile in, indicates that the set T is a subset of [K1], [K2]\j indicates that the set [K2] does not contain element j. Divide into A smaller file, recorded as:

[0046]

[0047] Among them, for the set R, ∏ R represents the full permutation of set R;

[0048] For t∈{1,2,...,|R}, |R| represents the number of elements in set R, ∏ R,t Represents all permutations and combinations of t subsets of the set R, denoted as:

[0049]

[0050] By dividing the file in formula (6), the sub-file coefficients are designed to eliminate interference among the users in set π′. Since the sub-file has been cached in the user set π, the user nodes in set π can use the cached files to eliminate interference and transmit t1+t2 sub-files to t1+t2 users. This communication process is interference-free.

[0051] Based on the above file division, all files to be transferred are divided into

[0052]

[0053] in, represents the circular arrangement of elements in set R. There are a total of (|R|-1)! circular arrangements in set R. For set R, π[a:b], b≥a is defined as

[0054]

[0055] in, For collections

[0056] For the sub-files divided by formula (8), the transmission coefficient of the relay node for each sub-file is designed to achieve interference-free transmission, S i The files sent are recorded as:

[0057]

[0058] Transmitted through different channels, U j The files received are:

[0059]

[0060] The divided files are shown in formula (6). For user node U j , j∈π[l+1:l+t2], Because Z 2,j The file is already cached in U j Available through Z2,j Perform interference cancellation; for user U j , j ∈ π[l + t2 + 1:l + t1 + t2 - 1], Z 2,j has not cached this file and needs to control the transmission coefficient for interference cancellation, that is:

[0061]

[0062] Step 3.2: When t1 + t2 ≤ K2, divide the file, the relay node cache capacity, and the user cache capacity so that αN, M1 and βM2 satisfy:

[0063]

[0064] where M1″ = γ2M1;

[0065] For γ1αN, use γ2M1 and γ3βM2 for weighted interference cancellation transmission;

[0066] For (1 - γ1)αN, use (1 - γ2)M1 and (1 - γ3)βM2 for traditional orthogonal unicast transmission.

[0067] As a preferred solution, in step 1.5, the cooperative decentralized coding transmission method includes the following steps:

[0068] Step 4.1: The relay node sends the encoded file

[0069]

[0070] where W k,R represents that this sub - file of W k is only cached in U j , j ∈ R, and R\{j} means that the set R does not contain j;

[0071] Step 4.2: When K1 ≥ K2, different relay nodes perform s = K2, K2 - 1,..., 1 coding transmissions. In the case of K1 ≥ K2, the maximum number of transmissions is K2;

[0072] When K1 < K2, all relay nodes cannot perform coding transmissions for all s values at one time. Therefore, when a relay node finishes transmitting, it jumps forward by K1 units until all s - value coding transmissions are completed.

[0073] The present invention also provides a multi - subsystem weighted interference cancellation coding cache system in a multi - relay interference network, applying the multi - subsystem weighted interference cancellation coding cache method described in any one of the above solutions. The multi - subsystem weighted interference cancellation coding cache system includes:

[0074] A file partitioning module is used to control the source node to execute the file partitioning method according to the user node cache capacity, relay node cache capacity, user node number and relay node number as input parameters to obtain the divided sub-files;

[0075] A storage module, used to take the divided sub-files as input parameters, control the relay nodes and the user end to store the sub-files in the corresponding cache according to the sub-file division;

[0076] A jointly designed weighted interference elimination and orthogonal unicast module is used to take the divided sub-files and the request index as input parameters, control the source node to execute the jointly designed weighted interference elimination and orthogonal unicast method, and obtain an XOR combination of the requested sub-files; wherein the request index represents the file sequence number requested by the user node;

[0077] The channel coding and modulation and demodulation module is used to perform the XOR combination of the request sub-files, execute the channel coding and modulation and demodulation methods in sequence, and then broadcast them to all relay nodes in free space via the transmitting antenna;

[0078] The collaborative decentralized coding transmission module is used to take the divided sub-files and the request index as input parameters, control the relay node to execute the collaborative decentralized coding transmission method, and obtain the XOR combination of the remaining request sub-files;

[0079] The channel coding and modulation and demodulation module is further used to perform XOR combination of the remaining request sub-files, execute the channel coding and modulation and demodulation methods in sequence, and then broadcast them to all relay nodes in free space via the transmitting antenna;

[0080] The channel coding and modulation and demodulation module is also used to sequentially execute the channel coding and modulation and demodulation methods on the subfiles received by the relay node to obtain the request subfiles after XOR combination;

[0081] The file processing integration module is used to control the user node to XOR the request sub-file combined with the sub-file cached by itself, obtain all the sub-files of the request file through XOR processing, and combine them to obtain the request file.

[0082] Compared with the prior art, the present invention has the following technical effects:

[0083] The present invention applies the coding cache to the multi-relay interference network, and proposes a multi-subsystem weighted interference elimination strategy. By weighting the sub-files to be transmitted at the transmitting end, the receiving end achieves interference alignment through different channel gains. In the case where the number of valid sub-files in a single interference-free transmission is greater than the number of users, compared to ignoring some relay nodes and user cache capacity, the present invention further improves the overall freedom of the system by jointly designing weighted interference elimination and orthogonal unicast transmission. In addition, in system two, based on the decentralized delivery strategy, each user decentralized collaborative transmission of the requested sub-files reduces the number of required transmissions, thereby reducing the transmission load while improving the overall freedom of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 is a system model diagram of an embodiment of the present invention;

[0085] Figure 2 A system partition model diagram of an embodiment of the present invention;

[0086] Figure 3 A transmission model diagram of a system when the number of files transmitted at a single time is greater than the number of users according to an embodiment of the present invention;

[0087] Figure 4 This is a transmission model diagram of system 2 of an embodiment of the present invention;

[0088] Figure 5 is a flow chart of a method according to an embodiment of the present invention;

[0089] Figure 6 A comparison diagram of the total degree of freedom of the system when the number of files transmitted at a single time is greater than the number of users according to an embodiment of the present invention;

[0090] Figure 7 This is a comparison diagram of the total degree of freedom of the system when the number of files transmitted at a single time is less than the number of users according to an embodiment of the present invention. DETAILED DESCRIPTION

[0091] In order to more clearly illustrate the embodiments of the present invention, specific implementations of the present invention will be described below with reference to the accompanying drawings.

[0092] like Figure 5 As shown, the multi-subsystem weighted interference elimination coding cache method in a multi-relay interference network of this embodiment is completed by the following steps:

[0093] Step 1.1, taking the user cache capacity, relay node cache capacity, number of users and number of relay nodes as input parameters, the source node executes the file partitioning method to obtain the partitioned sub-files;

[0094] Step 1.2, taking the divided sub-files obtained in step 1.1 as input parameters, the relay node and the user (same user node) divide the sub-files according to the sub-files and store them in the corresponding cache;

[0095] Step 1.3, source node file transfer step, is completed in the following sub-steps:

[0096] Step 1.3.1, taking the divided sub-files obtained in step 1.1 and the user request index as input parameters, where the request index represents the file number requested by the user, the source node performs a joint design of weighted interference elimination and orthogonal unicast method to obtain an XOR combination of the requested sub-files.

[0097] Step 1.3.2, the XOR combination of the request sub-files obtained in step 1.3.1 is sequentially executed with channel coding and modulation and demodulation methods, and then broadcasted to all relay nodes in free space via the transmitting antenna.

[0098] Step 1.4, relay node file transfer step, is completed as follows:

[0099] Step 1.4.1, using the divided sub-files and user request index obtained in step 1.1 as input parameters, the relay node executes the collaborative decentralized encoding transmission method to obtain the XOR combination of the remaining requested sub-files.

[0100] Step 1.4.2, XOR combination of the remaining request sub-files obtained in step 1.4.2, sequentially execute channel coding and modulation and demodulation methods, and then broadcast them to all relay nodes in free space through the transmitting antenna.

[0101] Step 1.5, the user node receives the file information step, which is completed in the following sub-steps:

[0102] Step 1.5.1, for the sub-files received in step 1.3.2 and step 1.4.2, the channel coding and modulation and demodulation methods described in the background technology are sequentially executed to obtain the requested sub-files after XOR combination.

[0103] Step 1.5.2, the user XORs the requested sub-files combined with the sub-files in its own cache to obtain all sub-files of the requested file, and combines them to obtain the requested file.

[0104] The file division method in step 1.1 above is completed by the following steps:

[0105] Step 2.1, the system model is as follows Figure 1 As shown, the system includes a source node and K1 relay nodes, denoted as S i , i is a positive integer, 0≤i≤K1, K1 is a positive integer. K2 users, denoted as U i, i is a positive integer, 0≤i≤K2, K2 is a positive integer. Assume that there are N files in the file library, N is a positive integer, denoted by W i , i is a positive integer, 0≤i≤N, each file is B bits, B is a real number greater than or equal to 0, the source node can access all files, each relay node has the same cache capacity, recorded as M1B bits, satisfying M1≤N, each user has the same cache capacity, recorded as M2B bits, satisfying M2≤M1, M1, M2 are real numbers greater than or equal to 0. Definition

[0106]

[0107] Among them, α represents the file division ratio, and β represents the user cache capacity division ratio. k , k∈[αN], divide the files into disjoint subfiles of equal size, denoted as

[0108]

[0109] W k,T,R Indicates that the subfile is only cached in S i and U j In the above equation, i∈T, j∈R. So S i and U j The cached content is

[0110] Z 1,i = {W k,T,R},i∈T,k∈[N] (31)

[0111] Z 2,j = {W k,T,R},j∈R,k∈[N] (32)

[0112] Z 1,i and Z 2,j Represents the contents of relay node and user cache respectively.

[0113] Step 2.2, for file W k ,k∈[(1-α)N], divided into disjoint subfiles, U j ,j∈[K2]fromW k Random Cache Sub-files. Users cache all sub-files of a file in sequence, and each user's cache does not affect each other.

[0114] In addition, among them, Figure 2 In the interference network shown, the joint design of weighted interference cancellation and orthogonal unicast method in step 1.3.1 above is completed by the following steps:

[0115] Step 3.1: When t1+t2≤K2, for the sub-files that need to be delivered

[0116]

[0117] in, Indicates that the user U j The requested file is cached in the relay node set T and the user node set R. It means that element j does not belong to the set R, j∈[K2] means that element j belongs to K2, [K2]={1,2,...,K2}.

[0118] First, the sub-file to be transferred Divide into several subsets of size t1+t2, and divide each subfile into smaller files. in, indicates that the set T is a subset of [K1], [K2]\j indicates that the set [K2] does not contain element j. Divide into A smaller file, denoted by

[0119]

[0120] Among them, for the set R, ∏ R represents the full permutation of set R. For t∈{1,2,...,|R|}, |R| represents the number of elements in set R, ∏ R,t Represents all permutations and combinations of t subsets of the set R, denoted by

[0121]

[0122] By dividing the files in formula (34), the sub-file coefficients are designed to eliminate interference among users in set π′. Since the sub-file has been cached in user set π, users in set π can use the cached files to eliminate interference. Therefore, the transmission of t1+t2 sub-files to t1+t2 users is interference-free.

[0123] Based on the above file division, all files to be transferred are divided into:

[0124]

[0125] in, represents the circular arrangement of elements in set R. There are a total of (|R|-1)! circular arrangements in set R. For set R, π[a:b], b≥a is defined as

[0126]

[0127] in, For collections

[0128] For the sub-files divided by formula (36), the transmission coefficient of the relay node for each sub-file is designed to achieve interference-free transmission, S i The files sent are recorded as:

[0129]

[0130] Transmitted through different channels, U j The files received are:

[0131]

[0132] The divided files are shown in formula (34). For user U j , j∈π[l+1:l+t2], Because Z 2,j The file is already cached in U j Available through Z 2,j Interference elimination; for user U j , j∈π[l+t2+1:l+t1+t2-1], Z 2,j The file is not cached and the transmission coefficient needs to be controlled for interference elimination, namely:

[0133]

[0134] Step 3.2: When t1+t2≤K2, the transmission model Figure 3 As shown, the file, relay node cache capacity and user cache capacity are divided so that αN,M1 and βM2 satisfy

[0135]

[0136] Where M1″=γ2M1. For γ1αN, the weighted interference cancellation transmission in step 4.1 is adopted using γ2M1 and γ3βM2. For (1-γ1)αN, the traditional orthogonal unicast transmission is adopted using (1-γ2)M1 and (1-γ3)βM2.

[0137] Among them, in the above step 1.4.1, the decentralized collaborative transmission method and transmission model Figure 4 As shown, the following steps are used to complete:

[0138] Step 4.1: The relay node sends the encoded file:

[0139]

[0140] Among them, Wk, RDenote as W k This sub - file of k is only cached in Uj, j ∈ R, and R\{j} means the set R does not contain j.

[0141] Step 4.2: When K1 ≥ K2, different relay nodes perform s = K2, K2 - 1,..., 1 coded transmissions. In the case of K1 ≥ K2, the maximum number of transmissions is K2.

[0142] When K1 < K2, all relay nodes cannot perform coded transmissions for all s values at one time. Therefore, after a relay node finishes transmitting, it jumps forward by K1 units until all s - value coded transmissions are completed.

[0143] Correspondingly, in the multi - relay interference network of this embodiment, the multi - subsystem weighted interference cancellation coded caching system applies the above - mentioned multi - subsystem weighted interference cancellation coded caching method. It is characterized in that the multi - subsystem weighted interference cancellation coded caching system includes a file partitioning module, a storage module, a joint design weighted interference cancellation and orthogonal unicast module, a channel coding / decoding and modulation / demodulation module, a cooperative decentralized coded transmission module, and a file processing and integration module.

[0144] Among them, the file partitioning module of this embodiment is used to control the source node to execute the file partitioning method with the cache capacity of user nodes, the cache capacity of relay nodes, the number of user nodes, and the number of relay nodes as input parameters to obtain the partitioned sub - files. Specifically, the file partitioning method has the following process:

[0145] (1) The system model is as Figure 1 shown. The system includes a source node, K1 relay nodes, denoted as S i , where i is a positive integer, 0 ≤ i ≤ K1, and K1 is a positive integer. K2 users, denoted as U i , where i is a positive integer, 0 ≤ i ≤ K2, and K2 is a positive integer. Assume that there are N files in the file library, N is a positive integer, denoted as W i , where i is a positive integer, 0 ≤ i ≤ N, each file is B bits, B is a real number greater than or equal to 0. The source node can access all files. Each relay node has the same cache capacity, denoted as M1B bits, satisfying M1 ≤ N. Each user has the same cache capacity, denoted as M2B bits, satisfying M2 ≤ M1, and M1, M2 are real numbers greater than or equal to 0. Define

[0146]

[0147] Among them, α represents the file partitioning ratio, and β represents the user cache capacity partitioning ratio. For the file W k , k ∈ [αN], the file is partitioned into non - overlapping sub - files of equal size, denoted as

[0148]

[0149] W k,T,R Indicates that the subfile is only cached in S i and U j In the above equation, i∈T, j∈R. So S i and U j The cached content is

[0150] Z 1,i = {W k,T,R},i∈T,k∈[N] (31)

[0151] Z 2,j = {W k,T,R},j∈R,k∈[N] (32)

[0152] Z 1,i and Z 2,j Represents the contents of relay node and user cache respectively.

[0153] (2) For file W k ,k∈[(1-α)N], divided into disjoint subfiles, U j ,j∈[K2]fromW k Random Cache Sub-files. Users cache all sub-files of a file in sequence, and each user's cache does not affect each other.

[0154] The storage module of this embodiment is used to use the divided sub-files as input parameters, control the relay node and the user end to store the sub-files in the corresponding cache according to the division.

[0155] The embodiment of the present invention jointly designs weighted interference cancellation and orthogonal unicast modules. Figure 2 In the interference network shown, the divided sub-files and request indexes are used as input parameters to control the source node to perform the jointly designed weighted interference elimination and orthogonal unicast method to obtain the XOR combination of the requested sub-files; wherein the request index represents the file sequence number requested by the user node.

[0156] Specifically, a weighted interference cancellation and orthogonal unicast method is jointly designed, and the specific process is as follows:

[0157] (1) When t1+t2≤K2, for the sub-files that need to be delivered

[0158]

[0159] in, Indicates that the user U j The requested file is cached in the relay node set T and the user node set R. It means that element j does not belong to the set R, j∈[K2] means that element j belongs to K2, [K2]={1,2,...,K2}.

[0160] First, the sub-file to be transferred Divide into several subsets of size t1+t2, and divide each subfile into smaller files. in, indicates that the set T is a subset of [K1], [K2]\j indicates that the set [K2] does not contain element j. Divide into A smaller file, denoted by

[0161]

[0162] Among them, for the set R, ∏ R represents the full permutation of set R. For t∈{1,2,...,|R|}, |R| represents the number of elements in set R, ∏ R,t Represents all permutations and combinations of t subsets of the set R, denoted by

[0163]

[0164] By dividing the files in formula (34), the sub-file coefficients are designed to eliminate interference among users in set π′. Since the sub-file has been cached in user set π, users in set π can use the cached files to eliminate interference. Therefore, the transmission of t1+t2 sub-files to t1+t2 users is interference-free.

[0165] Based on the above file division, all files to be transferred are divided into:

[0166]

[0167] in, represents the circular arrangement of elements in set R. There are a total of (|R|-1)! circular arrangements in set R. For set R, π[a:b], b≥a is defined as

[0168]

[0169] in, For collections

[0170] For the sub-files divided by formula (36), the transmission coefficient of the relay node for each sub-file is designed to achieve interference-free transmission, S i The files sent are recorded as:

[0171]

[0172] Transmitted through different channels, U j The files received are:

[0173]

[0174] The divided files are shown in formula (34). For user U j , j∈π[l+1:l+t2], Because Z 2,j The file is already cached in U j Available through Z 2,j Interference elimination; for user U j , j∈π[l+t2+1:l+t1+t2-1], Z 2,j The file is not cached and the transmission coefficient needs to be controlled for interference elimination, namely:

[0175]

[0176] Step 3.2: When t1+t2≤K2, the transmission model is as follows: Figure 3 As shown, the file, relay node cache capacity and user cache capacity are divided so that αN,M1 and βM2 satisfy

[0177]

[0178] Where M1″=γ2M1. For γ1αN, the weighted interference cancellation transmission in step 4.1 is adopted using γ2M1 and γ3βM2. For (1-γ1)αN, the traditional orthogonal unicast transmission is adopted using (1-γ2)M1 and (1-γ3)βM2.

[0179] The channel coding and modulation and demodulation module of this embodiment is used to perform XOR combination of the request sub-files, sequentially execute the channel coding and modulation and demodulation methods in the prior art, and then broadcast it to all relay nodes in free space via the transmitting antenna.

[0180] The collaborative decentralized coding transmission module of this embodiment is used to take the divided sub-files and request index as input parameters, control the relay node to execute the collaborative decentralized coding transmission method, and obtain the XOR combination of the remaining requested sub-files.

[0181] Specifically, the method of decentralized collaborative transmission, the specific process image:

[0182] (1) The relay node sends the encoded file:

[0183]

[0184] Among them, W k,R represents that this sub-file of W k is only cached in U j , j ∈ R, and R\{j} means that the set R does not contain j.

[0185] (2) When K1 ≥ K2, different relay nodes perform s = K2, K2 - 1,..., 1 coded transmissions. In the case of K1 ≥ K2, the maximum number of transmissions is K2;

[0186] When K1 < K2, all relay nodes cannot perform coded transmissions for all s values at one time. Therefore, after a relay node finishes transmitting, it jumps forward by K1 units until all s values are coded and transmitted.

[0187] Moreover, the channel coding / decoding and modulation / demodulation module of this embodiment is also used to perform exclusive-OR combination on the remaining part of the requested sub-files, successively execute the channel coding / decoding and modulation / demodulation methods in the prior art, and then broadcast it to all relay nodes in free space through the transmitting antenna.

[0188] In addition, the channel coding / decoding and modulation / demodulation module of this embodiment is also used to successively execute the channel coding / decoding and modulation / demodulation methods in the prior art on the sub-files received by the relay nodes to obtain the requested sub-files that have been exclusive-OR combined.

[0189] The file processing and integration module of this embodiment is used to control the user node to perform exclusive-OR processing on the requested sub-files that have been exclusive-OR combined and the sub-files cached by itself to obtain all sub-files of the requested file, and combine them to obtain the requested file.

[0190] As Figure 6 shown, it shows the total degrees of freedom of the caching strategy proposed by the present invention in the case of t1 + t2 ≤ K2 Total degrees of freedom of the single interference cancellation strategy and the total degrees of freedom of the traditional caching strategy varying with M2 under different K2. When N is constant, as M2 increases, within a certain range, the total degrees of freedom provided by weighted interference-free transmission increases. As M2 continues to increase, when it is the case, the DoF provided by interference-free transmission is K2 and remains constant

[0191] As Figure 7 shown, it shows the total degrees of freedom of the caching strategy proposed by the present invention in the case of t1 + t2 > K2 Total degrees of freedom of the single interference cancellation strategy and the total degrees of freedom of the traditional caching strategy varying with M2 under different K2. Since the strategy proposed in this paper can obtain the DoF gain provided by unicast when transmitting (1 - γ1)αN, therefore Always higher than

[0192] The present invention applies the coding cache to a multi-relay interference network, and proposes a multi-subsystem weighted interference elimination strategy. By weighting the sub-files to be transmitted at the transmitting end, the receiving end achieves interference alignment through different channel gains. In the case where the number of valid sub-files in a single interference-free transmission is greater than the number of users, compared to ignoring some relay nodes and user cache capacity, the present invention further improves the total freedom of the system by jointly designing weighted interference elimination and orthogonal unicast transmission. In addition, in system two, based on the decentralized delivery strategy, each user decentralized collaborative transmission of the requested sub-files reduces the number of required transmissions, and reduces the transmission load while increasing the total freedom of the system. By introducing relay nodes, a method of weighted interference elimination is invented in a multi-relay interference network, introducing coding cache, reducing the link load in the delivery stage, and increasing the number of valid files in a single transmission and the total freedom, which effectively improves the effectiveness and reliability of the system and has a high application value.

[0193] Although the embodiments of the present invention have been described, it is apparent to those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents. More other equivalent embodiments may also be included without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A multi-subsystem weighted interference elimination coding cache method in a multi-relay interference network, characterized in that: The steps include: Step 1.1, taking the user node cache capacity, the relay node cache capacity, the number of user nodes and the number of relay nodes as input parameters, the source node executes the file partitioning method to obtain the divided sub-files; Step 1.2, taking the divided sub-files as input parameters, the relay node and the user end divide the sub-files and store them in the corresponding cache; Step 1.3, using the divided sub-files and the request index as input parameters, the source node performs a joint design weighted interference cancellation and orthogonal unicast method to obtain an XOR combination of the requested sub-files; wherein the request index represents the file sequence number requested by the user node; Step 1.4, XOR combination of the requested sub-files, sequentially executing channel coding and modulation and demodulation methods, and then broadcasting them to all relay nodes in free space through the transmitting antenna; Step 1.5, using the divided sub-files and the request index as input parameters, the relay node executes the collaborative decentralized encoding transmission method to obtain the XOR combination of the remaining requested sub-files; Step 1.6, XOR combination of the remaining request sub-files, sequentially executing channel coding and modulation and demodulation methods, and then broadcasting them to all relay nodes in free space through the transmitting antenna; Step 1.7, for the sub-files received in step 1.4 and step 1.6, the channel coding and modulation and demodulation methods are sequentially performed to obtain the request sub-file after XOR combination; Step 1.8, the user node XORs the requested sub-files combined with the sub-files cached by itself to obtain all sub-files of the requested file, and combines them to obtain the requested file; In step 1.1, the file division method includes the following steps: Step 2.1, the system includes a source node and K1 relay nodes, denoted as S i , i is a positive integer, 0≤i≤K1, K1 is a positive integer; K2 user nodes, denoted as U i , i is a positive integer, 0≤i≤K2, K2 is a positive integer; Assume that there are N files in the file library, N is a positive integer, denoted by W i , i is a positive integer, 0≤i≤N, each file is B bits, B is a real number greater than or equal to 0, the source node can access all files, each relay node has the same cache capacity, recorded as M1B bits, satisfying M1≤N, each user has the same cache capacity, recorded as M2B bits, satisfying M2≤M1, M1, M2 are real numbers greater than or equal to 0; definition: Among them, α represents the file division ratio, β represents the user cache capacity division ratio, for file W k , k∈[αN], divide the files into disjoint subfiles of equal size, denoted as: Among them, W k,T,R Indicates that the subfile is only cached in S i and U j In the equation, i∈T, j∈R; Therefore, S i and U j The cached contents are: Z 1,i ={W k,T,R },i∈T,k∈[N] (3) Z 2,j ={W k,T,R },j∈R,k∈[N] (4) Among them, Z 1,i and Z 2,j Respectively represent the contents cached by relay nodes and user nodes; Step 2.2, for file W k ,k∈[(1-α)N], divided into disjoint subfiles, U j ,j∈[K2]fromW k Random Cache sub-files; user nodes cache the sub-files of all files in turn, and the caches of each user node do not affect each other; In step 1.3, the weighted interference cancellation and orthogonal unicast method are jointly designed, including the following steps: Step 3.1: When t1+t2≤K2, for the sub-files that need to be delivered: in, Indicates that the user U j The requested file is cached in the relay node set T and the user node set R. Indicates that element j does not belong to set R, j∈[K2] indicates that element j belongs to K2, [K2]={1,2,...,K2}; First, the sub-file to be transferred Divide into several subsets of size t1+t2, and divide each subfile into smaller files; for the requested subfile in, indicates that the set T is a subset of [K1], [K2]\j indicates that the set [K2] does not contain element j. Divide into A smaller file, recorded as: Among them, for the set R, ∏ R represents the full permutation of set R; For t∈{1,2,...,|R|}, |R| represents the number of elements in set R, ∏ R,t Represents all permutations and combinations of t subsets of the set R, denoted as: By dividing the file in formula (6), the sub-file coefficients are designed to eliminate interference among the users in set π′. Since the sub-file has been cached in the user set π, the user nodes in set π can use the cached files to eliminate interference and transmit t1+t2 sub-files to t1+t2 users. This communication process is interference-free. Based on the above file division, all files to be transferred are divided into in, represents the circular arrangement of elements in set R. There are a total of (|R|-1)! circular arrangements in set R. For set R, π[a:b], b≥a is defined as in, For collections For the sub-files divided by formula (8), the transmission coefficient of the relay node for each sub-file is designed to achieve interference-free transmission, S i The files sent are recorded as: Transmitted through different channels, U j The files received are: The divided files are shown in formula (6). For user node U j , Because Z 2,j The file is already cached in U j Available through Z 2,j Interference elimination; for user U j , Z 2,j The file is not cached, and the transmission coefficient needs to be controlled to eliminate interference, that is: Step 3.2: When t1+t2≤K2, divide the file, relay node cache capacity and user cache capacity so that αN,M1 and βM2 satisfy: Among them, M1″=γ2M1; For γ1αN, weighted interference cancellation transmission is adopted using γ2M1 and γ3βM2; For (1-γ1)αN, conventional orthogonal unicast transmission is adopted using (1-γ2)M1 and (1-γ3)βM2.

2. The method for weighted interference elimination coding and caching of multiple subsystems in a multi-relay interference network according to claim 1, characterized in that: In step 1.5, the collaborative decentralized coding transmission method includes the following steps: Step 4.1: Relay node sends the encoded file Among them, W k,R W k This sub-file is only cached in U j ,j∈R, R\{j} means that the set R does not contain j; Step 4.2: When K1≥K2, different relay nodes perform s=K2,K2-1,...,1 coded transmission. When K1≥K2, the maximum number of transmissions is K2. When K1 < K2, all relay nodes cannot perform coded transmission for all values of s at once. Therefore, after a relay node finishes transmitting, it jumps forward by K1 units until coded transmission for all values of s is completed.

3. A multi-subsystem weighted interference cancellation coding cache system in a multi-relay interference network, applying the multi-subsystem weighted interference cancellation coding cache method according to any one of claims 1-2, characterized in that: The multi-subsystem weighted interference cancellation coded caching system includes: A file partitioning module, which uses the cache capacities of user nodes, relay nodes, the number of user nodes, and the number of relay nodes as input parameters to control the source node to execute a file partitioning method to obtain partitioned sub-files; A storage module, which uses the partitioned sub-files as input parameters to control relay nodes and user terminals to store them in corresponding caches according to the sub-file partitioning; A joint design weighted interference cancellation and orthogonal unicast module, which uses the partitioned sub-files and request indexes as input parameters to control the source node to execute a joint design weighted interference cancellation and orthogonal unicast method to obtain the exclusive-OR combination of requested sub-files; where the request index represents the file number requested by a user node; A channel coding / decoding and modulation / demodulation module, which sequentially executes channel coding / decoding and modulation / demodulation methods on the exclusive-OR combination of requested sub-files, and then broadcasts it to all relay nodes in free space via a transmitting antenna; A cooperative decentralized coded transmission module, which uses the partitioned sub-files and request indexes as input parameters to control relay nodes to execute a cooperative decentralized coded transmission method to obtain the exclusive-OR combination of the remaining requested sub-files; The channel coding / decoding and modulation / demodulation module is also used to sequentially execute channel coding / decoding and modulation / demodulation methods on the exclusive-OR combination of the remaining requested sub-files, and then broadcast it to all relay nodes in free space via a transmitting antenna; The channel coding / decoding and modulation / demodulation module is also used to sequentially execute channel coding / decoding and modulation / demodulation methods on the sub-files received by relay nodes to obtain the exclusive-OR combined requested sub-files; A file processing and integration module, which controls user nodes to perform exclusive-OR processing on the exclusive-OR combined requested sub-files and the sub-files cached by themselves to obtain all sub-files of the requested file, and combines them to obtain the requested file.

Images