Information processing method and electronic equipment
By constructing a multi-model optimization system for the caching and update strategies of edge caching devices, the problem of insufficient content updates in edge caching technology is solved, thus ensuring content freshness and reducing network load.
Patent Information
- Application Number
- CN202510907996.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-02
- Publication Date
- 2025-10-17
AI Technical Summary
Existing edge caching technologies have failed to update content effectively, leading to network congestion and unmet user needs.
By constructing download cost models, aging cost models, and update cost models, and combining the cache capacity of edge caching devices with the backhaul link transmission capacity, caching and update strategies are optimized to ensure content freshness.
It effectively reduces the load on the backhaul link, lowers the latency for users to access content, and improves the user experience.
Smart Images
Figure CN120812129A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of wireless communication technology, and more particularly to an information processing method and electronic equipment. Background Art
[0002] With the rapid development of smart devices, data traffic has exploded. Due to limited backhaul capacity, acquiring large amounts of data traffic will lead to severe network congestion and unacceptable transmission delays.
[0003] To reduce the throughput overhead of base stations, device-to-device (D2D) communication technology has become a promising technology in wireless communication networks. With D2D technology, user devices can transmit content directly to each other without having to obtain it from the base station, thereby reducing the burden on the base station, reducing user transmission latency, and improving spectrum efficiency.
[0004] However, most of the research on edge caching focuses on traditional static content, without considering the update of edge cache content, and cannot meet users' demand for content. Summary of the Invention
[0005] The purpose of the present invention is to provide an information processing method and electronic device to address the deficiencies in the above-mentioned prior art, so as to update the content cached in the edge cache device, ensure the freshness of the content, and meet the user's demand for content.
[0006] To achieve the above objectives, the technical solutions adopted in the embodiments of the present invention are as follows: In a first aspect, an embodiment of the present invention provides an information processing method, the method comprising: caching corresponding contents in the plurality of edge cache devices respectively according to user preferences of a plurality of content requesting users for a plurality of contents, content popularity of the plurality of contents, and pairing relationships between the plurality of content requesting users and a plurality of edge cache devices, wherein the plurality of contents are stored in a core network; constructing a total cost model based on the download cost model, the aging cost model, and the update cost model of the plurality of contents in the target time slot; Solving the total cost model based on the cache capacity of each edge cache device, the backhaul link transmission capacity between the macro base station and the core network, and the maximum tolerable information age, to determine a cache policy and an update policy for each content in each edge cache device at the target time slot, wherein the cache policy is used to indicate whether to cache each content in each edge cache device at the target time slot, and the update policy is used to indicate whether to update each content in each edge cache device at the target time slot; According to the cache strategy and the update strategy, the content cached in each edge cache device is updated.
[0007] Optionally, before the corresponding content is cached in the plurality of edge cache devices respectively according to the user preference of the plurality of content request users for the plurality of contents, the content popularity of the plurality of contents, and the pairing relationship between the plurality of content request users and the plurality of edge cache devices, the method further comprises: The first preference degree of the plurality of edge cache devices for the plurality of contents and the second preference degree of the plurality of content request users for the plurality of contents are calculated respectively. According to the first preference degree and the second preference degree, the preference similarity between the plurality of content request users and the plurality of edge cache devices is calculated. According to the preference similarity between the plurality of content request users and the plurality of edge cache devices, and the distance between the plurality of content request users and the plurality of edge cache devices, the plurality of content request users and the plurality of edge cache devices are paired one by one.
[0008] Optionally, the first preference degree of the plurality of edge cache devices for the plurality of contents and the second preference degree of the plurality of content request users for the plurality of contents are calculated respectively, comprising: According to the score of each content under each theme, the normalized feature value of each content under each theme is calculated. According to the correlation between each content request user, each edge cache device and the plurality of themes, the normalized feature value of each content request user and each edge cache device under each theme is calculated respectively. According to the normalized feature value of each content under each theme and the normalized feature value of each edge cache device under each theme, the first preference degree of each edge cache device for each content is calculated. According to the normalized feature value of each content under each theme and the normalized feature value of each content request user under each theme, the second preference degree of each content request user for each content is calculated.
[0009] Optionally, the plurality of content request users and the plurality of edge cache devices are paired one by one according to the preference similarity between the plurality of content request users and the plurality of edge cache devices and the distance between user devices, comprising: According to the content request users contained in the coverage range of each edge cache device, the connection between each edge cache device and the content request users in the corresponding coverage range is formed to form a bipartite graph comprising the plurality of edge cache devices, the plurality of content request users and a plurality of connection edges. According to the preference similarity between the plurality of content request users and the plurality of edge cache devices, and the distance between the plurality of content request users and the plurality of edge cache devices, the weight of the plurality of connection edges is calculated; According to the bipartite graph and the weight of the plurality of connection edges, the Hungarian algorithm is used to determine the edge cache device matched by the each content request user.
[0010] Optionally, the download cost model is constructed in the following way: According to the request indication variable of the plurality of contents requested by the user, the cache indication variable of the plurality of contents, the size of the plurality of contents, the cost of downloading unit data from the edge cache device, and the cost of downloading the unit data from the core network, the download cost model is constructed.
[0011] Optionally, the aging cost model is constructed in the following way: According to the update indication variable of the plurality of contents, the information age of the plurality of contents in the target time slot is calculated; According to the cache indication variable of the plurality of contents, the information age of the plurality of contents in the target time slot, and the size of the plurality of contents, the aging cost model is constructed.
[0012] Optionally, the update cost model is constructed in the following way: According to the update indication variable of the plurality of contents, the cost of downloading the unit data from the core network, and the size of the plurality of contents, the update cost model is constructed.
[0013] Optionally, the total cost model is solved according to the cache capacity of the each edge cache device, the backhaul link transmission capacity between the macro base station and the core network, and the maximum tolerable information age, to determine the cache strategy and update strategy of the each content in the each edge cache device in the target time slot, including: The total cost model is solved according to the cache capacity of the each edge cache device, the backhaul link transmission capacity, and the maximum tolerable information age, to obtain the initial cache value and initial update value of the plurality of contents in the each edge cache device; The initial update values of the plurality of contents are sorted, and the content update quantity under the constraint of the backhaul link transmission capacity is calculated; The plurality of target contents corresponding to the content update quantity are determined as the to-be-updated contents of the each edge cache device in the target time slot; The to-be-updated contents are determined as the first to-be-cached contents; ranking initial cache values of other contents in the plurality of contents except the first to-be-cached content, determining a second to-be-cached content satisfying a maximum cache capacity of each edge cache device; The cache strategy includes the first to-be-cached content and the second to-be-cached content, and the update strategy includes the to-be-updated content.
[0014] Optionally, the method further includes: judging whether the target content is contained in a target edge cache device corresponding to a target user according to an acquisition request for target content sent by the target user; if the target content is contained in the target edge cache device, sending the target content to the target user through the target edge cache; if the target content is not contained in the target edge cache device, acquiring the target content from the core network through the macro base station and sending the target content to the target user.
[0015] In a second aspect, an information processing apparatus is also provided, and the apparatus includes: a content cache module configured to cache corresponding contents in a plurality of edge cache devices according to user preferences of a plurality of content request users for a plurality of contents, content popularity of the plurality of contents, and a pairing relationship between the plurality of content request users and the plurality of edge cache devices, wherein the plurality of contents are stored in a core network; a model construction module configured to construct a total cost model according to a download cost model, an aging cost model, and an update cost model of the plurality of contents in a target time slot; a strategy calculation module configured to solve the total cost model according to a cache capacity of each edge cache device, a backhaul link transmission capacity between a macro base station and the core network, and a maximum tolerable information age, to determine a cache strategy and an update strategy of each content in each edge cache device in the target time slot, wherein the cache strategy is used to indicate whether to cache each content in each edge cache device in the target time slot, and the update strategy is used to indicate whether to update each content in each edge cache device in the target time slot; a content update module configured to update contents cached in each edge cache device according to the cache strategy and the update strategy.
[0016] Optionally, the apparatus further includes: a preference degree calculation module configured to calculate a first preference degree of a plurality of edge cache devices for the plurality of contents and a second preference degree of a plurality of content request users for the plurality of contents, respectively; a similarity calculation module, configured to calculate a preference similarity between the plurality of content request users and the plurality of edge cache devices according to the first preference degree and the second preference degree; a pairing module, configured to pair the plurality of content request users and the plurality of edge cache devices one by one according to the preference similarity between the plurality of content request users and the plurality of edge cache devices and distances between the plurality of content request users and the plurality of edge cache devices.
[0017] Optionally, the preference degree calculation module is specifically configured to calculate a normalized feature value of each content in each theme according to a score of the each content in the plurality of themes; calculate a normalized feature value of each content request user and each edge cache device in the each theme according to a correlation between the each content request user and the each edge cache device and the plurality of themes; calculate a first preference degree of the each edge cache device for the each content according to the normalized feature value of the each content in the each theme and the normalized feature value of the each edge cache device in the each theme; and calculate a second preference degree of the each content request user for the each content according to the normalized feature value of the each content in the each theme and the normalized feature value of the each content request user in the each theme.
[0018] Optionally, the pairing module is specifically configured to connect each edge cache device and a content request user in a corresponding coverage range of the each edge cache device to form a bipartite graph including the plurality of edge cache devices, the plurality of content request users and a plurality of connection edges according to the content request users included in the coverage range of the each edge cache device; calculate a weight of the plurality of connection edges according to the preference similarity between the plurality of content request users and the plurality of edge cache devices and the distances between the plurality of content request users and the plurality of edge cache devices; and determine an edge cache device matched with the each content request user by using a Hungarian algorithm according to the bipartite graph and the weight of the plurality of connection edges.
[0019] Optionally, the download cost model is constructed in the following manner: The download cost model is constructed according to a request indication variable of the plurality of contents requested by the user, a cache indication variable of the plurality of contents, sizes of the plurality of contents, a cost of downloading unit data from the edge cache device and a cost of downloading the unit data from the core network.
[0020] Optionally, the aging cost model is constructed in the following manner: An information age of the plurality of contents in the target time slot is calculated according to an update indication variable of the plurality of contents. According to the cache indication variables of the plurality of contents, the information ages of the plurality of contents in the target time slot and the sizes of the plurality of contents, the aging cost model is constructed.
[0021] Optionally, the update cost model is constructed in the following manner: According to the update indication variables of the plurality of contents, the cost of downloading the unit data from the core network and the sizes of the plurality of contents, the update cost model is constructed.
[0022] Optionally, the strategy calculation module is specifically configured to: according to the cache capacities of the each edge cache device, the backhaul link transmission capacities and the maximum tolerable information ages, solve the total cost model to obtain initial cache values and initial update values of the plurality of contents in the each edge cache device; sort the initial update values of the plurality of contents, calculate the content update quantity under the constraint of the backhaul link transmission capacities; determine a plurality of target contents corresponding to the content update quantity as the to-be-updated contents of the each edge cache device in the target time slot; determine the to-be-updated contents as first to-be-cached contents; sort the initial cache values of the contents other than the first to-be-cached contents in the plurality of contents, and determine contents satisfying the maximum cache capacities of the each edge cache device as second to-be-cached contents; the cache strategy includes the first to-be-cached contents and the second to-be-cached contents, and the update strategy includes the to-be-updated contents.
[0023] Optionally, the apparatus further includes: The content transmission module is configured to: according to an acquisition request for target content sent by a target user, determine whether the target content is contained in a target edge cache device corresponding to the target user; if the target content is contained in the target edge cache device, send the target content to the target user through the target edge cache; and if the target content is not contained in the target edge cache device, acquire the target content from the core network through the macro base station and send the target content to the target user.
[0024] In a third aspect, an electronic device is provided, including a processor, a memory and a communication bus, wherein the memory stores machine readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory through the communication bus, and the processor executes the machine readable instructions to implement the steps of the information processing method in any of the first aspect.
[0025] The present application has the following advantages: The information processing method and the electronic device provided by the application solve the caching strategy and the updating strategy under the constraints of the caching capacity of each edge caching device, the transmission capacity of the backhaul link between the macro base station and the core network and the maximum tolerable information age, so as to update the content in each edge caching device, ensure the freshness of the cached content in each edge caching device, and make the cached content in the edge caching device meet the content demand of the paired content request user, greatly reduce the load of the backhaul link, reduce the time delay of the user in obtaining the content, and improve the user experience. BRIEF DESCRIPTION OF DRAWINGS
[0026] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the application, and therefore should not be regarded as a limitation to the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.
[0027] Figure 1 The application scenario of the embodiment of the application is shown in the figure. Figure 2 The flow of the information processing method provided by the embodiment of the application is shown in the figure. Figure One Figure 3 The flow of the information processing method provided by the embodiment of the application is shown in the figure. Figure Two Figure 4 The flow of the information processing method provided by the embodiment of the application is shown in the figure. Figure Three Figure 5 The flow of the information processing method provided by the embodiment of the application is shown in the figure. Figure Four Figure 6 The cost comparison of the information transmission method provided by the embodiment of the application is shown in the figure. Figure One Figure 7 The cost comparison of the information transmission method provided by the embodiment of the application is shown in the figure. Figure Two Figure 8 The structure of the information processing device provided by the embodiment of the application is shown in the figure. Figure 9 The schematic diagram of the electronic device provided by the embodiment of the application is shown in the figure. DETAILED DESCRIPTION
[0028] With reference to the drawings and embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present application, but not all embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.
[0029] The advantages and effects of the present application can be easily understood by those skilled in the art from the disclosure of the present application. The present application can also be implemented or applied by different specific embodiments. The details in the specification can also be modified and changed based on different views and applications without departing from the spirit of the present application. It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present application in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.
[0030] The drawings are only used for illustrative explanation, and the representation is only a schematic diagram, not a physical diagram, and cannot be understood as a limitation of the present application. For those skilled in the art, it is understandable that some known structures and their descriptions in the drawings can be omitted.
[0031] In order to better understand the scheme of the present application, the application scenario of the information processing method of the present application will be introduced first.
[0032] Figure 1 The application scenario of the embodiments of the present application is shown in FIG. 1, wherein a D2D wireless cache network is constructed, and the wireless cache network includes a core network, a macro base station, a content request user and an edge cache device. The edge cache device is a content provider, and communicates with the content request user through a D2D link. Figure 1
[0033] The terminal device used by the content request user is a user device without cache support, and the edge cache device is a user device with cache support.
[0034] Based on the above application scenario, the specific implementation of the information processing method provided by the present application will be described in combination with the embodiments.
[0035] Figure 2 The flow of the information processing method provided by the embodiments of the present application is shown in FIG. 2, and the information processing method can include the following steps. Figure One Figure 2 The information processing method can include the following steps. Step 101, according to the user preference of the plurality of content request users for the plurality of contents, the content popularity of the plurality of contents, and the pairing relationship between the plurality of content request users and the plurality of edge cache devices, respectively cache the corresponding contents in the plurality of edge cache devices, wherein the plurality of contents are stored in the core network.
[0036] Specifically, as known from the foregoing, the terminal device used by the content request user is a user device that does not support caching, and the edge cache device is a user device that supports caching. Pairing the content request user and the edge cache device is actually pairing the edge cache device and the terminal device corresponding to the content request user, so as to pair each content request user with a unique corresponding edge cache device.
[0037] According to the user preference of each content request user for the plurality of contents and the popularity of the plurality of contents, the user request probability of each content request user for the plurality of contents is calculated, the plurality of contents are sorted according to the user request probability, and part of the contents ranked in the front are cached in the edge cache device paired with each content request user, wherein the number of cached contents can be a preset number or can be determined according to the cache capacity of the edge cache device.
[0038] It should be noted that since the user preference of each content request user for the plurality of contents is different, the contents cached in the edge cache device paired with each content request user are different.
[0039] After caching some contents in the edge cache device, when the content requested by the content request user is cached in the edge cache device, the edge cache device can push the requested content to the content request user, without the need to obtain the requested content from the core network through the base station.
[0040] Step 102, constructing a total cost model according to the download cost model, the aging cost model and the update cost model of the plurality of contents in the target time slot.
[0041] Due to real-time updating of the contents and changes in user preferences, the contents cached in the edge cache device may be outdated or no longer favored by users. At this time, if the contents are still retained in the edge cache device, it not only cannot reduce the transmission pressure of the base station, but also wastes the cache capacity of the edge cache device. Therefore, it is necessary to update the contents cached in the edge cache device in a timely manner.
[0042] Specifically, the download cost model is used to represent the download cost of each content at the target time slot, the aging cost model is used to represent the aging cost of the outdated content cached in the edge cache device, the update cost model is used to represent the cost of updating the content cached in the edge cache device, and the total cost model is the sum of the download cost model, the aging cost model and the update cost model. The total cost model is used to represent the cost of caching and updating each content.
[0043] At step 103, the total cost model is solved according to the caching capacity of each edge cache device, the backhaul link transmission capacity between the macro base station and the core network, and the maximum tolerable information age, to determine the caching strategy and the updating strategy of each content in each edge cache device at the target time slot. The caching strategy is used to indicate whether each content is cached in each edge cache device at the target time slot, and the updating strategy is used to indicate whether each content is updated in each edge cache device at the target time slot.
[0044] Specifically, since the caching and updating of each content are limited by the caching capacity of the edge cache device, the backhaul link transmission capacity between the macro base station and the core network, and the maximum tolerable information age, the optimization problem of minimizing the total cost of the total cost model is constructed by taking the caching capacity of each edge cache device, the backhaul link transmission capacity between the macro base station and the core network, and the maximum tolerable information age as the constraints of the total cost model. The optimization problem of minimizing the total cost is solved to obtain the caching strategy and the updating strategy of each content in each edge cache device at the target time slot, wherein the caching strategy and the updating strategy are joint strategies, that is, the content updated in the edge cache device must be cached in the edge cache device, but the cached content is not necessarily updated.
[0045] At step 104, the content cached in each edge cache device is updated according to the caching strategy and the updating strategy.
[0046] Specifically, updating the content cached in the edge cache device includes determining whether to update the latest content re-downloaded into the edge cache device according to the updating strategy, and determining whether to delete the content cached in the edge cache device or still keep it in the cache according to the caching strategy.
[0047] If the updating strategy indicates that a content is updated, the caching strategy also needs to execute that the content is cached.
[0048] In a possible implementation manner, Figure 3 The flow of the information processing method provided by the embodiment of the present application Figure Two As Figure 3As shown, before the step 101, the method can further include: Step 201, respectively calculating a first preference degree of the plurality of edge cache devices for the plurality of contents, and a second preference degree of the plurality of content request users for the plurality of contents.
[0049] Step 202, calculating a preference similarity between the plurality of content request users and the plurality of edge cache devices according to the first preference degree and the second preference degree.
[0050] Step 203, pairing the plurality of content request users and the plurality of edge cache devices one by one according to the preference similarity between the plurality of content request users and the plurality of edge cache devices, and a distance between the plurality of content request users and the plurality of edge cache devices.
[0051] In the embodiment, the first preference degree of each edge cache device for the plurality of contents is determined according to the request situation of the user corresponding to each edge cache device for the plurality of contents, and the second preference degree of each content request user for the plurality of contents is determined according to the request situation of each content request user for the plurality of contents. The first preference degree of each edge cache device for the plurality of contents can be determined according to the request times of the user corresponding to each edge cache device for the plurality of contents within a preset time period, and the second preference degree of each content request user for the plurality of contents can be determined according to the request times of each content request user for the plurality of contents within the preset time period.
[0052] The first preference degree of each edge cache device for the plurality of contents can be represented as a preference distribution vector of each edge cache device, i.e. The second preference degree of each content request user for the plurality of contents can be represented as a preference distribution vector of each content request user, i.e. .
[0053] The pairing of the content request user and the edge cache device is actually the pairing between the content request user and the user corresponding to the edge cache device, and each content request user has a uniquely paired edge cache device.
[0054] According to the first preference degree of the plurality of edge cache devices for the plurality of contents and the second preference degree of the plurality of content request users for the plurality of contents, the cosine similarity is used to calculate the preference similarity between any one edge cache device and one content request user.
[0055] The calculation formula of preference similarity can be expressed as: (1) Request users based on multiple content With multiple edge cache devices Similarity of preferences between multiple content requesting users With multiple edge cache devices The distance between each content request user is determined The edge cache device whose distance is within a preset range and whose preference similarity is high is matched to each content requesting user, wherein the content requesting user and the edge cache device are paired one-to-one.
[0056] In one possible implementation, the process of calculating the first preferences of the plurality of edge cache devices for the plurality of contents and the second preferences of the plurality of content requesting users for the plurality of contents in step 201 may include: Based on the scores of each content under multiple topics, the normalized characteristic value of each content under each topic is calculated; based on the correlation between each content requesting user and each edge cache device and the multiple topics, the normalized characteristic value of each content requesting user and each edge cache device under each topic is calculated respectively; based on the normalized characteristic value of each content under each topic and the normalized characteristic value of each edge cache device under each topic, the first preference of each edge cache device for each content is calculated; based on the normalized characteristic value of each content under each topic and the normalized characteristic value of each content requesting user under each topic, the preference of each content requesting user for each content is calculated.
[0057] In this embodiment, the first preference and the second preference are calculated in the same manner. The following describes the method for calculating the user's preference for content by taking the calculation of the second preference as an example.
[0058] Specifically, each content has a theme, which can be regarded as a feature set that captures different contents in the content library. Content belongs to For example, take this topic Defined as content In Features The scores of all topics are normalized to get the topic Previous content The normalized eigenvalue of .
[0059] Similarly, define To indicate content request user On the topic The score in , the normalized feature value is .
[0060] Assume each content request user has a personalized preference, set to represent the preference of content request user to content , written as (2) Therefore, the normalized preference degree of content request user to content can be expressed as: (3) Wherein, the score of content in feature , the score of content request user u in theme is priori knowledge, the embodiment does not limit the specific implementation of calculating the score of content in feature , the score of content request user u in theme .
[0061] In one possible implementation, the above step 203 can include the following steps: According to the preference similarity between the plurality of content request users and the plurality of edge cache devices, and the distance between the plurality of content request users and the plurality of edge cache devices, the plurality of content request users and the plurality of edge cache devices are one-to-one matched.
[0062] In this embodiment, a weighted bipartite graph is defined, wherein is the set of content request users, is the set of edge cache devices covering at least one content request user, is the set of edges, is the set of edge weights, which is the weight of the connection edge formed by connecting the content request user and the edge cache device.
[0063] According to the preference similarity and the distance between each content request user and each edge cache device, a social awareness rate between each content request user and each edge cache device is calculated, and the social awareness rate represents a weight of a connection edge formed by the connection of the content request user and the edge cache device.
[0064] If the content request user is within the coverage of the edge cache device , there is an edge, denoted as . The weight of the edge is denoted as , which is actually a social awareness rate between the edge cache device and the content request user , and the calculation formula of the social awareness rate can be represented as: (4) wherein and represent weight parameters of the preference similarity and the distribution distance respectively, and satisfy ; is the distance between the edge cache device k and the content request user u .
[0065] According to the number of edge cache devices and the number of content request users, virtual points are added in the bipartite graph , so that the number of points representing the edge cache devices and the number of points representing the content request users are equal, and the edges are connected with the added points to form virtual edges, so as to convert the bipartite graph into a regular bipartite graph, denoted as .
[0066] The Hungarian algorithm (Kuhn-Munkres, KM) is used to perform one-to-one matching between the content request users and the edge cache devices in the regular bipartite graph, and finally the virtual points and the virtual edges are deleted, and the paired user pair set is defined as .
[0067] In a possible implementation manner, the download cost model is constructed in the following manner: According to the request indication variable of the user requesting multiple contents, the cache indication variable of the multiple contents, the size of the multiple contents, the cost of downloading unit data from the edge cache device, and the cost of downloading unit data from the core network, the download cost model is constructed.
[0068] In this embodiment, in the target time slot t , the download cost model of the content request user requesting the content f can be represented as: (5) wherein, denotes an indication variable of a user requesting the th content, and takes value ; denotes a binary cache indication variable, 0 represents not caching the content , and 1 represents caching the content ; denotes a cache file size; denotes a cost of downloading a unit data from the edge cache device; denotes a cost of downloading a unit data from the core network, and the unit data may be, for example, information of a preset size.
[0069] In a possible implementation, the aging cost model is constructed in the following manner: According to the update indication variables of the plurality of contents, the information ages of the plurality of contents in the target time slot are calculated; and according to the cache indication variables of the plurality of contents, the information ages of the plurality of contents in the target time slot, and the sizes of the plurality of contents, the aging cost model is constructed.
[0070] In the embodiment, in the target time slot , each content cached by the edge cache device incurs an aging cost due to being outdated, wherein the information age of each content increases and the aging cost of each content increases as the time each content remains in the edge cache device is longer.
[0071] The aging cost model of the aging cost incurred by the outdated content cached in the edge cache device can be expressed as: (6) wherein, denotes an information age of the cached content of each edge cache device in the time slot ; denotes a cost function of the AoI, and is a cost function factor.
[0072] According to the definition of the information age (Age of Information, AoI), the AoI evolution model can be expressed as: (7) wherein, denotes a binary update indication variable, denotes that the content is updated in the edge cache device in the target time slot , and the information age of the updated content is 1, Indicates the target time slot t Caching devices at the edge Content No updates, no updated content The information age is the age in the previous time slot The information age is based on plus 1.
[0073] In one possible implementation, the update cost model is constructed as follows: An update cost model is constructed according to update indicator variables of the plurality of contents, a cost of downloading unit data from a core network, and sizes of the plurality of contents.
[0074] In this embodiment, in the target time slot t ,The update cost model for updating the content in the cache in the edge cache device can be expressed as: (8) In one possible implementation, Figure 4 The process of the information processing method provided by the embodiment of the present invention Figure Three ,like Figure 4 As shown, the above step 103 solves the total cost model based on the cache capacity of each edge cache device, the transmission capacity of the backhaul link between the macro base station and the core network, and the maximum tolerable information age, and determines the caching strategy and update strategy for each content in each edge cache device at the target time slot. The process may include: Step 301: Solve the total cost model based on the cache capacity, backhaul link transmission capacity, and maximum tolerable information age of each edge cache device to obtain initial cache values and initial update values of multiple contents in each edge cache device.
[0075] Specifically, the construction minimizes the total cost The optimization problem can be expressed as: in, Represents the set of cache states of all edge cache devices; Represents the set of update status of all edge cache devices; Indicates the maximum tolerated information age threshold; and They represent the cache capacity of the edge cache device and the transmission capacity of the backhaul link respectively.
[0076] The transformation method of Taylor expansion, continuous convex approximation and McCormick envelope is used to transform the non-convex problem into a convex problem, which is then solved by the CVX solver to obtain the optimal caching and updating decisions.
[0077] Specifically, the SCA method is used to approximate the non-convex terms in the objective function to convex functions. In order to simplify the calculation, the and The subscript of , the first-order Taylor expansion expression for the non-convex term of the objective function can be expressed as: in, and Represents the AoI function pair and The partial derivative of and For the The optimal content caching and update decision variables obtained by iteration, A 0 is the initial AoI.
[0078] Use constant express , the objective function of the original optimization problem is transformed into: Introducing auxiliary variables To decouple the bilinear product of cache variables and update variables. Due to the non-convex constraints brought by the introduction of auxiliary variables, the McCormick envelope is used to convert the non-convex constraints Transformed into a convex constraint of the form: Therefore, the optimization problem of minimizing the total system cost can be transformed into: in, A collection of auxiliary variables representing all edge cache devices; , and They represent the continuous cache, update and auxiliary variables after relaxation respectively.
[0079] The above convex problem is solved by the CVX solver to obtain the optimal initial cache value after relaxation and the initial update value , initial cache value and the initial update value Unique to each edge cache device.
[0080] Step 302: Sort the initial update values of multiple contents and calculate the number of content updates under the transmission capacity constraint of the backhaul link.
[0081] Specifically, the initial update value of multiple contents The descending order is sorted, and the number of content updates that can be updated is determined according to the constraint of the backhaul link transmission capacity and the update size of the plurality of contents.
[0082] Step 303, determine the plurality of target contents corresponding to the content update quantity as the to-be-updated contents of each edge cache device under the target time slot.
[0083] Specifically, according to the sorting of the initial update values of the plurality of contents , the update indication variable of the content corresponding to the content update quantity is set to 1, and the update indication variables of other contents are set to 0. The content with the update indication variable of 1 is the to-be-updated content of the corresponding edge cache device under the target time slot.
[0084] Step 304, determine the to-be-updated content as the first to-be-cached content.
[0085] Specifically, since the to-be-updated content is necessarily cached in the edge cache device, when a content is the content that needs to be updated in the target time slot, the content still needs to be cached in the edge cache device in the target time slot. The cache indication variable corresponding to the to-be-updated content is set to 1, and the to-be-updated content is determined as the first to-be-cached content.
[0086] Step 305, sort the initial cache values of the contents other than the first to-be-cached content in the plurality of contents, and determine the content satisfying the maximum cache size of each edge cache device as the second to-be-cached content; the cache strategy includes: the first to-be-cached content and the second to-be-cached content, and the update strategy includes: the to-be-updated content.
[0087] Specifically, the initial cache values of the remaining contents other than the first to-be-cached content in the plurality of contents are sorted in descending order, and the second to-be-cached content that does not exceed the maximum cache capacity is determined according to the maximum cache capacity of the edge cache device and the size of the first to-be-cached content and the size of the remaining contents. The cache indication variable of the second to-be-cached content is set to 1, and the cache indication variables of other contents are set to 0.
[0088] In one possible implementation manner, Figure 5 the flow of the information processing method provided by the embodiment of the present application Figure Four , as shown in the figure Figure 5 , the information processing method can further include: Step 401, according to the acquisition request for the target content sent by the target user, judge whether the target content is contained in the target edge cache device corresponding to the target user.
[0089] Step 402, if the target edge cache device contains the target content, the target edge cache sends the target content to the target user.
[0090] Step 403: If the target edge cache device does not contain the target content, the target content is obtained from the core network through the macro base station and sent to the target user.
[0091] Specifically, the execution subject of steps 401 to 403 can be an edge cache device in the D2D wireless cache network or a relay device. If it is a relay device, the relay device needs to record the correspondence between content demand users and edge cache devices.
[0092] The target user sends a request to obtain the target content to the target edge cache device (or through a relay device). The target edge cache device determines whether the target content is cached. If the target content is cached, the target content is sent to the target user. If the target content is not cached, a request to obtain the target content can be sent to the core network through the macro base station. The core network can directly send the target content to the target user, or send the target content to the target user through the target edge cache device and cache the target content in the target edge cache device.
[0093] The information processing method provided by the embodiment of the present invention is based on a total cost model constructed based on a download cost model, an aging cost model, and an update cost model. The cache strategy and the update strategy are solved under the constraints of the cache capacity of each edge cache device, the transmission capacity of the backhaul link between the macro base station and the core network, and the maximum tolerated information age, so as to update the content in each edge cache device, ensure the freshness of the cached content in each edge cache device, so that the cached content in the edge cache device can meet the content needs of the paired content requesting users, greatly reduce the load of the backhaul link, reduce the delay for users to obtain content, and improve the user experience.
[0094] For example, Figure 6 Cost comparison of the information transmission method provided by the embodiment of the present invention Figure One ,like Figure 6 As shown, it can be seen that when the cache capacity of the edge cache device is the same, the information transmission method provided by the embodiment of the present invention greatly reduces the total system cost compared with the prior art.
[0095] In addition, as the cache capacity of the edge cache device increases, content request users request more requests from the edge cache device to provide services. However, since the return capacity of content updates is fixed, the more content is cached, the higher the cache aging cost, and the total cost will also increase.
[0096] For example, Figure 7 Cost comparison of the information transmission method provided by the embodiment of the present invention Figure Two ,like Figure 7As shown, it can be seen that, when the number of users is the same, the total system cost of the information transmission method provided by the embodiment of the present invention is greatly reduced compared with the prior art.
[0097] In addition, as the number of users increases, the total number of user requests will increase, which will increase the total cost of the system.
[0098] On the other hand, an embodiment of the present invention further provides an information processing device. Figure 8 A schematic diagram of the structure of the information processing device provided in the embodiment of the present application is shown in FIG. Figure 8 As shown, the device may include: A content caching module 501 is configured to cache corresponding content in a plurality of edge caching devices based on user preferences of a plurality of content requesting users for a plurality of content, content popularity of the plurality of content, and pairing relationships between the plurality of content requesting users and a plurality of edge caching devices, wherein the plurality of content is stored in a core network; A model building module 502 is configured to build a total cost model based on a download cost model, an aging cost model, and an update cost model of multiple contents in a target time slot; A policy calculation module 503 is configured to solve a total cost model based on the cache capacity of each edge cache device, the transmission capacity of the backhaul link between the macro base station and the core network, and the maximum tolerable information age, and determine a cache policy and an update policy for each content in each edge cache device at a target time slot. The cache policy indicates whether each content is cached in each edge cache device at the target time slot, and the update policy indicates whether each content is updated in each edge cache device at the target time slot. The content updating module 504 is configured to update the content cached in each edge cache device according to the cache policy and the update policy.
[0099] Optionally, the device may further include: a preference calculation module, configured to respectively calculate first preferences of multiple edge cache devices for multiple contents, and second preferences of multiple content requesting users for multiple contents; a similarity calculation module, configured to calculate preference similarities between a plurality of content requesting users and a plurality of edge cache devices based on the first preference and the second preference; The pairing module is used to pair multiple content requesting users with multiple edge cache devices one by one according to the preference similarities between the multiple content requesting users and the multiple edge cache devices and the distances between the multiple content requesting users and the multiple edge cache devices.
[0100] Optionally, the preference degree calculation module is specifically configured to calculate a normalized feature value of each content in each theme according to the score of each content in the multiple themes; calculate a normalized feature value of each content request user and each edge cache device in each theme according to the correlation between the multiple themes and each content request user and each edge cache device; and calculate a first preference degree of each edge cache device for each content according to the normalized feature value of each content in each theme and the normalized feature value of each edge cache device in each theme; and calculate a second preference degree of each content request user for each content according to the normalized feature value of each content in each theme and the normalized feature value of each content request user in each theme.
[0101] Optionally, the pairing module is specifically configured to connect each edge cache device and the content request user in the corresponding coverage range according to the content request user contained in the coverage range of each edge cache device, to form a bipartite graph including multiple edge cache devices, multiple content request users and multiple connection edges; calculate the weight of the multiple connection edges according to the preference similarity between the multiple content request users and the multiple edge cache devices and the distance between the multiple content request users and the multiple edge cache devices; and determine the edge cache device matched by each content request user by using the Hungarian algorithm according to the bipartite graph and the weight of the multiple connection edges.
[0102] Optionally, the download cost model is constructed in the following manner: The download cost model is constructed according to the request indication variable of the multiple contents requested by the user, the cache indication variable of the multiple contents, the size of the multiple contents, the cost of downloading unit data from the edge cache device and the cost of downloading unit data from the core network.
[0103] Optionally, the aging cost model is constructed in the following manner: The information age of the multiple contents in the target time slot is calculated according to the update indication variable of the multiple contents. The aging cost model is constructed according to the cache indication variable of the multiple contents, the information age of the multiple contents in the target time slot and the size of the multiple contents.
[0104] Optionally, the update cost model is constructed in the following manner: The update cost model is constructed according to the update indication variable of the multiple contents, the cost of downloading unit data from the core network and the size of the multiple contents.
[0105] Optionally, the policy calculation module 503 is specifically configured to solve the total cost model according to the cache capacity of each edge cache device, the backhaul link transmission capacity and the maximum tolerable information age, to obtain initial cache values and initial update values of a plurality of contents in each edge cache device; sort the initial update values of the plurality of contents, calculate the content update quantity under the constraint of the backhaul link transmission capacity; determine the plurality of target contents corresponding to the content update quantity as the to-be-updated contents of each edge cache device in the target time slot; determine the to-be-updated contents as the first to-be-cached contents; sort the initial cache values of the other contents in the plurality of contents except the first to-be-cached contents, and determine the contents satisfying the maximum cache capacity of each edge cache device as the second to-be-cached contents; the cache strategy includes the first to-be-cached contents and the second to-be-cached contents, and the update strategy includes the to-be-updated contents.
[0106] Optionally, the apparatus further includes: The content transmission module is configured to determine, according to the acquisition request for the target content sent by the target user, whether the target content is contained in the target edge cache device corresponding to the target user; if the target content is contained in the target edge cache device, the target content is sent to the target user through the target edge cache; if the target content is not contained in the target edge cache device, the target content is acquired from the core network through the macro base station, and the target content is sent to the target user.
[0107] In a possible implementation manner, the embodiment of the present application further provides an electronic device, Figure 9 As shown in the schematic diagram of the electronic device provided by the embodiment of the present application, Figure 9 The electronic device can include a processor 601, a memory 602 and a communication bus 603, wherein the memory 602 stores machine-readable instructions executable by the processor 601, when the electronic device is running, the processor 601 communicates with the memory 602 through the communication bus 603, and the processor 601 executes the machine-readable instructions to implement the information processing method of the above-mentioned embodiment.
[0108] The above is merely specific implementation of the present application, but the protection scope of the present application is not limited to this, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
Claims
1. An information processing method, characterized in that: The method comprises: caching corresponding contents in the plurality of edge cache devices according to user preferences of a plurality of content requesting users for a plurality of contents, content popularity of the plurality of contents, and pairing relationships between the plurality of content requesting users and a plurality of edge cache devices, wherein the plurality of contents are stored in a core network and each content requesting user is paired with a unique corresponding edge cache device; constructing a total cost model based on the download cost model, the aging cost model, and the update cost model of the plurality of contents in the target time slot; Solving the total cost model based on the cache capacity of each edge cache device, the backhaul link transmission capacity between the macro base station and the core network, and the maximum tolerable information age, to determine a cache policy and an update policy for each content in each edge cache device at the target time slot, wherein the cache policy is used to indicate whether to cache each content in each edge cache device at the target time slot, and the update policy is used to indicate whether to update each content in each edge cache device at the target time slot; updating the content cached in each edge cache device according to the cache strategy and the update strategy; Solving the total cost model based on the cache capacity of each edge cache device, the backhaul link transmission capacity between the macro base station and the core network, and the maximum tolerable information age to determine the caching strategy and update strategy for each content in each edge cache device at the target time slot includes: Solving the total cost model according to the cache capacity of each edge cache device, the backhaul link transmission capacity, and the maximum tolerable information age to obtain initial cache values and initial update values of the multiple contents in each edge cache device; sorting the initial update values of the plurality of contents, and calculating the number of content updates under the transmission capacity constraint of the backhaul link; Determine a plurality of target contents corresponding to the content update quantity as the contents to be updated for each edge cache device in the target time slot; Determining that the content to be updated is the first content to be cached; sorting initial cache values of the contents other than the first to-be-cached content among the plurality of contents, and determining the content that satisfies the maximum cache capacity of each edge cache device as the second to-be-cached content; The cache strategy includes: the first content to be cached and the second content to be cached, and the update strategy includes: the content to be updated.
2. The method according to claim 1, wherein Before caching corresponding contents in the plurality of edge cache devices respectively based on user preferences of the plurality of content requesting users for the plurality of contents, the content popularity of the plurality of contents, and the pairing relationships between the plurality of content requesting users and the plurality of edge cache devices, the method further includes: respectively calculating first preference degrees of a plurality of edge cache devices for the plurality of contents and second preference degrees of a plurality of content requesting users for the plurality of contents; Calculating preference similarities between the plurality of content requesting users and the plurality of edge cache devices based on the first preference and the second preference; The multiple content requesting users are paired with the multiple edge cache devices one-to-one according to the preference similarities between the multiple content requesting users and the multiple edge cache devices, and the distances between the multiple content requesting users and the multiple edge cache devices.
3. The method according to claim 2, wherein The step of respectively calculating first preferences of a plurality of edge cache devices for the plurality of contents and second preferences of a plurality of content requesting users for the plurality of contents includes: Calculating the normalized feature value of each content under each topic according to the score of each content under multiple topics; Calculating normalized feature values of each content requesting user and each edge cache device under each topic, respectively, based on correlations between each content requesting user and each edge cache device and the multiple topics; Calculating a first preference of each edge cache device for each content according to the normalized feature value of each content under each topic and the normalized feature value of each edge cache device under each topic; A second preference degree of each content requesting user for each content is calculated based on the normalized feature value of each content under each topic and the normalized feature value of each content requesting user under each topic.
4. The method according to claim 2, wherein The one-to-one pairing of the plurality of content requesting users and the plurality of edge cache devices according to the preference similarities between the plurality of content requesting users and the plurality of edge cache devices and the distances between the user devices includes: According to the content requesting users included in the coverage area of each edge cache device, each edge cache device is connected to the content requesting users within the corresponding coverage area to form a bipartite graph including the plurality of edge cache devices, the plurality of content requesting users, and a plurality of connecting edges; Calculating weights of the plurality of connection edges according to preference similarities between the plurality of content requesting users and the plurality of edge cache devices, and distances between the plurality of content requesting users and the plurality of edge cache devices; The Hungarian algorithm is used to determine an edge cache device that matches each content requesting user according to the bipartite graph and the weights of the multiple connecting edges.
5. The method according to claim 1, wherein The download cost model is constructed as follows: The download cost model is constructed according to a request indicator variable of a user requesting the multiple contents, a cache indicator variable of the multiple contents, a size of the multiple contents, a cost of downloading unit data from the edge cache device, and a cost of downloading the unit data from the core network.
6. The method according to claim 1, wherein The aging cost model is constructed in the following way: calculating, according to update indicator variables of the plurality of contents, information ages of the plurality of contents in the target time slot; The aging cost model is constructed according to the cache indicator variables of the multiple contents, the information age of the multiple contents in the target time slot, and the size of the multiple contents.
7. The method according to claim 1, wherein The update cost model is constructed in the following manner: The update cost model is constructed according to update indicator variables of the plurality of contents, a cost of downloading unit data from the core network, and sizes of the plurality of contents.
8. The method according to claim 1, wherein The method further comprises: Determining, based on a request for obtaining target content sent by a target user, whether a target edge cache device corresponding to the target user contains the target content; If the target edge cache device contains the target content, sending the target content to the target user through the target edge cache; If the target edge cache device does not include the target content, the target content is obtained from the core network through the macro base station, and the target content is sent to the target user.
9. An electronic device, characterized in that: The electronic device comprises a processor, a memory and a communication bus, wherein the memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the communication bus, and the processor executes the machine-readable instructions to implement the steps of the information processing method described in any one of claims 1 to 8.