Lunar communication network caching deployment scheme based on reliable transmission

Abstract

This invention provides a caching deployment scheme for a lunar communication network based on reliable transmission, comprising: constructing a caching cost model for each device node based on the size of the content to be cached on each device node; constructing an energy consumption model for each device node based on the energy consumed by the device node when caching data, maintaining probe activities, and transmitting data; constructing a latency cost model for each device node to retrieve requested content based on the location of the cached data; constructing a failure probability model for each device node based on its hardware state and the environmental state, and constructing an availability model for the device node based on the failure probability model; constructing a caching strategy optimization model for each device node based on its availability model, caching cost model, energy consumption model, and latency cost model; transforming the caching strategy optimization model into an unconstrained caching strategy optimization model, and solving the unconstrained caching strategy optimization model using the mayfly algorithm to obtain the optimal data caching strategy, and caching the data of the device nodes based on the optimal data caching strategy.

Description

Technical Field

[0001] This invention belongs to the field of communication technology, and in particular relates to a lunar surface communication network cache deployment scheme based on reliable transmission. Background Technology

[0002] As lunar exploration activities continue to deepen, the number and types of lunar surface exploration nodes will continue to increase, and persistent communication between devices will be necessary to establish a self-organizing network among lunar surface nodes and ensure communication between them. In the lunar environment, the deployment range of wired links is limited; a more practical approach is to combine a lunar surface wireless backbone network with a regional wireless self-organizing network. During lunar exploration missions, astronauts, rovers, and robots will form multiple exploration teams, creating a lunar surface wireless self-organizing communication network to conduct outward exploration tasks. The exploration teams will upload the acquired exploration data to the lander, while the lander maintains real-time communication with the exploration teams, monitoring their positions and astronaut vital signs and issuing command orders. Exploration teams can communicate with each other via multi-hop communication through the lander or directly.

[0003] However, lunar communication scenarios are completely different from those on Earth, such as... Figure 2 and Figure 3 As shown. Compared with the Earth mobile ad hoc network, the lunar surface wireless communication environment is poor. The lunar surface exploration nodes have the following characteristics: (1) Affected by the lunar surface topography, magnetic field, radiation, charged lunar dust and temperature, the communication links between exploration nodes are easily interrupted, making the network topology highly dynamic. At the same time, the node equipment ages faster and the failure probability is higher. (2) Due to the high cost of lunar equipment launch and the difficulty of adding equipment, the computing power and storage capacity of the lunar surface communication network nodes are limited. (3) The energy of the nodes is limited. The exploration nodes usually generate electricity through solar panels. They can work normally during the lunar day, but enter a dormant state during the lunar night. Only a few devices are on standby so that the lunar rover can be restarted when the lunar day arrives. (4) Since there is no atmosphere on the lunar surface, there is no infrastructure such as spacecraft to assist communication in the lunar surface communication network. In summary, the complex topography, magnetic field, radiation, charged lunar dust, temperature changes and the limitations of energy and resources on the lunar surface cause the reliable communication network to face many challenges, such as low link reliability, high node failure rate and easy data loss, resulting in low data transmission reliability and high transmission delay. Summary of the Invention

[0004] To address the problems existing in the background technology, this invention uses caching for data backup, and other nodes can obtain backup data from nearby cache nodes. Then, the mayfly algorithm is used to solve for the optimal cache node scheme to improve data transmission reliability and reduce latency.

[0005] To achieve the above technical objectives, this invention provides a lunar surface communication network caching deployment scheme based on reliable transmission, including:

[0006] S1: Construct a caching cost model for each device node based on the size of the content that needs to be cached on each device node;

[0007] S2: Construct an energy consumption model for the device node based on the energy consumed by the device node when caching data, maintaining probe activities, and transmitting data;

[0008] S3: Based on the data location cached by the device node, construct a latency cost model for the device node to retrieve the requested content;

[0009] S4: Construct a failure probability model for the device node based on its hardware status and the environmental status of the device, and construct an availability model for the device node based on the failure probability model.

[0010] S5: Construct a caching strategy optimization model for device nodes based on the availability model, caching cost model, latency cost model for retrieving requested content, and energy consumption model of device nodes.

[0011] S6: Transform the cache strategy optimization model into an unconstrained cache strategy optimization model, and use the mayfly algorithm to solve the unconstrained cache strategy optimization model to obtain the optimal data caching strategy. Based on the optimal data caching strategy, cache the data of the device nodes.

[0012] The present invention has at least the following beneficial effects

[0013] This invention constructs a caching cost model based on the size of the content to be cached on each device node, enabling precise quantification of the costs associated with caching data on different device nodes. Based on this cost model, the cached content of each device node can be rationally planned, avoiding unnecessary caching overhead and effectively controlling overall caching costs, thus achieving optimized resource allocation. The constructed device node energy consumption model considers in detail the energy consumption of devices when caching data, maintaining probe activities, and transmitting data, estimating the energy consumption of devices under different caching strategies and prioritizing caching schemes with lower energy consumption. The latency cost model constructed based on the data location cached on device nodes is crucial for improving data access efficiency. This model helps us cache data in locations that are more conducive to reducing the latency of retrieving requested content. The failure probability model and availability model constructed based on the hardware and environmental states of device nodes provide reliability considerations for caching strategy optimization. Through the availability model, when selecting caching device nodes, nodes with good hardware conditions, stable environments, and high availability are prioritized, thereby reducing the risk of data loss due to device failure and enhancing the reliability and stability of the entire system. A comprehensive caching strategy optimization model is constructed by integrating availability, caching cost, and energy consumption models. This multi-dimensional approach enables global system optimization, avoiding the problem of optimizing only a single factor while neglecting other important aspects. This ensures good performance across all aspects of the system. The mayfly algorithm is used for solution, which efficiently finds the optimal data caching strategy. It can quickly locate the data caching method that optimizes system performance in a complex solution space, and then cache the data on device nodes based on this optimal strategy, ensuring that the system achieves best performance in actual operation. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0015] Figure 2 This is a schematic diagram of a lunar surface communication scenario;

[0016] Figure 3 This is a schematic diagram of the lunar surface communication environment. Detailed Implementation

[0017] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0018] Please see Figure 1 This invention provides a lunar surface communication network caching deployment scheme based on reliable transmission, including:

[0019] S1: Construct a caching cost model for each device node based on the size of the content that needs to be cached on each device node;

[0020] S2: Construct an energy consumption model for the device node based on the energy consumed by the device node when caching data, maintaining probe activities, and transmitting data;

[0021] S3: Based on the data location cached by the device node, construct a latency cost model for the device node to retrieve the requested content;

[0022] S4: Construct a failure probability model for the device node based on its hardware status and the environmental status of the device, and construct an availability model for the device node based on the failure probability model.

[0023] S5: Construct a caching strategy optimization model for device nodes based on the availability model, caching cost model, latency cost model for retrieving requested content, and energy consumption model of device nodes.

[0024] S6: Transform the cache strategy optimization model into an unconstrained cache strategy optimization model, and use the mayfly algorithm to solve the unconstrained cache strategy optimization model to obtain the optimal data caching strategy. Based on the optimal data caching strategy, cache the data of the device nodes.

[0025] In this embodiment, the lunar communication network can be represented as graph G = (U, M, S, C), where U = {u1, u2, ..., u}. n} represents the set of all mobile nodes in the network (mobile nodes), M = {u1, u2, ..., u...} m} represents the set of all nodes in the network, m (mobile nodes + terminal nodes); This indicates the size of the content that needs to be cached on each device node; the cache capacity of a mobile node is... Nodes with cache space can cache content of an appropriate size, allowing each node to retrieve content from its local cache and neighboring cache nodes, reducing content retrieval latency and energy consumption; in the lunar communication network, when the mobile node u j When initiating a content request, it first checks if the content is available in its local cache; if not, it requests it through the cache node u within the communication scope. i Search in the middle; if the requested content cannot be obtained in all cache nodes, then search via the lunar base u. o The content cached by each cache node is transferred to the base for storage.

[0026] In the network, each node possesses mobility and autonomy, and the neighboring nodes of each node may dynamically change due to the node's mobility. A probe node has a communication coverage area of ​​radius r. A wireless link can be established between any two nodes if the communication distance between them is not greater than the maximum effective communication radius r. The link state between nodes is defined as a binary variable ξ. i,j ∈{0,1},u i ,u j ∈U, ξ i,j =1 represents node u i and node u j A communication link can be established; otherwise, ξ i,j =0, that is:

[0027]

[0028] Among them, h i,j Represented as node u i and node u j Channel gain between, h th It is the channel gain threshold for establishing a wireless communication link. Only when the gain is greater than this threshold can a link be established between two points.

[0029] To ensure transmission reliability, nodes can pre-cache data in a cache node using a binary variable. To represent node u j Select cache node u i The states are:

[0030]

[0031] in, Represents node u i As node u j Cache nodes, This indicates that node u i Not node u jThe cache nodes. When caching content, the nodes consume storage and energy resources, thus incurring caching costs. Due to the probe node u... i Different performance, different u i The caching costs also differ. (Regarding α) i Represented as node u i The cost of caching content per unit size is then node u i Cache u j The cost of caching data is:

[0032]

[0033] Therefore, a caching cost model for device nodes is constructed, which includes:

[0034]

[0035]

[0036] Among them, Ξ i,j Indicates device node u i Cache device node u j The cost of caching data; Time indicates device node u i As device node u j Cache nodes, Time indicates device node u i Not device node u j Cache nodes; α i Indicates device node u i The cost required to cache a unit of content.

[0037] During exploration activities, each node consumes energy for caching files, maintaining normal exploration activities, and transmitting data. In a lunar communication network, on-network caching can reduce energy consumption to some extent. Although solar panels can be used to generate electricity, the total energy of the equipment is limited due to poor lunar illumination conditions and may be depleted, leading to performance degradation. Node energy consumption mainly consists of three parts: fixed energy consumption, storage energy consumption, and transmission energy consumption.

[0038] Fixed energy consumption refers to the energy consumed to maintain node movement and exploration. This paper assumes that the fixed energy consumption of each node is the same. As a platform for lunar surface exploration, the lunar rover will carry various scientific instruments to conduct exploration of the lunar surface. Therefore, the rover's mobility is related to the execution process and effectiveness of the lunar exploration mission. The special driving environment on the lunar surface places higher demands on the performance of the rover's drive and steering systems, and imposes stricter limits on the energy consumption of the rover's driving system. Studies indicate that it is more energy-efficient than an independently steering lunar rover. The total energy consumption of the mobile device node UI is...

[0039] If nodes are within communication range, they can directly establish a wireless link for communication. If nodes are not within communication range, they can communicate through multi-hop communication. Therefore, node u i With node u j The transmission energy consumption is:

[0040]

[0041] Cache energy consumption: Storage energy consumption refers to the energy consumed for caching content, node u i Cache node u j Energy consumed by data The following formula can be used to calculate:

[0042]

[0043] in, This represents the energy required to cache each bit of data. Represents node u j The size of the data that needs to be cached.

[0044] Therefore, node u i As node u j Total energy consumed by cache nodes for:

[0045]

[0046] node u i The energy obtained in each time slot through solar energy, etc. is Then node u i Remaining energy The calculation is as follows:

[0047]

[0048] Among them, B i Let i represent the battery size of node i, and

[0049] Therefore, an energy consumption model for the device node is constructed, which includes:

[0050]

[0051]

[0052]

[0053]

[0054]

[0055]

[0056] Among them, B i Indicates device node u i The battery size; U represents the set of all mobile nodes in the lunar communication network; Indicates device node u i Energy obtained through solar energy, etc.; Indicates device node u i Cache device node u j Total energy consumption of the data; Indicates device node u i Total energy consumption for mobile applications; This represents the device node u. i Cache device node u j The energy consumption of the data; Indicates device node u j Transmit the data that needs to be cached to the device node u. i Transmission energy consumption, Indicates device node u i The remaining energy; Indicates device node u j The size of the data to be cached; J represents the energy required to cache each bit of data; J represents the device node u. i To device node u j The set of path nodes between them, where v represents an element in set J; γ represents the device node u. i Energy consumed per unit time to transmit data; r v,v-1 Indicates device node u i To device node u j The data transmission rate of each path segment, ξ i,j =1 indicates device node u i and device node u j Establish a communication link; ξ i,j =0 indicates device node u i and device node u j No communication link was established; i,j Indicates device node u i and device node u j Data transfer rate between; B i,j Indicates device node u i and device node u j The channel bandwidth between; log2() represents the logarithm to the base 2; p j Indicates device node uj The transmit power; u i′ ∈U,u i′ ≠u i Indicates that, except for device node u i Other mobile nodes; h i,j Indicates device node u i and device node u j The channel gain between; N0 represents the power spectral density of Gaussian white noise; h th This represents the channel gain threshold for establishing a wireless communication link.

[0057] In lunar communication networks, transmission latency is a key performance indicator. If data can be found in a cache node close to the requester, it avoids retrieving data from a distant data source (such as the base), thus significantly reducing transmission latency. Assume node u... i and node u j The coordinates are (x) i ,y i ) and (x j ,y j ), Represents node u i and node u i Shortest distance:

[0058]

[0059] Here is Then when node i Retrieve the requested content f from node j. m The data transmission rate is:

[0060]

[0061] Among them, nodes i The channel gain h between node j and node j i,j B i,j For node u i and node u j The channel bandwidth, p j U is the transmission power, N0 is the power spectral density of Gaussian white noise, and u i′ ∈U,u i′ ≠u i Indicates that, except for device node u i Other moving nodes. Then node u j From node u i The latency for retrieving the requested content is:

[0062]

[0063] Therefore, a high-latency cost model is constructed, which includes:

[0064]

[0065]

[0066] Among them, T i,j Represents node u j From node u i The latency of retrieving the requested content. Indicates device node u i and device node u v The Euclidean distance between them; Time indicates device node u i As device node u j Cache nodes, Time indicates device node u i Not device node u j cache nodes; r i,ν Indicates device node u i and device node u v Data transfer rate between, τ j Indicates device node u j The latency required to retrieve data from the base.

[0067] In lunar surface communication, on the one hand, the aging of communication equipment leading to hardware degradation may cause equipment failure; on the other hand, complex external factors on the lunar surface, such as temperature differences, lunar dust concentration, radiation intensity, and light intensity, can cause irreversible damage to communication equipment, severely impacting its normal operation. Therefore, we introduce a failure probability model for equipment nodes, which includes:

[0068]

[0069]

[0070]

[0071]

[0072]

[0073]

[0074] P m =1-e -FAΔt

[0075]

[0076]

[0077] in, Indicates device node u j The failure probability, k1 represents the equipment aging factor, and t represents the usage time of the equipment node; T exp The device node's lifespan is represented by e, the natural base, and D, which represents the radiation dose on the lunar surface. th Indicates the radiation resistance threshold of the device node; P m P represents the probability that a device on the lunar surface will be struck by a micrometeoroid. d|m The conditional probability of a device node being damaged by a micrometeoroid impact is represented by: F; the micrometeoroid flux on the lunar surface; A; the exposed area of ​​the device node; Δt; and T. min Indicates the minimum operating temperature of the device node; T max T represents the maximum operating temperature of the device node; T represents the ambient temperature of the lunar surface; T ref Indicates reference temperature; T r k1 represents the temperature sensitivity parameter of the device node; k2 represents the dust resistance parameter of the device node; C d This indicates the concentration of dust on the lunar surface.

[0078] An availability model for a device node is constructed based on the failure probability model of the device node. The availability model of the device node includes:

[0079]

[0080] Among them, L i Indicates device node u i Availability, Indicates device node u i The failure probability is ρ, where ρ represents the failure probability threshold.

[0081] If node u j Select node u i As a cache node, the total overhead required is calculated as follows:

[0082]

[0083] Where μ, λ, κ, and γ are all constants in the range [0,1], and μ+λ+k+γ=1.

[0084] In the scenario of deploying cache nodes in a lunar communication network, considering that the total overhead consists of factors such as energy consumption, latency, failure rate, and cache cost, and that these factors have different numerical ranges and units, a minimum-maximum normalization method can be used to process the total overhead in order to more reasonably compare the overhead of different nodes. The total overhead after normalization is:

[0085]

[0086] In this way, the total overhead of different nodes is normalized to an interval, allowing for a more intuitive comparison of the merits of each node as a cache node. Therefore, the cache node deployment problem within each region can be characterized as finding the optimal deployment strategy under constraints of node energy and failure probability, minimizing cache overhead while satisfying node availability constraints. This leads to the construction of a device node cache strategy optimization model, which includes:

[0087]

[0088]

[0089]

[0090] in, Time indicates device node u i As device node u j Cache nodes, Time indicates device node u i Not device node u j Cache nodes; ξ i,j =1 indicates device node u i and device node u j Establish a communication link; ξ i,j =0 indicates device node u i and device node u j No communication link was established; M represents the set of probe device nodes, containing m devices; U represents the set of all mobile nodes, containing n mobile devices; constraint C1 represents the constraint on establishing a communication link between nodes and the binary buffer decision variable; C i Indicates device node u i C2 indicates that the cached data on a node cannot exceed the cache capacity. Indicates device node u i The remaining energy, Indicates device node u i Total mobile energy consumption, C3 indicates that the node's remaining energy must be sufficient to support its normal exploration mission; Indicates device node u iEnergy obtained through solar energy Indicates device node u i Cache device node u j The total energy consumption of the data, C4 represents node u i The total energy consumption of cached data is no greater than that of node u. i Energy generated by the solar panel; T i,j Represents node u j From node u i The latency of retrieving the requested content, τ j Indicates device node u j The latency required to retrieve data from the base, C5 represents node u. j From cache node u i The latency for retrieving content is much shorter than that for node u. j Delay τ is obtained from the base. j ; and t represents the minimum and maximum energy consumption of the node cache, respectively. min and t max L represents the minimum and maximum delay values, respectively. min and L max These represent the minimum and maximum values ​​of the availability of energy-saving equipment, respectively. min and Ξ max λ and κ represent the minimum and maximum caching costs of the device nodes, respectively. μ, λ, κ and γ are weighting coefficients, and their values ​​range from [0,1]. These coefficients are used to adjust the importance of different factors in the optimization model, and μ+λ+κ+γ=1.

[0091] After modeling the system, it can be seen from the objective function and constraints that the above mathematical model has high complexity, making it difficult to find its optimal solution in polynomial time. Inspired by the life cycle and group behavior of mayflies, the mayfly algorithm has strong global search capabilities and is often used to optimize problems. Therefore, this invention uses the mayfly algorithm to solve the problem and obtain the globally optimal cache set. In the lunar communication network scenario of this invention, we regard the cache overhead function as the fitness function of the mayflies, and the position of each mayfly represents a set of cache node selection schemes. For example, the position of a mayfly can be represented by a vector. This indicates that the selection scheme for cache nodes in the network is achieved by the movement and mating of mayflies to obtain the globally optimal position, which corresponds to the best set of cache nodes in the lunar communication network.

[0092] To solve objective optimization problems using the mayfly algorithm, it is necessary to transform the original problem. To transform the constrained optimization problem into an unconstrained optimization problem, we utilize the concept of a penalty method. The transformation process is described below. The penalty function is defined as:

[0093] P(x)=θ∑g(f(x))f 2 (x)+ζ∑g(h(x))h 2 (x)

[0094] Here, θ and ζ are penalty terms, and x is the optimization variable. The equality constraint is transformed into the form h(x) = 0. When h′(x) = 0, the constraint condition is satisfied, and g(h(x)) = 0, meaning the new objective function value remains unchanged. h′(x) ≠ 0 indicates that the constraint is not satisfied, and g(h(x)) = 1. Due to the existence of the penalty term, the value of the new objective function is greatly improved. In this way, solutions with high overhead can be eliminated. Using the same principle, the inequality constraint is transformed into the form f(x) ≤ 0, which facilitates subsequent optimization. Therefore, the original problem can be transformed into:

[0095]

[0096] Finally, the optimization problem is given as follows:

[0097] minφ(x)

[0098]

[0099]

[0100]

[0101] Therefore, the process described above transforms the cache strategy optimization model into an unconstrained cache strategy optimization model, including:

[0102] Define the penalty function for the caching strategy optimization model:

[0103] P(x)=θ∑g(f(x))f 2 (x)+ζ∑g(h(x))h 2 (x)

[0104] Where θ and ζ are penalty coefficients, x is the optimization variable, and the inequality constraints are transformed into the form f(x) ≤ 0, g() represents the control function, h(x) represents the equality constraint, and f(x) represents the inequality constraint. The original problem is then transformed into:

[0105]

[0106] The unconstrained caching strategy optimization model is obtained as follows:

[0107] minφ(x)

[0108]

[0109]

[0110] Where φ(x) represents the unconstrained caching strategy optimization model, and T i,j Represents node u j From node u i The latency of retrieving the requested content, τ j Indicates device node u j The latency required to retrieve data from the base.

[0111] Each mayfly's position represents a potential solution to the problem. First, there are two random groups of mayflies, representing the male and female populations respectively. This means the mayflies are randomly placed in the problem space, their positions represented by vector x, and their performance is evaluated according to a fitness function f(x). Each mayfly's flight direction is a dynamic interaction of individual and social flight experience. Each mayfly adjusts its position by dynamically updating its velocity, moving towards the optimal location.

[0112] Let υ denote the velocity of the mayfly, and po(x) denote the optimization target. The position of the mayfly in the problem space is equivalent to the potential solution of the problem. The swarming of male mayflies means that the position of each male mayfly is adjusted based on its own experience and that of its neighbors. Let... This represents the position of the male mayfly m in the e-th iteration, with the velocity added to the current position. The movement process of male mayflies is as follows:

[0113]

[0114] Assuming the male mayflies are not moving very fast, they continue to move. Then the speed of the male mayfly is defined as:

[0115]

[0116] in, Let represent the movement speed in the e-th iteration, a1 be a constant measuring cognitive contribution, a2 be a constant measuring social contribution, and pbest. m represents the best historical position of the male mayfly (m), gbest represents the best global position, β is a fixed visibility factor that limits the visibility of the mayfly, and r p This is the mayfly's current location and pbest. m Cartesian distance between them, r g The Cartesian distance between the current position and gbest is:

[0117]

[0118] gbest=min{φ(pbest1),φ(pbest2),...,φ(pbest M )}

[0119] In the formula, M represents the total number of male mayflies. Unlike males, female mayflies do not gather in groups. They fly to males to reproduce. The movement process of female mayflies is as follows:

[0120]

[0121] Since the attraction process is random, we decided to model it as a deterministic process. That is, based on fitness values, the best females are attracted to the best males, and the second best females are attracted to the second best males, as shown below:

[0122]

[0123] Where, r m,f Let be the distance between the male and female mayflies, and fe be a random walk coefficient. According to the fitness rule, when a female mayfly is attracted to a male, it moves towards him. When a female mayfly is not attracted, it will perform a random walk, moving randomly in the search space, where r∈[-1,1] is a random value.

[0124] The method of reproduction also follows the principle of mating the best female with the best male, the second best female with the second best male, and so on. The result of this crossbreeding is two offspring, produced as follows:

[0125] spring1=off*male+(1-off)*female,

[0126] spring2=off*female+(1-off)*male,

[0127] Here, off is a random value within a certain range, the initial velocity of the offspring is 0, and the iteration is repeated until the difference between the optimal fitness values ​​obtained from two consecutive iterations is less than ε (ε is a very small positive number), or the number of iterations reaches the preset maximum number of iterations, causing the algorithm to converge.

[0128] Therefore, summarizing the above process, the solution to the unconstrained caching strategy optimization model using the mayfly algorithm includes:

[0129] S611: Initialize the mayfly population S, the offspring population F, the convergence error ε, and the maximum number of iterations I. max ;

[0130] S612: Using the unconstrained caching strategy optimization model as the objective function, initialize the position of each mayfly, where the position of the mayfly represents the selection scheme of the cache node;

[0131] S613: Calculate the initial fitness value φ(x) of each mayfly using the objective function based on the initial position po(x) of the mayfly;

[0132] S614: Male mayflies based on their historical best position (pbest). m and the update speed of the globally optimal position gbest According to speed Update its location

[0133] S615: Male mayflies based on updated location Calculate fitness value using objective function And with the historical best position pbest m In comparison, if Then update the historical best position pbest m ;

[0134] S616: Based on the best historical position of all male mayflies (pbest) m Select the position with the lowest fitness and update the global optimal position gbest;

[0135] S617: Female mayflies calculate fitness values ​​based on an objective function. Update speed Update its location

[0136] S618: Mating mayfly individuals in offspring population F and assessing the fitness of the offspring;

[0137] S619: Create new populations, retain the best males and females, and ensure that the best individuals in the population can be passed on to the next generation;

[0138] S620: Repeat the above operation until the convergence criterion is met, and output the optimal data caching strategy of the cache node corresponding to the globally optimal position gbest.

[0139] Simulating the mayfly's courtship dance and random flight enhances the balance between exploration and exploitation properties, two effective methods to help the algorithm escape local optima. The number of algorithm iterations is set to I. max Let the population size be S, and d be the data dimension. The algorithm's complexity is O(1). max *S*d). This scheme establishes a system model that comprehensively considers cache cost, energy consumption, latency, and availability, taking into account the characteristics of lunar surface communication networks. It also utilizes the mayfly algorithm to solve the cache node deployment problem, providing reliable communication support for future lunar exploration missions.

[0140] In summary, this invention constructs a caching cost model based on the size of the content to be cached on each device node, enabling precise quantification of the costs associated with caching data on different device nodes. Based on this cost model, the cached content of each device node can be rationally planned, avoiding unnecessary caching overhead and effectively controlling overall caching costs, thus achieving optimized resource allocation. The constructed device node energy consumption model considers in detail the energy consumption of devices when caching data, maintaining probe activities, and transmitting data, estimating the energy consumption of devices under different caching strategies and prioritizing caching schemes with lower energy consumption. The latency cost model constructed based on the data location cached on device nodes is crucial for improving data access efficiency. This model helps us cache data in locations that are more conducive to reducing the latency of retrieving requested content. The failure probability model and availability model constructed based on the hardware and environmental states of device nodes provide reliability considerations for caching strategy optimization. Through the availability model, when selecting caching device nodes, nodes with good hardware conditions, stable environments, and high availability are prioritized, thereby reducing the risk of data loss due to device failure and enhancing the reliability and stability of the entire system. A comprehensive caching strategy optimization model is constructed by integrating availability, caching cost, and energy consumption models. This multi-dimensional approach enables global system optimization, avoiding the problem of optimizing only a single factor while neglecting other important aspects. This ensures good performance across all aspects of the system. The mayfly algorithm is used for solving the problem. This method efficiently finds the optimal data caching strategy, quickly locating the data caching method that optimizes system performance within a complex solution space. Then, based on this optimal strategy, data is cached on device nodes, ensuring the system achieves best performance in actual operation.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A lunar surface communication network cache deployment scheme based on reliable transmission, characterized in that, include: S1: Construct a caching cost model for each device node based on the size of the content that needs to be cached on each device node; S2: Construct an energy consumption model for the device node based on the energy consumed by the device node when caching data, maintaining probe activities, and transmitting data; S3: Based on the data location cached by the device node, construct a latency cost model for the device node to retrieve the requested content; S4: Construct a failure probability model for the device node based on its hardware status and the environmental status of the device, and construct an availability model for the device node based on the failure probability model. S5: Construct a caching strategy optimization model for device nodes based on the availability model, caching cost model, latency cost model for retrieving requested content, and energy consumption model of device nodes. S6: Transform the cache strategy optimization model into an unconstrained cache strategy optimization model, and use the mayfly algorithm to solve the unconstrained cache strategy optimization model to obtain the optimal data caching strategy. Based on the optimal data caching strategy, cache the data of the device nodes.

2. The lunar surface communication network cache deployment scheme based on reliable transmission according to claim 1, characterized in that, The caching cost model for the device nodes includes: Among them, Ξ i,j Indicates device node u i Cache device node u j The cost of caching data; Time indicates device node u i As device node u j Cache nodes, Time indicates device node u i Not device node u j Cache nodes; α i Indicates device node u i The cost required to cache a unit of content.

3. The lunar surface communication network cache deployment scheme based on reliable transmission according to claim 1, characterized in that, The energy consumption model of the device node includes: Among them, B i Indicates device node u i The battery size; U represents the set of all mobile nodes in the lunar communication network; Indicates device node u i Energy obtained through solar energy, etc.; Indicates device node u i Cache device node u j Total energy consumption of the data; Indicates device node u i Total energy consumption for mobile applications; This represents the device node u. i Cache device node u j The energy consumption of the data; Indicates device node u j Transmit the data that needs to be cached to the device node u. i Transmission energy consumption, Indicates device node u i The remaining energy; Indicates device node u j The size of the data to be cached; J represents the energy required to cache each bit of data; J represents the device node u. i To device node u j The set of path nodes between them, where v represents an element in set J; Υ represents a device node u. i Energy consumed per unit time to transmit data; r v,v-1 Indicates device node u i To device node u j The data transmission rate of each path segment, ξ i,j =1 indicates device node u i and device node u j Establish a communication link; ξ i,j =0 indicates device node u i and device node u j No communication link was established; i,j Indicates device node u i and device node u j Data transfer rate between; B i,j Indicates device node u i and device node u j The channel bandwidth between; log2() represents the logarithm to the base 2; p j Indicates device node u j The transmit power; u i′ ∈U,u i′ ≠u i Indicates that, except for device node u i Other mobile nodes; h i,j Indicates device node u i and device node u j The channel gain between; N0 represents the power spectral density of Gaussian white noise; h th This represents the channel gain threshold for establishing a wireless communication link.

4. The lunar surface communication network cache deployment scheme based on reliable transmission according to claim 1, characterized in that, The time delay cost model includes: Among them, T i,j Represents node u j From node u i The latency of retrieving the requested content. Indicates device node u i and device node u v The Euclidean distance between them; Time indicates device node u i As device node u j Cache nodes, Time indicates device node u i Not device node u j cache nodes; r i,v Indicates device node u i and device node u v Data transfer rate between, τ j Indicates device node u j The latency required to retrieve data from the base.

5. The lunar surface communication network cache deployment scheme based on reliable transmission according to claim 1, characterized in that, The failure probability model of the device node includes: P m =1-e -FAΔt in, Indicates device node u j The failure probability, k1 represents the equipment aging factor, and t represents the usage time of the equipment node; T exp The device node's lifespan is represented by e, the natural base, and D, which represents the radiation dose on the lunar surface. th Indicates the radiation resistance threshold of the device node; P m P represents the probability that a device on the lunar surface will be struck by a micrometeoroid. d|m The conditional probability of a device node being damaged by a micrometeoroid impact is represented by: F; the micrometeoroid flux on the lunar surface; A; the exposed area of ​​the device node; Δt; and T. min Indicates the minimum operating temperature of the device node; T max T represents the maximum operating temperature of the device node; T represents the ambient temperature of the lunar surface; T ref Indicates reference temperature; T r k1 represents the temperature sensitivity parameter of the device node; k2 represents the dust resistance parameter of the device node; C d This indicates the concentration of dust on the lunar surface.

6. The lunar surface communication network cache deployment scheme based on reliable transmission according to claim 5, characterized in that, The availability model of the device node includes: Among them, L i Indicates device node u i Availability, Indicates device node u i The failure probability is ρ, where ρ represents the failure probability threshold.

7. The lunar surface communication network cache deployment scheme based on reliable transmission according to claim 1, characterized in that, The caching strategy optimization model for the device nodes includes: in, Time indicates device node u i As device node u j Cache nodes, Time indicates device node u i Not device node u j Cache nodes; ξ i,j =1 indicates device node u i and device node u j Establish a communication link; ξ i,j =0 indicates device node u i and device node u j No communication link was established; M represents the set of probe device nodes, containing m devices; U represents the set of all mobile nodes, containing n mobile devices; constraint C1 represents the constraint on establishing a communication link between nodes and the binary buffer decision variable; C i Indicates device node u i C2 indicates that the cached data on a node cannot exceed the cache capacity. Indicates device node u i The remaining energy, Indicates device node u i Total mobile energy consumption, C3 indicates that the node's remaining energy must be sufficient to support its normal exploration mission; Indicates device node u i Energy obtained through solar energy Indicates device node u i Cache device node u j The total energy consumption of the data, C4 represents node u i The total energy consumption of cached data is no greater than that of node u. i Energy generated by the solar panel; T i,j Represents node u j From node u i The latency of retrieving the requested content, τ j Indicates device node u j The latency required to retrieve data from the base, C5 represents node u. j From cache node u i The latency for retrieving content is much shorter than that for node u. j Delay τ is obtained from the base. j ; and t represents the minimum and maximum energy consumption of the node cache, respectively. min and t max L represents the minimum and maximum delay values, respectively. min and L max These represent the minimum and maximum values ​​of the availability of energy-saving equipment, respectively. min and Ξ max λ and κ represent the minimum and maximum caching costs of the device nodes, respectively. μ, λ, κ and γ are weighting coefficients, and their values ​​range from [0,1]. These coefficients are used to adjust the importance of different factors in the optimization model, and μ+λ+κ+γ=1.

8. The lunar surface communication network cache deployment scheme based on reliable transmission according to claim 7, characterized in that, The process of transforming the cache strategy optimization model into an unconstrained cache strategy optimization model includes: Define the penalty function for the caching strategy optimization model: P(x)=θΣg(f(x))f 2 (x)+ζ∑g(h(x))h 2 (x) Where θ and ζ are penalty coefficients, x is the optimization variable, and the inequality constraints are transformed into the form f(x) ≤ 0, g() represents the control function, h(x) represents the equality constraint, and f(x) represents the inequality constraint. The original problem is then transformed into: The unconstrained caching strategy optimization model is obtained as follows: minφ(x) Where φ(x) represents the unconstrained caching strategy optimization model, and T i,j Represents node u j From node u i The latency of retrieving the requested content, τ j Indicates device node u j The latency required to retrieve data from the base.

9. The lunar surface communication network cache deployment scheme based on reliable transmission according to claim 8, characterized in that, The process of solving the unconstrained caching strategy optimization model using the mayfly algorithm includes: S611: Initialize the mayfly population S, the offspring population F, the convergence error ε, and the maximum number of iterations I. max ; S612: Using the unconstrained caching strategy optimization model as the objective function, initialize the position of each mayfly, where the position of the mayfly represents the selection scheme of the cache node; S613: Calculate the initial fitness value φ(x) of each mayfly using the objective function based on the initial position po(x) of the mayfly; S614: Male mayflies based on their historical best position (pbest). m and the update speed of the globally optimal position gbest According to speed Update its location S615: Male mayflies based on updated location Calculate fitness value using objective function And with the historical best position pbest m In comparison, if Then update the historical best position pbest m ; S616: Based on all historical best positions of male mayflies (pbest) m Select the position with the lowest fitness and update the global optimal position gbest; S617: Female mayflies calculate fitness values ​​based on an objective function. Update speed Update its location S618: Mating mayfly individuals in offspring population F and assessing the fitness of the offspring; S619: Create new populations, retain the best males and females, and ensure that the best individuals in the population can be passed on to the next generation; S620: Repeat the above operation until the convergence criterion is met, and output the optimal data caching strategy of the cache node corresponding to the globally optimal position gbest.

Images