CN113543067B

CN113543067B - Data issuing method and device based on vehicle-mounted network

Info

Publication number: CN113543067B
Application number: CN202110633075.3A
Authority: CN
Inventors: 徐连明; 王莉; 费爱国; 崔鹤文; 田泽宇; 魏青
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2021-06-07
Filing date: 2021-06-07
Publication date: 2023-10-20
Anticipated expiration: 2041-06-07
Also published as: CN113543067A

Abstract

The invention provides a data issuing method and device based on a vehicle-mounted network, comprising the following steps: encoding the hot content into a plurality of fragments by using an erasure code encoding technology, performing redundancy encoding on the fragments to obtain a plurality of content fragments, and respectively caching the content fragments into each VU in the cluster; receiving a loading request of a target VU in a vehicle-mounted network for hot content, sending content fragments to the target VU, and controlling other k' VUs to send the content fragments cached by each to the target VU until the target VU decodes and generates the hot content according to the received k content fragments. According to the data issuing method and device based on the vehicle-mounted network, the RSU is utilized to cache the hot spot resources, the distributed coding caching technology is adopted to code and segment the content to be cached in each VU, the robustness of content transmission is guaranteed, the space utilization rate of cache nodes is improved, the network load and the backhaul link pressure are reduced, and the time delay of acquiring the content resources by the vehicle is reduced while the base station pressure is relieved.

Description

Data issuing method and device based on vehicle-mounted network

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a data issuing method and apparatus based on a vehicle network.

Background

With the popularization of public transportation, there are more and more families with private cars, and cars have even played an indispensable role in the life of the present invention. However, with the dramatic increase in the number of vehicles, existing road infrastructure has not been able to meet these vehicle user demands. When a large number of vehicle nodes send out requests aiming at the same hot content, the phenomenon of repeated downloading of resources occurs, and the wireless link resources are wasted. And, repeated transmission of large amounts of hot content can further increase network load and backhaul link pressure, causing peak network congestion. Compared with the mobile handheld equipment, the roadside unit and the vehicle have better data storage capacity, content files are cached on the equipment, the data transmission efficiency can be effectively improved, and the pressure of a return link is reduced.

Therefore, in the prior art, hot spot resources are cached on surrounding vehicles and roadside units in advance, so that the nearby vehicles and the roadside units can realize close-range content sharing, and the time delay of acquiring the content resources by the vehicles is reduced while the pressure of the base station is relieved.

However, due to rapid changes in vehicle network topology and communication link quality, complete content may not be transferred in a short time, thereby making it impossible for a user to complete a complete download of a large volume of files.

Disclosure of Invention

Aiming at the problems existing in the prior art, the embodiment of the invention provides a data issuing method and device based on a vehicle-mounted network.

In one aspect, the present invention provides a data issuing method based on a vehicle network, including: encoding any hot content into a plurality of fragments by using an erasure code encoding technology, performing redundancy encoding on all fragments to obtain a plurality of content fragments, and respectively caching the content fragments to VU (virtual private units) of each vehicle user in a cluster; each VU stores only one piece of content shard; the clusters are determined by dividing all the VUs according to the positions of the VUs in the vehicle-mounted network and the request information of the VUs for the hot content in the scene; receiving a loading request of any target VU in the vehicle-mounted network for the hot content, and sending content fragments to the target VU; after the number of the content fragments sent to the target VU by the road side unit RSU is k-k', if the content fragments are stopped to be sent to the target VU, other k-vehicle VU are controlled to send the content fragments cached by each to the target VU until the target VU generates the hot content according to the received k content fragments.

On the other hand, the invention also provides a data issuing device based on the vehicle-mounted network, which comprises the following steps: the content issuing unit is used for encoding any hot content into a plurality of fragments by using an erasure code encoding technology, performing redundancy encoding on all the fragments to obtain a plurality of content fragments, and respectively caching the content fragments to VU (virtual Unit) of each vehicle user in a cluster; each VU stores only one piece of content shard; the clusters are determined by dividing all the VUs according to the positions of the VUs in the vehicle-mounted network and the request information of the VUs for the hot content in the scene; a first data issuing unit, configured to receive a loading request of any target VU in the vehicle-mounted network for the hot content, and send a content fragment to the target VU, so that the target VU decodes the content fragment to generate the hot content; and the second data issuing unit is used for controlling other k 'vehicles VU to send the content fragments cached by each to the target VU if the content fragments are stopped to be sent to the target VU after the RSU sends the content fragments with the number of k-k' to the target VU, and the target VU generates the hot content according to the received k content fragments.

In another aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the steps of the data issuing method based on the vehicle network as any one of the above are implemented when the processor executes the program.

In another aspect, the present invention also provides a non-transitory computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the data distribution method based on an on-board network as described in any one of the above.

According to the data issuing method and device based on the vehicle-mounted network, the RSU is utilized to cache the hot spot resources, the distributed coding caching technology is adopted to code and segment the content to be cached in each VU, the robustness of content transmission is guaranteed, the space utilization rate of cache nodes is improved, the network load and the backhaul link pressure are reduced, and the time delay of acquiring the content resources by the vehicle is reduced while the base station pressure is relieved.

Drawings

In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

Fig. 1 is a flow chart of a data issuing method based on a vehicle-mounted network provided by the invention;

fig. 2 is a diagram of a distributed content delivery scenario under the assistance of a non-orthogonal multiple access technique provided by the present invention;

FIG. 3 is a schematic diagram of a process for solving a user pair problem based on a matching theory provided by the invention;

FIG. 4 is a schematic diagram of a process for solving a V2V link by using a matching theory provided by the invention;

FIG. 5 is a schematic diagram of a process for solving resource allocation using a graph coloring algorithm provided by the present invention;

FIG. 6 is a flow chart of a pairing scheme of resource allocation and users based on the matching theory provided by the invention;

FIG. 7 is a schematic diagram of a section algorithm provided by the present invention;

fig. 8 is a schematic structural diagram of a data issuing device based on a vehicle-mounted network provided by the invention;

FIG. 9 is a schematic diagram of one of the Vus and RSU distributions provided by the present invention;

fig. 10 is a schematic diagram showing the variation of average time delay of all users with the number of users according to the present invention;

FIG. 11 is a graph showing the average traversal rate of all users as a function of the number of users according to the present invention;

FIG. 12 is a graph showing the average traversal rate of all users as a function of the number of users and the upper and lower limits;

FIG. 13 is a diagram showing the variation of average time delay of all users according to the number of users and coding parameters;

FIG. 14 is a graph showing the average spectral efficiency as a function of the number of users provided by the present invention;

FIG. 15 is a second schematic diagram of the distribution of Vus and RSU provided by the present invention;

FIG. 16 is a second diagram showing the variation of the average time delay of all users with the number of users according to the present invention;

FIG. 17 is a second diagram showing the variation of the average traversal rate of all users with the number of users according to the present invention;

fig. 18 is a schematic structural diagram of an electronic device provided by the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The existing internet of vehicles data distribution is generally implemented by caching hot spot resources on a Road Side Unit (RSU) and distributing the hot spot resources to Vehicle Users (VU), but due to rapid changes of Vehicle network topology and communication link quality, complete content may not be completely transmitted in a short time.

According to the data issuing method based on the vehicle-mounted network, a distributed coding caching technology is considered to be introduced, the content is coded and segmented by utilizing an erasure code before the content data is issued, then the coded content segments are stored in a vehicle caching node in advance, and when a content request vehicle (called a target VU) initiates a request to a certain hot content, a plurality of surrounding VUs send the respective cached content segments to the target VU so as to jointly assist the target VU in downloading the hot content. After the content requesting vehicle obtains a certain number of content pieces, the complete content can be obtained through decoding.

In view of this, two phases, content deployment (deployment of content shards) and content shard transmission (transmission of content shards), are critical to vehicle code caching.

In the content deployment stage, when the traditional orthogonal multiple access (Orthogonal multiple access, OMA) is adopted to transmit the content fragments, the content fragments are required to be issued to different content cache vehicle nodes in sequence, and the problem of low spectral efficiency exists, so the invention considers that the content fragments are issued to each VU in a scene by utilizing a Non-orthogonal multiple access (Non-orthogonal multiple access, NOMA) technology, thereby improving the content fragment deployment efficiency, and in the NOMA, which receiving nodes are selected to be matched and paired has a remarkable influence on the accessibility of the NOMA system.

Second, how the vehicle selects a communication link to acquire a content piece and how to plan the resource usage of the two phases also has a significant impact on the performance of the system during the content transfer phase.

Finally, the decision problem in the whole process is time consuming, and how to make a fast decision by the target VU is also an important issue.

Currently, many research works only consider the deployment of content shards, but do not consider both the content shard delivery and transmission processes at the same time, and do not consider the time taken by the decision problem.

Therefore, the invention is based on the whole process of content deployment and content transmission, comprehensively considers the communication conditions of the VUs in the two processes and reasonably manages the resources so that all the VUs in the scene can recover to obtain the required complete content as soon as possible.

The following describes a data issuing method and device based on a vehicle-mounted network provided by the embodiment of the invention with reference to fig. 1 to 18.

Fig. 1 is a flow chart of a data issuing method based on a vehicle-mounted network, as shown in fig. 1, including but not limited to the following steps:

step 101: encoding any hot content into a plurality of fragments by using an erasure code encoding technology, performing redundancy encoding on all fragments to obtain a plurality of content fragments, and respectively caching the content fragments to VU (virtual private units) of each vehicle user in a cluster; each VU stores only one piece of content shard.

The clusters are determined by dividing all the VUs according to the positions of the VUs in the vehicle-mounted network and the request information of the VUs for the hot content in the scene;

step 102: and the RSU receives a loading request of any target VU in the vehicle-mounted network for the hot content and sends the content fragments to the target VU.

Step 103: after the RSU sends the content fragments to the target VU, if the content fragments are stopped to be sent to the target VU, the RSU controls other k' VU to send the content fragments cached by the RSU to the target VU until the target VU decodes and generates the hot content according to the received k content fragments.

The invention assumes that there are U RSUs with buffering function in the scene, and defines the index of the RSU asAt the same time, there are a certain number of VUs in the scene.

In step 101, according to the locations of the VUs and the request conditions for hot content in the scene, all the VUs are divided into several clusters (assuming that the vehicles in each cluster are running in the same direction), all the RSUs store all the hot content required by the VUs in the scene, and the hot content is defined asAssume that the size of the f-th content is C _f 。

The hot content is generally larger content such as road condition high definition map, each RSU issues content for all Vus in a current cluster, and the number of Vus in a vehicle cluster served by a u-th RSU is assumed to be K in a certain time slot _u 。

Because the mobility of the vehicle can hardly realize the transmission of the complete hot content to the target VU in the communication coverage area of the RSU, the RSU firstly divides the hot content into k pieces by using the erasure code coding technology at the RSU side, then obtains n pieces of content fragments by redundant coding, and then transmits the content fragments to any target VU needing content downloading in a scene by combining the NOMA technology with the V2I communication technology according to the condition of a communication link.

After receiving the content fragments, the target VU sends the content fragments cached by the target VU to the target VU (and also sends the content fragments to other VUs) according to the content fragments owned by the target VU and surrounding VUs (namely, the set formed by multiple VUs), and obtains the required content fragments from other VUs, if a certain target VU owns k different content fragments, the target VU can restore the required complete content through decoding.

The erasure code encoding technique may be an (n, k) maximum distance separable code (Maximum Distance Separable, MDS) encoding technique, specifically, n pieces of content with redundancy are obtained by encoding the content, and stored in different VUs, so that the buffer size of each vehicle user for one content is reduced, and meanwhile, any target VU needing to download the content can obtain the original content by decoding only by obtaining any k pieces of the n pieces of content.

Specifically, in step 102, firstly, a certain target RSU issues content fragments to VUs in its corresponding cluster, for any target VU, since the target VU is always in an operation state, after receiving a k-k' fragment of content sent by the RSU, it is no longer in the cluster where the target RSU is located, and then the target RSU terminates sending content fragments to the target VU (or fails to send), and step 103 is entered.

In step 103, after the road side unit RSU sends the number of content fragments to the target VU as k-k ', if the sending of the content fragments to the target VU is terminated, the base station controls other k' VUs to send the content fragments cached by each to the target VU until the target VU receives k content fragments in total. At this time, the target VU may generate the hot content according to the received k content fragment decodes.

According to the data issuing method based on the vehicle-mounted network, the RSU is utilized to cache the hot spot resources, the distributed coding caching technology is adopted to code and segment the content to be cached in each VU, the robustness of content transmission is guaranteed, the space utilization rate of the caching nodes is improved, the network load and the backhaul link pressure are reduced, and the time delay of acquiring the content resources by the vehicle is reduced while the base station pressure is relieved.

Based on the content of the above embodiment, as an alternative embodiment, before caching the content pieces at the respective vehicle users VU in one cluster, further includes:

dividing all VUs in the cluster into strong users and weak users according to channel conditions;

performing pairwise team formation on the strong user and the weak user to obtain a plurality of NOMA user pairs; each of said NOMA user pairs consists of a strong user and a weak user;

and taking the strong user in the NOMA user pair as the NOMA strong user, taking the weak user in the NOMA user pair as the NOMA weak user, and taking the rest strong users or weak users which cannot be formed into a team as V2I users.

The data issuing method based on the vehicle-mounted network also considers the problem that the frequency spectrum utilization rate is low when the vehicle singly uses the special frequency spectrum in the communication process, and therefore, the NOMA technology is adopted to assist data issuing.

Fig. 2 is a schematic diagram of a distributed content delivery scenario under the assistance of a non-orthogonal multiple access technology provided in the present invention, as shown in fig. 2, assuming that VUs in the same cluster applies for the same hot content in one time slot. The complete communication process is shown in fig. 1, and the complete process is shown in fig. 2 by taking (4, 3) codes as an example.

Firstly, considering channel conditions, classifying Vus in each cluster into three types of users, namely NOMA strong users, NOMA weak users and V2I users, and then distributing content fragments to the NOMA strong users, the NOMA weak users and the V2I users in different modes according to classification conditions of the Vus.

Then, the Vus in the scene shares the own content fragments to other Vus according to the received content fragments and the V2V link conditions among surrounding Vus, and receives other content fragments from other Vus to restore the complete content through decoding, so that the Vus in the whole cluster can obtain the required content.

(1) Regarding the content delivery stage (hereinafter referred to as "first stage"):

the invention firstly considers that NOMA technology is used in the content delivery process, namely how to divide Vus in clusters into: NOMA strong users, NOMA weak users, and V2I users, such that the RSU delivers content fragments to the VUs in different ways.

Specifically, consider the u-th RSU and the corresponding cluster first, consider that there is K in the cluster _u Vehicle VU, noted asRecord h _i The channel power gain for the RSU to send a message to the i-th VUs, considering that the channel obeys the rayleigh distribution, can be expressed as:

h _i ＝δ _i d _i ^-α ；

Wherein d is _i Delta is the distance from the RSU to the ith Vus _i For this reason the channel fading coefficients RSU to the i-th VUs, a is the path loss index.

Dividing the Vus in the cluster into two parts of strong users and weak users according to the channel condition, for example, dividing the Vus with the channel intensity larger than a certain threshold value into strong users, dividing the Vus with the channel intensity smaller than or equal to the threshold value into weak users, and respectively marking asWherein->Satisfy m+n=k _u . There are two paired methods between the strong and weak users:

a) NOMA pairs: the strong user and the weak user respectively act as the strong user and the weak user in the NOMA pair, namely respectively work in the NOMA strong user mode and the NOMA weak user mode, and can use the same frequency spectrum resource under the condition of interference.

b) Virtual pair: that is, when the number of strong users is greater than the number of weak users, or the number of weak users is greater than the number of strong users, there is no way for some VUs to form NOMA pairs with another type of VUs, and then the remaining strong users or weak users that cannot be grouped are used as V2I users (as V2I users operate in V2I mode), that is, the V2I users directly receive content clips from the RSU through the V2I link.

Based on the content of the foregoing embodiment, as an optional embodiment, the buffering the content segments to the respective vehicle users VU in one cluster includes:

And transmitting the content fragments to each NOMA strong user, each NOMA weak user and each V2I user by adopting a V2I link.

Further, the controlling the other k' VUs to send the respective cached content pieces to the target VU includes:

when the target VU is a NOMA strong user or a NOMA weak user, any other NOMA strong user or any other NOMA weak user adopts a NOMA communication mode to send the content fragments to the target VU, and the total number of any other NOMA strong user and any other NOMA weak user is k'; correspondingly, if the target VU is a V2I user, the roadside unit RSU continues to adopt a V2I link to send the content fragments to the target user.

In addition, the NOMA strong users and the NOMA weak users in the same sub-cluster adopt a multiplexing spectrum mode to perform interactive transmission of content fragments; the number of VUs in the sub-cluster is greater than or equal to 2.

Recording deviceIndex of user pair formed for strong and weak users, < ->And g=max { M, N }, is satisfied.

Assuming that the spectrum is divided into Resource Blocks (RBs) having the same bandwidth, the total bandwidth of the spectrum Resource is B, and assuming that the V2I link does not multiplex the spectrum in the data transmission stage (first stage), the spectrum Resource is divided into G RBs, and the bandwidth of each RB is B/G. Record h _g，S And h _g，W RSU and g user pair respectivelyChannel power gains of communication links between the medium-strength and weak users are respectively as follows:

h _g，S ＝δ _g，S d _g，S ^-α (1a)

h _g，W ＝δ _g，W d _g，W ^-α (1b)

wherein d is _g，S 、d _g，W The distances between RSU and strong and weak users in g-th user pair are respectively delta _g，S 、δ _g，W Channel fading coefficients of strong and weak users in the RSU to g-th user pair, respectively, and obey rayleigh fading.

Definition of gamma _g，S And gamma _g，W And the signal to noise ratios (Signal to Interference plus Noise Ratio, SINR) of the strong and weak users in the g-th user pair are respectively, and the signal to noise ratios of the strong and weak users in the g-th user pair are respectively:

wherein beta is _g E (0, 1/2) is the power allocation factor, P, in the g-th user pair _RSU Is the total power, σ, of the RSU in sending messages to each user pair ² Is additive white gaussian noise (Additive White Gaussian Noise AWGN).

When the g-th user pair is a virtual pair, if the VU in the g-th user pair is a strong user, beta _g =1, if this VU is a weak user, β _g =0. The rates of the strong and weak users in the g-th NOMA pair are respectively:

definition of variablesRepresenting whether the VU is in the g-th user pair (the case that the variable X is the VU constitutes the user pair), if the q-th Vus is in the g-th user pair, X _q，g =1, otherwise x _q，g =0, wherein>

Assuming that the content required in the cluster is f, after (n, k) encoding, each content slice has a size of C _f In order to fully utilize the performance superiority of the RSU for issuing the content, the k-k' piece of content is issued in the issuing process, and the issuing process is required to be carried out at t ₁ Completed in time.

(2) Regarding the V2V communication phase (hereinafter referred to as the second phase):

taking the scenario shown in fig. 2 as an example, the whole process of content distribution and sharing is described by taking (4, 3) code as an example (i.e. n=3, k=3), and as can be seen from fig. 2, VUs selects different modes of receiving content fragments in the content distribution process, namely, NOMA strong user, NOMA weak user, and V2I user, wherein the NOMA strong user, NOMA weak user, and V2I user all receive content fragments from RSUs.

After the content distribution is finished, when the content sharing process is entered, the target VU already receives k-k 'pieces of content fragments from the RSU, so that the complete content can be recovered by decoding only by receiving the content fragments from Vus of other k' vehicles having different fragments.

Considering that the f hot content required by the Vus in the cluster is coded by erasure codes to obtain n pieces of content fragments, and marking the n pieces of content fragments as Record variable->If the q-th Vus receives the content fragment j from the RSU, y is the storage condition of the content fragment by the Vus after the RSU issues the content fragment _qj =1, otherwise y _qj =0, wherein>

It is assumed that each VU can acquire a desired piece of content fragmentation from any different VU. Recording variableRepresenting the sharing of content segments between VUs, if the qth VUs sends a content segment to the ith VUs, then z _qi =1, otherwise z _qi =0, wherein>

Recording h when the qth Vus sends a content fragment to the ith Vus _qi The channel power gain when a content fragment is sent to the ith VUs for the qth VU can be expressed as:

h _qi ＝δ _qi d _qi ^-α (4)

wherein d is _qi Delta is the distance between the qth and the ith Vus _qi Is the channel fading coefficient between the q-th Vus and the i-th Vus.

Optionally, in order to improve the spectrum utilization rate, the data issuing method based on the vehicle-mounted network provided by the invention assumes that the VUs and the content issuing stage use the same spectrum resource when in V2V communication in the content sharing stage, and re-divide the spectrum resource into RBs with equal bandwidths, and the RBs can be multiplexed between the VUs in the V2V process.

Considering that all VUs in a current cluster are clustered again according to interference relation among the VUs, and the obtained VUs in the same sub-cluster use the same RB, wherein the number of the Vus in each sub-cluster is not more than 2, and the Vus is divided into R sub-clusters, and the cluster set is recorded as Definition variable A ε {0,1} ^Q×R Representing the situation that the Vus is divided into sub-clusters, if the qth Vus is divided into the (r) th sub-cluster when transmitting content fragmentation, a _qr =1, otherwise a _qr ＝0。

Definition of gamma _qi The received SINR for the qth VUs when sending the content fragmentation to the ith VUs can be expressed as:

wherein p is _qi Transmitting power g when transmitting content fragments for the qth VU to the ith VU _q′i′ Interference channel power gain for the (i) th Vus when transmitting content fragmentation to the (i) th Vus for the (q) th Vus, then p _q′i′ For the power at which the q 'th vehicle VU transmits content clips to the i' th vehicle VU.

Thus, the rate at which the ith VU receives content slices is:

based on the foregoing embodiment, as an optional embodiment, before any hot content is encoded into a plurality of slices by using the erasure coding encoding technique, performing redundancy encoding on all the slices to obtain a plurality of content slices, and buffering the content slices to VU of each vehicle user in a cluster, the method further includes:

constructing a data issuing optimization model by taking the minimum total time delay as a target; solving the data issuing optimization model to obtain the condition and beta of the user pair composed of the optimal VU _g An optimal power distribution factor matrix, an optimal transmission content fragmentation situation matrix between VUs, an optimal RB allocation situation matrix, an optimal transmission power distribution matrix in a V2V process, the number of k and the number of k' in the composed user pair; wherein beta is _g Indicating that the g-th user pair is a NOMA user pair.

Concrete embodimentsThe invention uses the traversing speed to evaluate the communication time delay. In order to minimize the maximum of all VUs communication delays in both phases, the problem can be modeled. T-shaped memory _i(pair) For the communication delay of the ith VU in the first stage, T _i(pair) The expression of (2) is:

accordingly, the data rate of the ith VU in the first phase can be defined as R _i(pair) And the communication time delay of the ith VU in the second stage is T _i(V2V) The data rate is R _i(V2V) . In the first phase, the VU has received the content fragments of k-k 'slices from the RSU, so that in the second phase, only one content fragment is received from each of the k' other VUs to decode to obtain the complete content, T _i(V2V) The expression of (2) is:

in order to minimize the total delay of all VUs in both phases, the present invention models the problem as

P _RSU ＞P _V2V (9l)

k-k′≥k′ (9m)

Specifically, in formula (9), the optimization variable X is the case where VU constitutes a user pair; optimizing variablesContent fragmentation by VU after content fragmentation is issued for RSUIf the qth VU receives the content fragment j from the RSU, y _qj =1, otherwise y _qj =0; the optimization variable beta is beta _g A user centering power distribution factor matrix is formed; the optimization variable Z is a matrix for representing the situation of transmitting content fragmentation among the Vus; the optimization variable A is an RB allocation condition matrix; the optimization variable P is the transmission power allocation matrix in the V2V process.

Specifically, the constraint (9 a) of the above-mentioned optimization problem indicates that each VU can be in only one user pair; constraint (9 b) indicates that there is at most one strong user in each user pair; constraint (9 c) indicates that there is at most one weak user in each user pair; constraint (9 d) indicates that if the g-th user pair is a NOMA pair, the transmit power allocation factor is greater than 0 and less than 1/2; constraint (9 e) indicates that the first-stage content delivery stage is to be at t ₁ Completing in time; constraint (9 f) indicates that the second phase V2V communication phase is to be at t ₂ Completing in time; the constraint (9 g) indicates that each VU can only be in one sub-cluster when sending content fragments; constraint (9 h) indicates that there are at most two VUs in each sub-cluster; constraint (9 i) indicates that the VU of the second stage V2V process has a transmit power less than P _V2V The method comprises the steps of carrying out a first treatment on the surface of the Constraint (9 j) indicates that the VU is required to be able to receive content slices from k' users in the second phase to obtain enough content slices for decoding to obtain complete content; the constraint (9 k) indicates the first stage VU transmit power P _RSU Transmit power P with VU of second phase V2V procedure _V2V The sum isThe constraint (91) indicates the first stage VU transmit power P _RSU Transmit power P of VU to be greater than that of the second stage V2V process _V2V The method comprises the steps of carrying out a first treatment on the surface of the The constraint (9 m) indicates that the number of content slices delivered in the first phase is greater than the number of content slices received in the second phase to ensure that the VU can obtain the desired number of slices.

As an alternative embodiment, the invention decomposes the problem related to the formula (9) after constructing the data issuing optimization model shown in the formula (9), including but not limited to the following steps:

acquiring position information of each VU and a maximum value of transmitting power, wherein the maximum value of transmitting powerDividing all VU in the cluster into strong users and weak users according to channel conditions, calculating the communication time delay sum of NOMA user pairs formed by each strong user and each weak user, and determining the weight of the strong user and the weak user according to the communication time delay sum of all the strong users and the communication time delay sum of all the weak users; based on Hungary algorithm, taking strong and weak user weight under the condition that the traversing speed of all VUs and the user when the maximum are determined to be met as user maximum weight matching x ₁ The method comprises the steps of carrying out a first treatment on the surface of the Matching x according to user maximum weight ₁ Calculating the time delay t of each VU ₁ The method comprises the steps of carrying out a first treatment on the surface of the Taking two groups of same VUs as two point sets, and assigning an edge weight value to infinity when the two point sets are the same VU; when the two point sets are different VUs, setting the edge weight as the time delay of the two VUs for transmitting the message; acquiring minimum weight matching x by using many-to-many matching algorithm ₂ To determine all VU reception delays and minimum V2V link pairings; taking all the VUs as a set, and calculating time delay when any two VUs in the set multiplex resources; if the time delay is larger than the sum of the time delays when each VU uses the resources independently, constructing an edge between points corresponding to every two VUs, and thus obtaining a traditional graph; coloring the traditional graph based on a hypergraph coloring algorithm, wherein two points with edges are colored with different colors, and points corresponding to the same resource are colored with the same color; acquiring the resource multiplexing condition x according to the coloring result ₃ And according to x ₃ Calculating the V2V communication time delay t of each VU ₂ The method comprises the steps of carrying out a first treatment on the surface of the According to t ₁ 、t ₂ Calculating the total time delay t of all VU in the issuing stage and the communication stage; method for obtaining maximum value based on single variable combined with x ₁ 、x ₂ 、x ₃ P pair P _V2V And P _RSU Optimization is performed so that the total delay t is minimized.

Specifically, the solution can be performed by dividing the equation (9) into two sub-problems according to content distribution (first stage) and content sharing (second stage).

First, considering the content delivery process, a sub-problem one can be obtained:

s.t. constraints (9 a) to (9 e) (12 a)

In the first sub-problem, the maximization can be achieved by optimizing X and beta first

When X to exp (θ), P is a constant, it is possible to obtain:

wherein, the liquid crystal display device comprises a liquid crystal display device,is an exponential integral function. Suppose strong user channel fading obeys Rayleigh distribution and has mean value lambda _S Weak user channel fading obeys Rayleigh distribution and has mean value lambda _W Definition mu _g，S ＝λ _S d _g，S ^-α /σ ² ，μ _g，W ＝λ _W d _g，W ^-α /σ ² . The traversal rates of strong and weak users in the g-th user pair during Rayleigh fading can be obtained as follows:

the total traversal rate of the g-th user pair is r _g ＝r _g，S +r _g，W 。

If the g-th user pair is a virtual pair and the VU in the g-th user pair is a strong user, beta _g =1, if VU in this virtual pair is a weak user, β _g ＝0。

As can be seen from the reference, when channel fading follows the rayleigh distribution, to reduce the complexity in finding the traversal rate, the upper and lower limits of the traversal rate can be relaxed, and defined:

wherein:

the upper and lower limits of the strong user obtained by the invention are respectively as follows:

the upper and lower limits of the weak users are respectively:

FIG. 3 is a schematic diagram of a process for solving a user pair problem based on a matching theory, as shown in FIG. 3, the weight between strong and weak users determined by the above formulas (18 a) - (18 b) and formulas (19 a) - (19 b) is calculated; then, a user pair is formed between users using a matching algorithm to solve equation (12) using a matching theory.

Further, the present invention considers the process of sharing content shards (second stage), and the available problem two:

s.t. constraint (9 f) to constraint (9 j) (19 a)

In sub-problem two, Z, A, P can be optimized first to maximize

When spectrum resources are allocated to the VUs, a graph coloring algorithm can be used, firstly, an edge is established according to the data rate condition when two VUs use the same spectrum resources at the same time, when the qth VU sends a message to the ith VU, the same resource block is used when the qth VU sends a message to the ith VU, if the data rate is smaller than a threshold value, the edge is constructed between the qth VU and the qth VU, and the data rates of the qth VU and the qth' VU are respectively:

according to the reference, when X ₁ ～exp(α ₁ )，X ₂ ～exp(α ₂ ) At this time, it is possible to obtain:

wherein the method comprises the steps ofAssume channel power gainThe data rates of the q-th and q' -th VUs can be written as:

fig. 4 is a schematic diagram of a process of solving a V2V link by using a matching theory provided by the present invention, and fig. 5 is a schematic diagram of a process of solving a resource allocation by using a graph coloring algorithm provided by the present invention, where the present invention uses the schematic diagram as a weight when matching a V2V link and an edge weight relationship in a conventional graph, and solves the V2V link by using the matching theory shown in fig. 4, and solves a resource allocation situation by using the graph coloring algorithm shown in fig. 5, so as to solve formula (19).

Fig. 6 is a flowchart of a resource allocation and user pairing scheme based on the matching theory provided by the invention, and in summary, the invention uses a one-to-one matching algorithm to solve equation (12), and then uses a graph coloring algorithm to solve equation (19).

FIG. 7 is a schematic diagram of a section algorithm provided by the present invention, and with reference to FIG. 7, the whole solving process is as follows:

first, the strong users and the weak users are seen as two point sets, and the sum of the traversing speeds of the user pairs of each strong user and each weak user in the user pair is calculated. When solving for the traversal rate sum, due to R _g，S Is along with beta _g Monotonically increasing, R _g，W Is along with beta _g Monotonically decreasing, the sum of the traversal rates follows β _g Monotonically increasing, the present invention can use the binary algorithm shown in FIG. 7 for β _g Optimizing and then optimizing beta _g Maximizing the traversing speed sum in the g-th user pair, and taking the traversing speed sum as the edge weight between two users.

Then, the huntarian algorithm is used to find the maximum weight match to determine NOMA pairwise.

Further, consider the set of VUs as two sets of points, each containing all VUs, assuming one set is the set of transmitting users and the other set is the set of receiving users. Since each receiving user needs to receive content clips from k' transmitting users, which is a many-to-many problem, a many-to-many matching algorithm can be utilized to find the minimum weight match to determine the V2V communication link formation.

And calculating the traversing rate of the q-th VU and the q '-th VU when multiplexing the resource block based on the V2V link formation condition, and if the traversing rate is smaller than a threshold value, considering that an edge exists between the q-th VU and the q' -th VU, so as to construct the traditional graph.

Then, solving is carried out by using a graph coloring algorithm to obtain the multiplexing condition of power among the Vus. In the above process, the present invention divides power into P _RSU P _V2V Then, according to the method of using single variable to calculate the maximum value of the paired condition, the V2V link formation condition and the resource multiplexing condition, under the constraint condition (9 k) to the constraint condition (9 m), P is calculated _V2V 、P _RSU And optimizing to minimize the communication delay of the Vus in the whole process.

Fig. 8 is a schematic structural diagram of a data issuing device based on a vehicle-mounted network, where, as shown in fig. 8, the data issuing device mainly includes: a content issuing unit 81, a first data issuing unit 82, and a second data issuing unit 83, wherein:

the content issuing unit 81 is mainly configured to encode any hot content into a plurality of slices by using erasure code encoding technology, perform redundancy encoding on all the slices to obtain a plurality of content slices, and respectively cache the content slices to each vehicle user VU in a cluster; each VU stores only one piece of content shard; the clusters are determined by dividing all the VUs according to the positions of the VUs in the vehicle-mounted network and the request information of the VUs for the hot content in the scene.

The first data issuing unit 82 is mainly configured to receive a loading request of any target VU in the vehicle-mounted network for the hot content, and send a content fragment to the target VU, so that the target VU decodes the content fragment to generate the hot content.

The second data issuing unit 83 mainly controls the other k 'VUs to send the content fragments cached by each to the target VU after the road side unit RSU sends the content fragments k-k' to the target VU, if the sending of the content fragments to the target VU is terminated, until the target VU decodes and generates the hot content according to the received k content fragments.

It should be noted that, when the vehicle network-based data issuing device provided in the embodiment of the present invention specifically operates, the vehicle network-based data issuing method described in any one of the foregoing embodiments may be executed, which is not described in detail in this embodiment.

In order to fully show the beneficial effects of the data issuing method and device based on the vehicle-mounted network, the performance of the invention is verified through experimental simulation results.

Specifically, in fig. 6, the present invention adopts a hungarian matching algorithm, and uses HK algorithm and GS algorithm as references in the simulation verification, and compares the total time delay and running time of all vehicle users when using the three matching algorithms based on different vehicle distribution scenes, so as to compare the advantages and disadvantages of the different matching algorithms in the scenes considered in this chapter.

Wherein when the matching algorithm is used for the first time, the two matched parties are the strong channel Vus and the weak channel Vus. When the GS algorithm is used, the preference list of the strong channel Vus to the weak channel Vus is set as the transmission delay of the strong channel Vus when the strong Vus and the weak Vus form user pairs; the preference list of the weak channel Vus to the strong channel Vus is set as the transmission delay of the weak channel Vus when the strong Vus and the weak Vus form a user pair.

For the HK algorithm, the side weight relationship of the matched two parties is not considered, and only connectivity of the two parties is considered, so that the side weight between each strong VU and each weak VU is set to be a logic value 1, which represents that the two parties can be communicated. And (3) aiming at the Hungary algorithm, minimizing the weight sum, and setting the edge weight between the strong channel Vus and the weak channel Vus as the transmission delay sum of the strong and weak users when the Hungary algorithm is used.

When the matching algorithm is used for the second time, the two matched sides are all Vus, when the GS algorithm is used, if the two Vus are the same Vus, the preference is set to be minus infinity, the representation cannot be matched, if the two Vus are not the same Vus, the preference is set to be transmission delay, and the preference list matrixes of the two Vus are in a transposed relation. When the HK algorithm is used, the edge weight values of different VUs on two sides are set to be logic value 1, which means that the two VUs can be communicated, and the same VUs are set to be logic value 0, which means that the two VUs cannot be communicated. When using the hungarian algorithm, the edge weights between different VUs on two sides are set as transmission delays, and the edge weights between the same VUs are set as minus infinity, which means that the two VUs cannot be matched.

In order to simplify the experimental flow, the invention considers the situation that only one cluster exists in the scene, and respectively aims at different maximum values of the transmitting powerIn contrast, the path loss factor is set to be alpha=3, and the Gaussian white noise power is set to be sigma ² The data rate threshold of the vehicle user forming the NOMA pair is the data rate when the vehicle user is operating in OMA state, the threshold of the data rate of the vehicle user during V2V is the data rate when the vehicle user is not multiplexing resource blocks with other users, the current result is (3, 2) code (i.e. n=3, k=2), k' =1, simulation result and analysis are as follows:

(one) when RSUs are evenly distributed on one side of a lane:

fig. 9 is a schematic diagram of distribution of VUs and RSU provided by the present invention, considering that the width of the lane is 20m, the length is 80m, the RSU is located at the middle position on one side of the lane, wherein considering that the safety distance between the vehicle size and the workshop, considering that the position distance between the center of each vehicle is 8m, the width is 5m, and the RSU is located at the position of coordinates (40, 0), in which case the distribution of vehicles and RSUs is shown in fig. 9.

FIG. 10 is a diagram of the present inventionAs shown in FIG. 10, the triangle dotted line is the case of the Hungary matching algorithm used in the invention, the open circle is the case of the GS matching algorithm, the solid circle is the case of the HK matching algorithm, and the solid line represents the corresponding relationship The dashed line bar represents the corresponding equivalent +.>In this case.

From simulation analysis, it is known that the use of the hungarian algorithm in the proposed solution is superior to other matching algorithms. At the position ofIn the case of (2), the average time delay of the Hungary algorithm is reduced by 14.4% compared with the average time delay of the GS algorithm, and the average time delay of the HK algorithm is reduced by 39.7%. At->In the case of (2), the average time delay of the Hungary algorithm is reduced by 19.9% compared with the average time delay of the GS algorithm, and is reduced by 50.3% compared with the average time delay of the HK algorithm.

Fig. 11 is a schematic diagram showing the variation of the average traversal rate of all users with the number of users, and as shown in fig. 11, the hungarian algorithm used in the present invention is superior to other matching algorithms. At the position ofUnder the condition, the Hungary algorithm is improved by 4.7% relative to the GS algorithm traversing speed, and is improved by 58.9% relative to the HK algorithm traversing speed. At->Under the condition, the Hungary algorithm is improved by 6.8% relative to the GS algorithm traversing speed, and is improved relative to the HK algorithm traversing speedThe improvement is 83.9%.

FIG. 12 is a graph showing the average traversal rate of all users according to the number of users and the upper and lower limits, as shown in FIG. 12, the traversal rate is calculated by comparing the upper and lower limits in the first stage, which is used in the simulation experiment Triangle lines adopt Hungary matching algorithm, and the upper and lower limits are not used; the hollow circle line is the case using the hungarian matching algorithm and the upper limit in the first stage; the solid bars are the case where the hungarian matching algorithm is used and the lower limit is used. From the simulation analysis, the upper and lower limits of the relaxation are not much different from the results of the upper and lower limits of the non-use. The result of using the lower limit to calculate the traversal rate differs by 4.6% compared to the result without the upper and lower limits. The results using the upper limit calculated traversal rate were 3.7% different than those without the upper and lower limits.

Fig. 13 is a schematic diagram showing the variation of average time delay of all users along with the number of users and coding parameters, and the hungarian algorithm is used in the simulation experiment. Wherein the triangle lines representIn the case of k=2, k' =1, the open circles represent +.>In the case of k=4, k' =1, the solid circles representk=2, k' =0, i.e. only the first phase V2I communication, the solid line represents the corresponding timeThe dashed lines represent the corresponding cases +.>When (1)And (3) the situation.

From simulation analysis, the delay is minimal when k=2 and k' =1. At the position ofIn the case where k=2 and k ' =1, the time delay is reduced by 19.6% with respect to the case where k=4 and k ' =1, and the time delay is reduced by 32.6% with respect to the case where k=2 and k ' =0. At- >In the case where k=2 and k ' =1, the time delay is reduced by 22.7% with respect to the case where k=4 and k ' =1, and the time delay is reduced by 37.0% with respect to the case where k=2 and k ' =0.

Fig. 14 is a schematic diagram showing the variation of average spectrum efficiency with the number of users, where the average spectrum efficiency is expressed as the ratio of the total spectrum efficiency to the number of users, and the hungarian algorithm is used in the simulation experiment. The triangular bars represent the average spectral efficiency for the first phase of the transmission using the NOMA technique, the solid circles represent the average spectral efficiency for the first phase of the transmission using the OMA technique, and the solid lines represent the corresponding circlesThe dashed lines represent the corresponding cases +.>In this case.

From simulation analysis, NOMA delivery is more advantageous than OMA delivery in terms of average spectral efficiency. At the position ofUnder the condition, NOMA issue is improved by 4.9% compared with OMA issue average spectrum effect. At->Under the condition, NOMA issue is improved by 4.7% compared with OMA issue average spectrum effect.

Two) when the RSUs are equally distributed on both sides of the lane:

the vehicle and RSU distribution in this case is shown in fig. 14.

Fig. 15 is a schematic diagram of another VUs and RSU distribution provided by the present invention, considering that the width of the lane is 20m, the length is 80m, the RSU is located at the middle position of the lane, considering that the safety distance between the vehicle size and the workshop, considering that the position distance between the center of each vehicle is 8m, the width is 5m, and the RSU is located at the position of coordinates (40, 10), in which case the vehicle and RSU distribution is shown in fig. 15.

FIG. 16 is a diagram showing the total delay of all users according to the number of users, as shown in FIG. 16, the triangle dotted line is the case of the Hungary matching algorithm used in the present invention, the open circle is the case of the GS matching algorithm, the solid circle is the case of the HK matching algorithm, and the solid line represents the corresponding relationshipThe dashed line bar represents the corresponding equivalent +.>In this case.

From simulation analysis, it is known that the use of the hungarian algorithm in the proposed solution is superior to other matching algorithms. At the position ofIn the case of (2), the average time delay of the Hungary algorithm is reduced by 10.1% compared with the average time delay of the GS algorithm, and the average time delay of the HK algorithm is reduced by 42.1%. At->In the case of (2), the average time delay of the Hungary algorithm is reduced by 16.4% compared with the average time delay of the GS algorithm, and the average time delay of the HK algorithm is reduced by 54.0%.

Fig. 17 is a schematic diagram showing the variation of the average traversal rate of all users with the number of users, and as shown in fig. 17, the hungarian algorithm is superior to other matching algorithms.At the position ofUnder the condition, the Hungary algorithm is improved by 4.5% relative to the GS algorithm traversing speed, and is improved by 54.1% relative to the HK algorithm traversing speed. At the position of Under the condition, the Hungary algorithm is improved by 6.4% relative to the GS algorithm traversing speed, and is improved by 75.8% relative to the HK algorithm traversing speed.

In summary, as can be seen from the results of different vehicle distributions, the effect of the different positions of the RSUs on the total delay in the first stage is obvious, because the vehicle and the RSUs are distributed differently, there is no effect on the V2V link, and only the distance between the vehicle and the RSUs is affected, i.e. the effect on the link of the RSU for issuing the content fragment is greater, so that the trend of the results of the two vehicle distributions is the same.

Fig. 18 is a schematic structural diagram of an electronic device provided by the present invention, and as shown in fig. 18, the electronic device may include: processor 810, communication interface (Communications Interface) 820, memory 830, and communication bus 840, wherein processor 810, communication interface 820, memory 830 accomplish communication with each other through communication bus 840. The processor 810 may invoke logic instructions in the memory 830 to perform an in-vehicle network based data delivery method comprising: encoding any hot content into a plurality of fragments by using an erasure code encoding technology, performing redundancy encoding on all fragments to obtain a plurality of content fragments, and respectively caching the content fragments to VU (virtual private units) of each vehicle user in a cluster; each VU stores only one piece of content shard; the clusters are determined by dividing all the VUs according to the positions of the VUs in the vehicle-mounted network and the request information of the VUs for the hot content in the scene; receiving a loading request of any target VU in the vehicle-mounted network for the hot content, and sending content fragments to the target VU; after the number of the content fragments sent to the target VU by the road side unit RSU is k-k', if the content fragments are stopped to be sent to the target VU, other k-vehicle VU are controlled to send the content fragments cached by each to the target VU until the target VU generates the hot content according to the received k content fragments.

Further, the logic instructions in the memory 830 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method for data delivery based on an on-board network provided by the above methods, the method comprising: encoding any hot content into a plurality of fragments by using an erasure code encoding technology, performing redundancy encoding on all fragments to obtain a plurality of content fragments, and respectively caching the content fragments to VU (virtual private units) of each vehicle user in a cluster; each VU stores only one piece of content shard; the clusters are determined by dividing all the VUs according to the positions of the VUs in the vehicle-mounted network and the request information of the VUs for the hot content in the scene; receiving a loading request of any target VU in the vehicle-mounted network for the hot content, and sending content fragments to the target VU; after the number of the content fragments sent to the target VU by the road side unit RSU is k-k', if the content fragments are stopped to be sent to the target VU, other k-vehicle VU are controlled to send the content fragments cached by each to the target VU until the target VU generates the hot content according to the received k content fragments.

In still another aspect, the present invention further provides a non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor is implemented to perform the data issuing method based on the on-vehicle network provided by the above embodiments, the method includes: encoding any hot content into a plurality of fragments by using an erasure code encoding technology, performing redundancy encoding on all fragments to obtain a plurality of content fragments, and respectively caching the content fragments to VU (virtual private units) of each vehicle user in a cluster; each VU stores only one piece of content shard; the clusters are determined by dividing all the VUs according to the positions of the VUs in the vehicle-mounted network and the request information of the VUs for the hot content in the scene; receiving a loading request of any target VU in the vehicle-mounted network for the hot content, and sending content fragments to the target VU; after the number of the content fragments sent to the target VU by the road side unit RSU is k-k', if the content fragments are stopped to be sent to the target VU, other k-vehicle VU are controlled to send the content fragments cached by each to the target VU until the target VU generates the hot content according to the received k content fragments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. The data issuing method based on the vehicle-mounted network is characterized by comprising the following steps of:

encoding any hot content into a plurality of fragments by using an erasure code encoding technology, performing redundancy encoding on all fragments to obtain a plurality of content fragments, and respectively caching the content fragments to VU (virtual private units) of each vehicle user in a cluster; each VU stores only one piece of content shard;

receiving a loading request of any target VU in the vehicle-mounted network for the hot content, and sending content fragments to the target VU;

After the number of the content fragments sent to the target VU by the road side unit RSU is k-k ', if the content fragments are stopped to be sent to the target VU, other k' vehicles VU send the content fragments cached by each to the target VU until the target VU decodes and generates the hot content according to the received k content fragments.

2. The data distribution method based on the on-vehicle network according to claim 1, further comprising, before buffering the content pieces to the respective vehicle users VU in one cluster:

3. The method for transmitting data based on an on-vehicle network according to claim 2, wherein the buffering the content fragments to each vehicle user VU in one cluster respectively includes:

Using V2I link to each NOMA strong user, each NOMA weak P _RSU ＞P _V2V User and each of the V2I usersAnd sending the content fragments.

4. The data distribution method based on the on-vehicle network according to claim 3, wherein the sending, by the other k' VUs, the respective cached pieces of the content to the target VU includes:

when the target VU is a NOMA strong user or a NOMA weak user, any other NOMA strong user or any other NOMA weak user adopts a NOMA communication mode to send the content fragments to the target VU, and the total number of any other NOMA strong user and any other NOMA weak user is k';

accordingly, if the target VU is a V2I user, the roadside unit RSU continues to adopt a V2I link to send the content fragments to the target VU.

5. The data distribution method based on the vehicle-mounted network according to claim 3, further comprising:

the NOMA strong users and the NOMA weak users in the same sub-cluster perform interactive transmission of content fragments in a multiplexing frequency spectrum mode;

the number of VUs in the sub-cluster is greater than or equal to 2.

6. The method for transmitting data based on an on-vehicle network according to claim 1, wherein before encoding any hot content into a plurality of slices by using erasure coding encoding technique, performing redundancy encoding on all slices to obtain a plurality of content slices, and buffering the content slices to each vehicle user VU in a cluster, further comprising:

Constructing a data issuing optimization model by taking the minimum total time delay as a target;

solving the data issuing optimization model to obtain the condition and beta of the user pair composed of the optimal VU _g Optimal power distribution factor matrix in composed user pair, optimal transmission content fragmentation situation matrix between VU, optimal RB allocation situation matrix, optimal transmission power distribution matrix in V2V process, k number and k' numberAn amount of;

wherein beta is _g Indicating that the g-th user pair is a NOMA user pair;

the data issuing optimization model is as follows:

the data issuing optimization modelThe constraint conditions of (2) are:

P _RSU ＞P _V2V (9l)

k-k'≥k' (9m)

wherein B is the total bandwidth of the spectrum resource; beta _g Indicating that the g-th user pair is a NOMA user pair;index of user pair formed for strong and weak users, < ->And g=max { M, N }; h is a _g,S And h _g,W Channel power gains for communication links between strong and weak users in RSU and g-th user pair, respectively, where h _g,S ＝δ _g,S d _g,S ^-α ，h _g,W ＝δ _g,W d _g,W ^-α ，d _g,S 、d _g,W The distances between RSU and strong and weak users in g-th user pair are respectively delta _g,S 、δ _g,W Channel fading coefficients of strong and weak users in the RSU-g user pair are respectively, and alpha is a path loss index; gamma ray _g,S And gamma _g,W The received signal to noise ratios of strong and weak users in the g-th user pair are respectively;

P _RSU is the total power, σ, of the RSU in sending messages to each user pair ² Is additive white gaussian noise; p is p _qi A transmission power at the time of transmitting the content fragment for the qth VU to the ith VU; h is a _qi Channel power gain when content fragmentation is sent to the ith Vus for the qth Vus; g _q'i' Interference channel power gain for the (i) th Vus when transmitting content fragmentation to the (i) th Vus for the (q) th Vus, then p _q'i' For the power at which the q 'th VU transmits content fragmentation to the i' th VU; t (T) _i(pair) For the communication delay of the ith VU in the first phase; the data rate of the ith VU in the first stage is R _i(pair) The communication time delay of the ith VU in the second stage is T _i(V2V) The data rate is R _i(V2V) The method comprises the steps of carrying out a first treatment on the surface of the The optimization variable X is the situation that the VU forms a user pair; optimizing variablesIf the q-th VU receives the content fragment j from the RSU, y is the storage condition of the content fragment by the VU after the content fragment is issued by the RSU _qj =1, otherwise y _qj =0; the optimization variable beta is beta _g A user centering power distribution factor matrix is formed; the optimization variable Z is a matrix for representing the situation of transmitting content fragmentation among the Vus; the optimization variable A is an RB allocation condition matrix; the optimization variable P is a transmission power distribution matrix in the V2V process; constraint (9 a) indicates that each VU can only be in one user pair; constraint (9 b) indicates that there is at most one strong user in each user pair; constraint (9 c) indicates that there is at most one weak user in each user pair; constraint (9 d) indicates that if the g-th user pair is a NOMA pair, the transmit power allocation factor is greater than 0 small 1/2; constraint (9 e) indicates that the first-stage content delivery stage is to be at t ₁ Completing in time; constraint (9 f) indicates that the second phase V2V communication phase is to be at t ₂ Completing in time; the constraint (9 g) indicates that each VU can only be in one sub-cluster when sending content fragments; constraint (9 h) indicates that there are at most two VUs in each sub-cluster; constraint (9 i) indicates that the transmit power of VU of the second stage V2V process is less than P _V2V The method comprises the steps of carrying out a first treatment on the surface of the Constraint (9 j) indicates that the VU is capable of receiving content slices from k' users in the second phase to obtain enough content slices for decoding to obtain complete content; the constraint (9 k) indicates the first stage VU transmit power P _RSU Transmit power P with VU of second phase V2V procedure _V2V The sum is->The constraint (9 l) indicates the first stage VU transmit power P _RSU Transmit power P of VU greater than second stage V2V process _V2V The method comprises the steps of carrying out a first treatment on the surface of the The constraint (9 m) indicates that the number of content slices delivered in the first phase is greater than the number of content slices received in the second phase so that the VU can obtain the required number of slices.

7. The method for distributing data based on the vehicle network according to claim 6, wherein the solving the data distribution optimization model includes:

Acquiring position information of each VU and a maximum value of transmitting power, wherein the maximum value of transmitting power

Dividing all VU in the cluster into strong users and weak users according to channel conditions, calculating the communication time delay sum of NOMA user pairs formed by each strong user and each weak user, and determining the weight of the strong user and the weak user according to the communication time delay sum of all the strong users and the communication time delay sum of all the weak users;

based on the hungarian algorithm, the user who satisfies the traversal rate and maximum of all VUs will be determinedThe strong and weak user weight under the paired condition is used as the maximum user weight to match x ₁ The method comprises the steps of carrying out a first treatment on the surface of the Matching x according to user maximum weight ₁ Calculating the time delay t of each VU ₁ ；

Taking two groups of same VUs as two point sets, and assigning an edge weight value to infinity when the two point sets are the same VU; when the two point sets are different VUs, setting the edge weight as the time delay of the two VUs for transmitting the message;

acquiring minimum weight matching x by using many-to-many matching algorithm ₂ To determine all VU reception delays and minimum V2V link pairings;

taking all the VUs as a set, and calculating time delay when any two VUs in the set multiplex resources; if the time delay is larger than the sum of the time delays when each VU uses the resources independently, constructing an edge between points corresponding to every two VUs, and thus obtaining a traditional graph;

Coloring the traditional graph based on a hypergraph coloring algorithm, wherein two points with edges are colored with different colors, and points corresponding to the same resource are colored with the same color;

acquiring the resource multiplexing condition x according to the coloring result ₃ And according to x ₃ Calculating the V2V communication time delay t of each VU ₂ ；

According to t ₁ 、t ₂ Calculating the total time delay t of all VU in the issuing stage and the communication stage;

method for obtaining maximum value based on single variable combined with x ₁ 、x ₂ 、x ₃ P pair P _V2V And P _RSU Optimization is performed so that the total delay t is minimized.

8. A data issuing device based on a vehicle-mounted network, comprising:

the content issuing unit is used for encoding any hot content into a plurality of fragments by using an erasure code encoding technology, performing redundancy encoding on all the fragments to obtain a plurality of content fragments, and respectively caching the content fragments to VU (virtual Unit) of each vehicle user in a cluster; each VU stores only one piece of content shard; the clusters are determined by dividing all the VUs according to the positions of the VUs in the vehicle-mounted network and the request information of the VUs for the hot content in the scene;

a first data issuing unit, configured to receive a loading request of any target VU in the vehicle-mounted network for the hot content, and send a content fragment to the target VU, so that the target VU decodes the content fragment to generate the hot content;

And the second data issuing unit is used for controlling other k 'vehicles VU to send the content fragments cached by each to the target VU if the content fragments are stopped to be sent to the target VU after the RSU sends the content fragments with the number of k-k' to the target VU, and the target VU generates the hot content according to the received k content fragments.

9. A non-transitory computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the data distribution method steps based on an on-board network according to any one of claims 1 to 7.