A method for predicting a lunar surface dynamic network topology

Abstract

The application belongs to the technical field of network analysis, and particularly relates to a lunar surface dynamic network topology prediction method, which comprises the following steps: when a node is accelerated or decelerated along a lunar surface, a two-dimensional rectangular coordinate system is constructed with a fixed base station on the lunar surface as a coordinate origin, and a Wiener process with a drift parameter is used to simulate the movement of the node; SINR, node energy consumption, buffer queue length and link availability are selected as parameter indexes, and a network topology is established for the fixed base station and the mobile node; the probability of a link "connection" in the network topology of the parameter indexes in the mobile simulation state of the mobile node is calculated, and a link state transition matrix is generated; and the network topology state transition probability table is calculated based on the generated link state transition matrix to represent the evolution of the network topology. The application jointly considers node attributes and link attributes, quantifies the link connection state change together, and more accurately predicts the network topology change.

Description

Technical Field

[0001] This invention belongs to the field of network analysis technology, specifically relating to a method for predicting the topology of a dynamic network on the lunar surface. Background Technology

[0002] Deep space exploration is a major trend in current aviation development, and the Moon, with its unique location and resources, has attracted widespread attention. In lunar exploration missions, probe nodes need to transmit data back to lunar base stations in real time to ensure the mission's success. Therefore, research on communication networks in lunar exploration is of great significance. Currently, lunar surface communication is mainly point-to-point wireless communication between the lander and the rover. For limited lunar exploration missions, the number of nodes is small, and the link states between nodes remain relatively stable, with no long-term link interruptions or new connections being established, making network topology maintenance relatively simple. However, due to the unknown environment and diverse node types on the lunar surface, future lunar exploration will involve more unmanned or manned lunar rover equipment collaborating in network communication. The mobility of lunar surface nodes and the temporal variability of link states between nodes mean that traditional point-to-point communication methods cannot characterize changes in lunar surface network topology. Therefore, a predictable dynamic network topology method is needed.

[0003] When conducting exploration missions on the lunar surface, data from lunar nodes needs to be transmitted back to the lunar base via links between nodes, such as... Figure 1 When lunar rovers and astronauts perform exploration missions, they establish links with neighboring nodes and base stations to transmit data. Lunar exploration nodes on the lunar surface are divided into two categories based on mobility: 1) Fixed lunar exploration nodes, which are deployed in smaller numbers, such as... Figure 1 Medium-sized base stations and fixed nodes are generally deployed in locations suitable for building fixed communication infrastructure, such as lunar surface stations or lunar highlands. Their applications include path planning information transmission and regional coverage communication. 2) Mobile lunar exploration nodes, such as Figure 1 In general, rovers, astronauts, and other similar personnel are used in scenarios where planned exploration missions are carried out, such as long-distance lunar surface exploration and lunar crater exploration. Their applications include real-time video monitoring of lunar craters and telemetry image data.

[0004] In lunar surface communication, node communication can be networked in several ways: 1) a combination of fixed and mobile nodes, and 2) a combination of mobile nodes with each other. Fixed nodes, due to their fixed location and typically fixed range, experience minimal network topology changes over a period of time. However, mobile nodes in lunar surface communication experience highly variable motion status information. Considering factors such as mobile node buffer queue length and node energy consumption, the network topology changes frequently. Furthermore, the lunar environment necessitates path loss during data transmission, making it difficult to establish or prone to interruption of wireless links between nodes, leading to link unavailability and delayed updates to network topology information. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention proposes a method for predicting the topology of a dynamic lunar network, characterized by comprising:

[0006] When the node accelerates or decelerates along the lunar surface, a two-dimensional rectangular coordinate system is constructed with the fixed base station on the lunar surface as the origin, and the movement of the node is simulated using the Wiener process with drift parameters.

[0007] SINR, node energy consumption, buffer queue length, and link availability were selected as parameters to establish a network topology for fixed base stations and mobile nodes.

[0008] The probability of link "connection" in the network topology under the mobile node's simulated mobile state is calculated based on the parameter index. The probability of link "connection" in the next time step is calculated based on the calculated parameter index probability. The link state transition matrix is ​​generated based on the probability of link "connection" in the next time step.

[0009] The network topology state transition probability is calculated based on the generated link state transition matrix to characterize the evolution of the network topology.

[0010] The beneficial effects of this invention are:

[0011] 1. This invention establishes a two-dimensional rectangular coordinate system for node motion, which can better characterize the motion state of the nodes;

[0012] 2. This invention jointly considers node attributes and link attributes to quantify changes in link connection states, and characterizes the evolution of network topology through network topology state transition probabilities, enabling timely and accurate prediction of network topology changes. Attached Figure Description

[0013] Figure 1 A schematic diagram of the lunar exploration node network;

[0014] Figure 2 An abstract network topology for a dynamic lunar network scenario;

[0015] Figure 3 This is an example diagram of an invalid network state under the maximum network state condition. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] To address the issue of reliable transmission in lunar surface communication, a lunar dynamic network topology prediction method is proposed, based on the motion state information of lunar rover nodes and considering the characteristics of the link state and content itself. This method constructs the topology using node performance and link state, and then forwards data based on the topology information.

[0018] like Figure 2 As shown, an abstract network topology for a dynamic lunar network scenario is presented. Nodes in the lunar scenario use celestial navigation for positioning, acquiring information such as their position, direction, and velocity. They periodically broadcast beacons to build a set of neighboring nodes, where neighboring nodes are defined as nodes within the node's wireless transmission range that can communicate directly. However, due to the high mobility of nodes, the constantly changing distances between nodes, signal interference, and shadow fading, the wireless links in the network are highly susceptible to interruption, as shown by link l in the figure. sj Due to signal interference and fading, the received signal strength at node j is too low, and the link cannot be connected. ce High-speed movement of nodes at both ends can easily cause the distance between nodes to exceed their respective transmission range. Furthermore, the time-varying data flow and network resources in the network cause the node's buffer space to constantly change. When the remaining buffer space of the receiving node is insufficient to store or process incoming data packets, the data packets will be discarded, and the communication link between nodes cannot be successfully established, as shown in link l in the figure. bg The interrupt occurred because node g ran out of cache space.

[0019] Therefore, in order to achieve accurate prediction and evolutionary characterization of dynamic network topology, it is necessary to start from the micro-causes that cause network topology to change over time, that is, to characterize the topology evolution process by analyzing the changes in the state of network nodes and links and the mutual influence between links.

[0020] 1. Nodal motion model

[0021] In dynamic network scenarios, the high mobility of network nodes is one of the fundamental reasons for the intermittency of links and the time-varying nature of network topology. Furthermore, the high-speed movement of network nodes causes the distances between nodes to constantly change, leading to continuous changes in network characteristics, such as the SINR parameters of nodes. Therefore, it is necessary to establish a motion model to simulate the movement of nodes in dynamic networks.

[0022] In lunar surface communication, nodes do not travel along the lunar surface at a uniform speed, and their direction of motion is variable and influenced by neighboring nodes. When a node accelerates or decelerates along the lunar surface, a two-dimensional Cartesian coordinate system is constructed with a fixed lunar base station as the origin. The Wiener process with drift parameters is used to simulate the node's movement. Based on the characteristics of the Wiener process, the node's velocity change is calculated as follows:

[0023]

[0024]

[0025]

[0026] in, V represents the change in velocity of moving node i along the x-axis during time Δt. x,i,t' V represents the x-axis velocity component at time t'. x,i,t This represents the x-axis velocity component at time t. V represents the change in velocity of moving node i along the y-axis during time Δt. y,i,t' V represents the y-axis velocity component at time t'. y,i,t μ represents the y-axis velocity component at time t. i Denotes the drift parameter, δ i Let represent the parameters that make the Wiener process follow a standard Gaussian distribution, and Δt represent the time from time t1 to time t2.

[0027] In the above formula, and Let μ represent the velocities of vehicle i at times t1 and t2, respectively, and Δt = t2 - t1. i For the drift parameter, δ i It follows a standard Gaussian distribution. The change in the distance between nodes and the distance between nodes after Δt can be calculated from the change in node velocity between nodes:

[0028]

[0029] d(ij,Δt)=Δd ij,Δt +d(ij,t1)

[0030] Where, Δd ij,ΔtThis represents the change in the distance between nodes at interval Δt. This represents the velocity of node i at time t1. μ represents the velocity of node j at time t1. i μ represents the average velocity of node i. j Let Δv represent the average velocity of node j. i,Δt Δv represents the velocity change of node i between points Δt and Δt. j,Δt Let d(ij, Δt) represent the velocity change of node j between Δt, d(ij, Δt) represent the distance between nodes after Δt, d(ij, t1) represent the distance between nodes at time t1, and Δt represent the time from time t1 to time t2.

[0031] Due to Δv i,Δt and Δv j,Δt If it follows a Gaussian distribution, then based on the principle of linear combination of Gaussian variables, Δd ij,Δt Both d(ij,Δt) and d(ij,Δt) follow a Gaussian distribution.

[0032] 2. Network parameter model

[0033] In the complex dynamic network scenario on the lunar surface, the wireless link status is constantly changing. Based on the analysis of network characteristics, the following factors directly affect link connection changes: First, from the node's perspective, the SINR of the receiving node needs to be greater than the reception threshold to ensure that the receiving node can successfully receive data transmitted by the sending node. In addition, network nodes need sufficient buffer queue length to perform store-and-forward processing on the received data. Second, from the link's perspective, to ensure the stability of the wireless link, the link needs to have availability, that is, even when nodes are moving at high speed, they still need to remain within each other's communication range for a specified period of time to ensure the link's continuous availability. Therefore, this invention will use SINR, node power consumption, buffer queue length, and link availability as parameters to construct the network topology.

[0034] Define a set M = {S, E, B, L}, where S = 1 indicates that the SINR of the receiving node is greater than the acceptance threshold; E = 1 indicates that the node's energy can support normal operation; B = 1 indicates that the receiving node has no buffer space to accept transmitted data; and L = 1 indicates that the link is available. These four network parameters jointly determine the link connection state. Only when all four parameters are simultaneously 1 can the conditions for establishing a link be met. Define the link connection state space as C = {0, 1}, where C = 1 indicates that the link between nodes i and j is in a connected state.

[0035] The threshold is a critical value at which the received useful signal is greater than the interference noise.

[0036] 2.1 SINR

[0037] In dynamic networks, the high mobility of nodes leads to frequent changes in the distance between nodes and signal interference, thus affecting the SINR value of the nodes. Furthermore, during communication on the lunar surface, due to the long signal propagation distance, the signal may be affected by interference and noise. Therefore, when calculating the signal-to-noise ratio (SNR), factors such as signal propagation loss, noise power, and signal power need to be considered. Thus, the SINR calculation formula is:

[0038]

[0039] When node i sends a data packet to node j at time t1, the received signal strength of node j after time Δt, after long-distance path loss, is:

[0040]

[0041] In the above formula, P t The transmission power of the sending node, The path propagation loss is calculated using the following formula:

[0042]

[0043] In the above formula, the parameter ρ ij ∈{ρ L ,ρ NL} represents the shadow fading situation between nodes i and j, where ρ L =1,ρ NL =ρ represents the signal propagation between nodes, which are line-of-sight (LOS) propagation and non-line-of-sight (NLOS) propagation respectively. ρ is the shadow fading factor, calculated as ρ = exp(-σ 2 / 2+σψ) and σ=σ (dB) log(10) / 10, σ (dB) Let be the standard deviation of the log-normal shading, and ψ be the standard normal quantity. Additionally... Here, α represents the propagation parameters affected by the lunar environment, d represents the path loss exponent, and d represents the node spacing.

[0044] The strength of the interference signal received by node j after Δt is:

[0045]

[0046] Among them, P t Indicates the transmission power of the sending node. This represents the propagation loss value, and N0 represents additive white Gaussian noise. Let represent the number of nodes that interfere with node j at time t1+Δt, defined as the number of neighboring nodes of node j at that time. Therefore, the number of interfering nodes is calculated as follows:

[0047]

[0048]

[0049] In the two formulas above, N is the number of nodes in the dynamic network, and λ k Indicates whether node k is a neighbor node, λ k ∈{0,1}, when node k is within the transmission range of node j, λ k =1, otherwise λ k =0, Let be the probability that node k is within the transmission range of node j after passing through Δt. This represents the distance between node k and node j after Δt, and the maximum communication range of node R.

[0050] Based on the above analysis, if node i sends a data packet to node j at time t1, then the SINR strength of node j after Δt is:

[0051]

[0052] in, P represents the SINR intensity of node j after Δt. t Indicates the transmission power of the sending node. Indicates path propagation loss. This represents the strength of the interference signal received by node j after Δt.

[0053] For receiving node j to successfully receive data, its SINR value must be greater than the receiving threshold γ0. Therefore, the probability that the receiving node's SINR value is greater than the threshold is calculated as follows:

[0054]

[0055] Where U(s) represents the probability that the SINR value is greater than the threshold, and P{} represents the probability calculation operation, P t Indicates the transmission power of the sending node. This represents the propagation loss of path ij. γ represents the strength of the interference signal received by node j after time Δt, and γ0 represents the threshold.

[0056] 2.2 Node Energy Consumption

[0057] When performing missions on the lunar surface, the rover is equipped with solar panels to provide energy. Once the energy is depleted, it needs to wait for replenishment to continue exploration and data transmission. Therefore, node energy consumption is defined as the energy remaining on the rover at the current moment. The power generation of the solar panels is related to the area of ​​the solar panels and the intensity of solar radiation. Assuming the solar panels receive sunlight at the optimal angle, the formula for calculating the output power of the solar panels is P. t as follows:

[0058] P v =ξ(t)×S

[0059] Among them, P v Let ξ(t) represent the output power of the solar panel, which is the input power for charging the rover at the current moment, and S represent the area of ​​the solar panel. The solar radiation intensity on the lunar surface is affected by factors such as time and the lunar environment; therefore, the formula for calculating the solar radiation intensity on the lunar surface is:

[0060] ξ(t)=ξ0cos 2 (theta(t))cosθ

[0061] Where ξ(t) is the solar radiation intensity on the lunar surface, ξ0 is the solar coefficient, θ is the lunar angular velocity, and theta(t) is the solar altitude angle, calculated as follows:

[0062] sin(theta(t))=sinπsindelta+cosπcosdelta

[0063] In the above formula, delta represents the solar declination, and the calculation formula is as follows:

[0064] sin(delta(t))=sinαsin(θ e t+θ o )

[0065] In the above formula, α is the lunar inclination angle, and θ is the lunar tilt angle. e Let θ be the angular velocity of the moon's rotation. o This is the solar right ascension. Therefore, the energy consumption calculation formula for the probe node is:

[0066] P = P0 + (P v -P e )×Δt

[0067] In the above formula, P represents the remaining energy of the node at the current time after a time interval Δt, P0 represents the initial energy of the probe, and P e This represents the rated output power of the probe. Therefore, the probability that the node's remaining energy at the current moment can ensure the node's normal operation is:

[0068] U(e) = P{P > 0}

[0069] 2.3 Link Availability

[0070] When the distance between nodes is less than the communication range, direct communication between the nodes is considered possible. Link availability is defined as the probability that direct communication between two nodes remains available for a specified period of time, denoted as a time interval Δt. The probability density function f(T) represents the probability that nodes can still communicate for link duration T, as shown in the following equation:

[0071]

[0072] Where R represents the maximum communicable range between nodes, and T represents the duration. Let be the mean and variance of the velocity difference between node i and node j, respectively.

[0073]

[0074]

[0075] Where, μ i μ j Let i and j represent the average velocities of nodes i and j, respectively. Let i and j represent the velocity variances of nodes i and j, respectively.

[0076] If a link exists between two nodes, and the duration of that link's availability depends on their current positions and relative velocities, then the duration of availability can be expressed as:

[0077]

[0078] Among them, T P Indicates the duration of link availability. Let μ represent the velocities of nodes i and j at the current time. i μ j Let Δv represent the velocity variances of nodes i and j, respectively. i,Δt Δv j,Δt These represent the changes in nodal velocity after a time interval Δt. This represents the distance between nodes i and j within the communication range of r.

[0079] Therefore, according to the definition of link availability, in lunar communication, when node i sends a data packet to node j, the link availability, i.e., the probability that the nodes can still communicate after a period of time, is:

[0080]

[0081] Where U(a) represents the probability that the link between nodes will still be available after a period of time, and T PLet f(T) represent the duration of link availability, f(T) represent the probability density function that nodes can still communicate during the link's duration T, and R represent the maximum communicable range between nodes. Let represent the mean and variance of the velocity difference between node i and node j, respectively.

[0082] 2.4 Cache queue length

[0083] To ensure successful data reception, receiving nodes need a buffer queue length to receive or process data. Assuming all nodes in the network have the same maximum buffer queue length F, and at time t1, the buffer queue length used by the receiving node is a, then the node's buffer queue length is b = Fa. For receiving vehicle node j, the reception and transmission of data packets within time Δt can change the vehicle's buffer queue length. This invention will analyze the changes in the buffer queue in the following two cases:

[0084] Case 1: Within time Δt, receiving vehicle j successfully receives data sent from its neighboring vehicle, and the buffer queue of receiving vehicle j increases.

[0085] Scenario 2: Within time Δt, receiving vehicle j successfully sends a data packet to other vehicles, and the buffer queue of receiving vehicle j is reduced.

[0086] Assume that the arrival of data packets at nodes in the network approximately follows a Poisson distribution, with an average arrival probability of λ packets / s; assume that the processing speed of network nodes in the buffer queue is h packets / s. The probability of successful packet transmission follows a binomial distribution. Given that a node processes m data packets, the probability of a node sending a data packet incorrectly is p. e The following formula can be obtained:

[0087]

[0088]

[0089] Among them, f x and f y Let x and y be the probability mass functions for the number of data packets received and sent, respectively. f x and f y Let be the Poisson distribution function and the binomial distribution function, respectively. Therefore, the probability that the length of the receiving vehicle j's buffer queue is still less than the maximum allowed buffer queue length after time Δt is:

[0090]

[0091] Where U(q) represents the probability that the buffer queue length is still less than the maximum allowed buffer queue, x and y represent the number of received data packets and the number of sent data packets, respectively, b represents the buffer queue length of the node, and f x and f y Let p represent the probability mass functions for the number of data packets and the number of data packets sent, respectively; λ represent the average arrival probability of data packets; Δt represent the time from time t1 to time t2; a represent the length of the receiving node queue; and p represent the probability mass functions for the number of data packets and the number of data packets sent, respectively. e This represents the probability of an error in the data packets sent by a node. This represents a factorial operation on the data already present in the current node i. p represents the number of permutations and combinations containing data packet errors in the received data packets. s This represents the probability of successfully receiving a data packet, and S represents the maximum number of data packets.

[0092] 3. Dynamic topology prediction modeling

[0093] The link connection status is determined by the SINR status of the receiving node, the link availability, and the buffer queue length. When all three parameters are in state 1, the link connection status in the dynamic network is considered to be "connected"; otherwise, the link is in a "faulted" state.

[0094] Wireless link l between node i and node j ij The current state is "connected", that is, C. ij When = 1, the probability that the next link connection state is still "connected" is:

[0095] P c (C ij =1,C ij =1) =U(s) 11 U(e) 11 U(a) 11 U(q) 11

[0096] Therefore, we can calculate the link l ij The transition probability from the "connected" state to the "faulty" state is:

[0097] P c =(C ij =1,C ij '=0)=1-U(s) 11 U(e) 11 U(a) 11 U(q) 11

[0098] Similarly, link l can be calculated. ij The transition probability from the "fault" state to the "connected" state is:

[0099] P c (C ij =0,C ij =1) =U(s) 01 U(e) 01 U(a) 01 U(q) 01

[0100] +U(s) 01 U(e) 11 (U(a) 11 U(q) 11 +U(a) 01 U(q) 11 +U(a) 11 U(q) 01 )

[0101] +U(s) 11 U(e) 11 (U(a) 01 U(q) 11 +U(a) 11 U(q) 01 +U(a) 01 U(q) 01 )

[0102] +U(s) 11 U(e) 01 +U(s) 01 U(e) 11

[0103] The link l can be obtained from the above calculations. ij The connection state transition matrix is:

[0104]

[0105] Among them, C ij Represents the link state transition matrix. This represents the probability that the link was unavailable at the previous time step but will be available at the next time step. P represents the probability that the link is available at both the previous and next time steps. c (C ij =1,C ij '=1) represents the probability that the next link connection state will still be "connected", C ij The wireless link l between node i and node j ij The current state, U(s) 11 U(e) 11 U(a) 11 U(q) 11These represent the probabilities that the link signal-to-noise ratio is greater than the receiving threshold at the current moment and will still be greater than the receiving threshold at the next moment, the probabilities that the current node's energy consumption and the next moment can both support normal operation, the probabilities that the link is available at the current moment and will still be available at the next moment, and the probabilities that the current node's buffer queue length can complete data buffering and forwarding and will still be available at the next moment.

[0106] In dynamic network scenarios, considering a network with N nodes, the maximum number of links between nodes is E ≤ N(N-1) / 2. In reality, some nodes in the network will never be able to establish links for data transmission. Therefore, the maximum set of potential links E in a dynamic network is... max (Including "faulty" and "connected" links) and node set V max Defined as a "maximum network," and assumed to be a simple undirected graph, as shown in Figure x, an example of a maximum network is given. A network is considered invalid if a link in the network does not belong to the maximum set of potential links. Figure 3 The diagram illustrates an example of an invalid network state under the maximum network state condition, in which link l 23 This indicates an invalid link. In the network status example diagram, solid lines represent links that are connected, while dashed lines represent links that are faulty.

[0107] Define an N×N 0,1 adjacency matrix A = [a ij ] N×N Describes the state of dynamic network topology, where a ij ∈{0,1},i,j=1,2,...N, when a ij =1 indicates link l ij Maintain the connection state, otherwise, when a ij =0 indicates link l ij This is a faulty link. Additionally, using a set... Denotes the set of all network topology states, where |E max | represents the number of links in the potential link set of the maximum network state. Since each link has both connected and faulty states, the total number of network topology states is...

[0108] Based on the characteristics of dynamic networks, a link connection state transition matrix is ​​derived to characterize link changes. Based on accurate predictions of link state changes, a dynamic topology evolution model is constructed to represent the evolution of the network topology. By traversing all link connection state changes in the largest network, the network topology state transition matrix is ​​calculated, where the network topology state transition probabilities are calculated as follows:

[0109]

[0110] Where q(A,A') represents the network topology state transition probability, A and A' represent the current network topology state and the next time step network topology state, respectively, and l ij E represents the wireless link between node i and node j. max Represents the largest set of potential links in a dynamic network. Let a represent the probability of the link transition matrix. ij and a ij These are the links l corresponding to network topology states A and A′, respectively. ij state.

[0111] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the topology of a dynamic lunar network, characterized in that, include: When the node accelerates or decelerates along the lunar surface, a two-dimensional rectangular coordinate system is constructed with the fixed base station on the lunar surface as the origin, and the movement of the node is simulated using the Wiener process with drift parameters. SINR, node energy consumption, buffer queue length, and link availability were selected as parameters to establish a network topology for fixed base stations and mobile nodes. The network topology is constructed using SINR, node energy consumption, buffer queue length, and link availability as parameters, including: SINR, node energy consumption, buffer queue length, and link availability are defined as a set. Define the link connection state space as , This indicates that the link between nodes i and j is in a connected state. This indicates that the SINR of the receiving node is greater than the receiving threshold; This indicates that the node's energy is sufficient to support normal operation; This indicates that the receiving node does not have buffer space to accept transmitted data. The availability of a link is determined by four parameters: SINR, node energy consumption, buffer queue length, and link availability. When all four parameters are in state 1, the link connection status in the dynamic network is considered "connected"; otherwise, the link is in "faulted" state. Furthermore, the conditions for establishing a link can only be met when all four parameters are 1. The probability of link "connection" in the network topology under the mobile node's simulated mobile state is calculated based on the parameter index. The probability of link "connection" in the next time step is calculated based on the calculated parameter index probability. The link state transition matrix is ​​generated based on the probability of link "connection" in the next time step. The network topology state transition probability is calculated based on the generated link state transition matrix to characterize the evolution of the network topology.

2. The lunar surface dynamic network topology prediction method according to claim 1, characterized in that, The movement of the nodes is simulated using a Wiener process with drift parameters, including: Based on the characteristics of the Wiener process, determine the velocity changes of the moving nodes; The change in the distance between the moving nodes is calculated based on the determined speed change of the moving nodes.

3. The lunar dynamic network topology prediction method according to claim 2, characterized in that, Determine the speed changes of the moving node, including: in, Indicates a moving node In time internal velocity change, Indicates a moving node In time The change in velocity along the x-axis within the space. Indicates a moving node In time The change in velocity along the y-axis within the space. Indicates the drift parameter, Denotes the parameters that make the Wiener process follow a standard Gaussian distribution. Indicates from Time's up The time of a moment.

4. The lunar dynamic network topology prediction method according to claim 2, characterized in that, Calculate the change in the distance between moving nodes, including: in, Indicates the process Spacing between subsequent nodes Indicates the process Spacing between subsequent nodes express The spacing between time nodes Representing separate nodes and nodes , Indicates from Time's up The time of a moment.

5. The lunar dynamic network topology prediction method according to claim 1, characterized in that, The probability of a link "connection" in the network topology under the simulated mobility state of a mobile node is calculated, including the following parameters: The probability that the SINR value is greater than the threshold: The probability that node energy consumption ensures normal node operation: Link availability probability: The probability that the buffer queue length is still less than the maximum allowed buffer queue length: in, This indicates the probability that the SINR value is greater than the threshold. This indicates an operation to calculate probability. Indicates the transmission power of the sending node. Representing a path Propagation loss, Indicates that node j is in time The strength of the interference signal received later, Indicates the threshold; This represents the probability that the node's energy consumption ensures its normal operation. Indicates the process Remaining energy at the current time point after the specified time; This represents the probability that the link between nodes will still be available after a certain period of time. Indicates the duration of link availability. This represents the probability density function indicating that nodes can still communicate within a link's duration T. Indicates the maximum communication range between nodes. , Let represent the mean and variance of the velocity difference between node i and node j, respectively. This indicates the probability that the buffer queue length is still less than the maximum allowed buffer queue length. , These represent the number of data packets received and the number of data packets sent, respectively. Indicates the length of the node's cache queue. and The probability mass functions representing the number of data packets and the number of data packets sent, respectively. This represents the average probability of data packet arrival. Indicates from Time's up The time of moment, Indicates the length of the receiving node queue. This represents the probability of an error in the data packets sent by a node. This represents a factorial operation on the data already present in the current node i. This represents the number of permutations and combinations of data packets containing errors in the received data packets.

6. The lunar dynamic network topology prediction method according to claim 1, characterized in that, The probability of the link "connection" at the next time step is calculated based on the calculated parameter index probability, including: in, This indicates the probability that the next link connection will remain in the "connected" state. The wireless link l between node i and node j ij The current state, These represent the probabilities that the link signal-to-noise ratio is greater than the receiving threshold at the current moment and will still be greater than the receiving threshold at the next moment, the probabilities that the current node's energy consumption and the next moment can both support normal operation, the probabilities that the link is available at the current moment and will still be available at the next moment, and the probabilities that the current node's buffer queue length can complete data buffering and forwarding and will still be available at the next moment.

7. The lunar dynamic network topology prediction method according to claim 1, characterized in that, Generate a link state transition matrix based on the probability of the link "connecting" at the next time step, including: in, Represents the link state transition matrix. This represents the probability that the link was unavailable at the previous time step but will be available at the next time step. This represents the probability that the link is available at both the previous and next time points.

8. The lunar dynamic network topology prediction method according to claim 1, characterized in that, The network topology state transition probability is calculated based on the generated link state transition matrix to characterize the evolution of the network topology, including: in, Represents the network topology state transition probability. and These represent the current network topology state and the network topology state at the next time step, respectively. This represents the wireless link between node i and node j. Represents the largest set of potential links in a dynamic network. Represents the probability of the link transition matrix. and These are the network topology states. and The corresponding link l ij state.

Images