A real-time ground network simulation system for dynamic topology

Abstract

A real-time ground network simulation system for dynamic topology, the system consists of two parts: (1) a plurality of network nodes, (2) a real-time network simulation system; the plurality of network nodes are connected to the real-time network simulation system; wherein the network nodes: 1) generate real data packets; 2) receive data packets from the real-time network simulation system through real network ports and parse the data packets; 3) receive the IDs of connected nodes in a collision domain and the real-time packet loss rate of links between connected nodes in the collision domain sent by the real-time network simulation system; the system flow consists of eight steps: step 1, user parameter setting; step 2, collision domain node perception; step 3, network link packet loss rate calculation; step 4, network topology update judgment; step 5, network link parameter configuration; step 6, network node data packet sending; step 7, node position, wireless channel parameters, simulation time granularity update; step 8, whether the set simulation total duration is reached.

Description

Technical Field

[0001] The invention designs a real-time ground network simulation method and system oriented to dynamic topology, and relates to the technical field of network communication simulation. Background Art

[0002] When a new communication device is designed and produced, it needs to go through a series of tests to determine whether it meets the design requirements and design goals of the communication device. Among them, the communication performance of the new device under large-scale communication conditions is an important verification indicator. In order to solve the problem of device communication performance verification, people use network simulation technology.

[0003] The existing network simulation methods are mainly discrete event-based simulation software such as NS-3, OPNET, OMNET++, QualNet, and virtualization-based network simulation platforms. NS-3 and other discrete event simulator-based simulation software have good simulation effects, but they themselves run according to the tick inside the program and are not real-time. When a data packet is sent to another node, the simulator puts it into the event queue according to the time. After processing the events in the queue, the global variables in the tool will change, and then the next event will be processed.

[0004] The virtualized network simulation platform instantiates nodes in the network and isolates the resources between these nodes. The simulation tool uses virtual switches to connect virtual nodes, generate custom network topology relationships, and can run real applications in virtual nodes to achieve real-time simulation of real data flows in the network topology. Therefore, compared with discrete event-based network simulation software, the virtualized network simulation platform has the characteristics of high scalability, high fidelity, and high flexibility.

[0005] In real networks, 1) the geographic location of network nodes 2) wireless channel parameters 3) network topology connectivity are all quantities that change over time. Changes in these three variables will affect changes in network simulation link parameters. Existing network simulation software generally sets network simulation link parameters as fixed parameters when performing network simulation, and does not modify the network link parameters or only modifies them based on one or two of the above three variables.

[0006] The applicant previously disclosed a large-scale network simulation system and a dynamic resource scheduling method in the application “A large-scale network simulation system and a resource dynamic scheduling and allocation method” (application publication number: CN114050961A), in which network link parameters are all pre-set fixed values ​​when performing network simulation. The inventor provides an SDN-based space information network semi-physical centralized simulation platform and an implementation method thereof in the “SDN-based space information network semi-physical centralized simulation platform and an implementation method thereof” (application publication number: CN106301911A), in which the simulated network nodes are satellites and the changes in the geographical location of the network nodes and the network topology connectivity are taken into account when performing network simulation, but the influence of the changes in wireless channel parameters over time is not considered. Summary of the invention

[0007] The purpose of the present invention is to solve the problem that 1) the simulation authenticity of event-driven network simulation software such as NS-3 and OPNET is insufficient; 2) the network link parameters of the existing real-time network simulation system fail to fully consider the changes of a) the geographical location of network nodes b) the wireless channel parameters c) the network topology connectivity over time. The present invention discloses a real-time ground network simulation method and system for dynamic topology, and proposes a network simulation method with variable network link parameters that comprehensively considers the link signal-to-noise ratio and the number of nodes in the collision domain. The method 1) can calculate the network link parameter according to the network link packet loss rate PL ij (t) The rate of change compared to the previous moment, adaptively adjust the network update time 2) Compared with the traditional complex network channel modeling, it can i (t)b) The signal-to-noise ratio (SNR) of the link between nodes at time t ij (t)c) The number of nodes m within the collision domain of each node at time t and the packet loss rate PL of the network link between each node ij (t) to establish a simple mapping relationship between them, reducing the simulation resource overhead 3) can be based on PL ij (t) By comparing with the preset threshold, the real-time network link is judged to achieve dynamic network topology update, which improves the authenticity and efficiency of network simulation results.

[0008] To achieve the above object, the technical solution adopted by the present invention is: a real-time ground network simulation system for dynamic topology, the block diagram of the embodiment is as follows Figure 1 As shown, the system consists of two parts: (1) a number of network nodes, (2) a real-time network simulation system; the number of network nodes are connected to the real-time network simulation system;

[0009] Among them, the network node: 1) generates a real data packet and sends the data packet to the real-time network simulation system through the real network port; 2) receives the data packet from the real-time network simulation system through the real network port and parses the data packet; 3) receives the connected node ID in the collision domain and the real-time packet loss rate of the link between the connected nodes in the collision domain sent by the real-time network simulation system; 4) configures the network link parameters reaching each network node according to the connected node ID in the collision domain and the real-time packet loss rate of the link between the connected nodes in the collision domain received from the real-time network simulation system.

[0010] The functions of the real-time network simulation system include: 1) allowing users to set the total duration of network simulation, the number of simulation nodes, the virtual IP address of each network node during simulation, the geographical location of the network nodes at each time during simulation, the signal-to-noise ratio between the geographical locations at each time, the collision domain range of each network node, and the modulation and coding scheme (MCS) of each network node sending packets; 2) counting the number of remaining network nodes in the collision domain of each network node; 3) calculating the real-time packet loss rate of the network link between each node and the connected nodes in its collision domain; 4) calculating the network topology connectivity status at each time; 5) receiving data packets sent by network nodes from real network ports; 6) forwarding data packets sent by network nodes to real network ports.

[0011] The dynamic topology represents a time-varying graph (VTG). A VTG can be composed of To characterize. Among them, represents a VTG, V represents the node parameter set in the VTG, E represents the link parameter set in the VTG, represents the life cycle of the VTG, and ρ represents the connectivity relationship of the nodes in the VTG.

[0012] Furthermore, the network node parameter set V in the present invention is composed of the geographical location Loc of the network node related to time. i (t) and the MCS when the node sends a data packet i (t) characterization, that is, V = {Loc i (t),MCS i (t)}; In the present invention, the link parameter set E is composed of the signal-to-noise ratio SNR between each node link related to time ij (t) and the packet loss rate PL between each connected node ij (t) characterization, that is, E = {SNR ij (t),PL ij (t)}; The life cycle of VTG in the present invention It is represented by the total network simulation time T set by the user during initialization, that is, In the present invention, the node connectivity relationship ρ is represented by the network topology connectivity state matrix Topo(t) related to time, that is, ρ=Topo(t).

[0013] The workflow of the real-time ground network simulation system for dynamic topology is shown in the figure below: Figure 2 As shown, the system workflow is mainly divided into the following eight steps: Step 1, user parameter setting; Step 2, collision domain node perception; Step 3, network link packet loss rate calculation; Step 4, network topology update judgment; Step 5, network link parameter configuration; Step 6, network node data packet sending; Step 7, node location, wireless channel parameters, simulation time granularity update; Step 8, whether to reach the set total simulation time judgment;

[0014] Step 1: User parameter settings

[0015] The user can set the following parameters through the parameter configuration script: 1) the total duration of network simulation T 2) the number of simulation nodes n 3) the virtual IP address of each network node during simulation 4) the geographical location function Loc of each network node during simulation i (t)5) Signal-to-noise ratio function SNR between network nodes ij (t) 6) The connectivity matrix of the initial network topology Topo (0) 7) The collision domain range Dmax of each network node i 8) Modulation and coding function MCS of packets sent by each network node i (t)9) Initialize the network simulation time granularity dt.

[0016] After completing the above configuration, the time-varying topology VTG can be obtained by

[0017] Step 2: Collision domain node perception

[0018] From the time-varying topology VTG, take the network topology G at time t = (V(t), E(t)) = ({Loc i (t),MCS i (t)},{SNR ij (t),PL ij (t)}) and the network topology connectivity matrix Topo(t) at time t, through the geographical location function Loc i (t) The geographical location information (x, y, z) of each node at time t and the collision domain range Dmax of each node i As well as the connectivity relationship of each node in the connectivity state matrix Topo(t) of the network topology at time t, the number of connected nodes m in the collision domain of each node is counted.

[0019] The specific steps include:

[0020] Step 2.1: Calculation of the distance matrix D(t) of each node

[0021] The system calculates the distance between each node and obtains the node distance matrix D(t).

[0022] Assume that the geographical coordinates of two nodes i and j at time t are Loc i (t) = (x i ,y i ,z i ), Loc j (t) = (x j ,y j ,z j ), the distance between the two nodes is d ij (t):

[0023]

[0024] Then the distance matrix D(t) is:

[0025]

[0026] Step 2.2: The connected node set P in the collision domain of each node i (t) Calculation

[0027] Traverse the node distance matrix D(t) and the network topology connectivity matrix Topo(t) row by row, and determine whether each element in D(t) has a collision domain range Dmax. i and whether the value of each element in Topo(t) is 1 (the value of the element in the i-th row and j-th column in Topo(t) is 1, indicating that nodes i and j are connected), and the statistics are less than the collision domain range Dmax i And the connected node ids, get the connected node set P in the collision domain of each node i (t)(i=1,2,…,n).

[0028] P i (t) = {id1, id2, ..., id m},i=1,2,…,n

[0029] Step 3: Calculate the packet loss rate of the network link

[0030] According to the signal-to-noise ratio function matrix SNR between each network node at time t ij (t), the modulation and coding function MCS of each node i (t) and the connected node set P in the collision domain of each node i (t), calculate the packet loss rate PL between each node link ij (t).

[0031] The specific steps include:

[0032] Step 3.1: Baseline packet loss rate p between network links of each node loss (MCS i (t),SNR ij (t))Calculation

[0033] According to the preset signal-to-noise ratio function SNR between each network node ij (t) and the MCS of the sending node i (t), calculate each MCS i (t) and SNR ij The baseline packet loss rate p under (t) loss (MCS i (t),SNR ij (t)):

[0034]

[0035]

[0036]

[0037]

[0038] Step 3.2: Correction of baseline packet loss rate between network links of each node

[0039] According to the connected node set P in the collision domain of each node at time t obtained in step 2 i (t) the number of elements, the distance d between node i and node j ij (t) and the collision domain range Dmax of node i i For the baseline packet loss rate p loss (MCS i (t),SNR ij (t)) is corrected to obtain the packet loss rate PL of each link ij (t):

[0040]

[0041] Among them, m is the connected node set P in the collision domain of each node at time t i (t) is the number of elements, PLC is the packet loss rate increasing constant, Dmax i is the collision domain range of the i-th node.

[0042] Step 4: Network topology update judgment

[0043] Determine the PL calculated in step 3 ij (t) Is it greater than the set network link connectivity threshold P? connect , if PLij (t)≥P connect , then the link between node i and node j is considered to be disconnected, Topo(t) is updated, the element in the i-th row and j-th column and the element in the j-th row and i-th column in Topo(t) are set to 0, and the process jumps to step 2; otherwise, Topo(t) is not updated, and the process jumps to step 5.

[0044] Step 5: Network link parameter configuration

[0045] According to the IDs of the connected nodes in the collision domain of each node at time t obtained in step 2 and the packet loss rate of each link at time t calculated in step 3, flow control configuration is performed on each link to achieve dynamic adjustment of the network link status.

[0046] The specific steps include:

[0047] Step 5.1: Send the configuration parameters

[0048] The network simulation system sends 1) the ID of the connected nodes in the collision domain of each node at each time and 2) the link packet loss rate between each connected node at each time calculated in step 2 to each network node.

[0049] Step 5.2: Clear the network link configuration of each node at the last moment

[0050] Each network node clears the link configuration parameters at the last moment.

[0051] Step 5.3: Configure network link parameters between nodes

[0052] According to the connected node IDs in the collision domain and the packet loss rates between the connected nodes transmitted by the network simulation system, each network node resets the parameters of each network link.

[0053] Step 6: Network node data packet sending

[0054] Each sending network node sends a network data packet to the currently connected node.

[0055] Step 7: Update node location, wireless channel parameters and simulation time granularity

[0056] Determine the packet loss rate PL of each node link calculated at the current moment ij (t) and the packet loss rate PL of each node link at the previous moment ij Rate of change of (t-dt) Is it less than the set network simulation time granularity threshold P? change ,like Then dt=1.5×dt, otherwise, dt=10ms.

[0057] After completing the simulation time granularity dt update, the real-time simulation system changes the geographical location of the nodes in the network from Loc i (t) becomes Loc i (t+dt), the signal-to-noise ratio of the wireless channel between nodes in the network is SNR ij (t) becomes SNR ij (t+dt).

[0058] Step 8: Determine whether the total simulation time has been reached

[0059] The real-time network simulation system determines whether the current time t+dt is greater than the network simulation time T set by the user. If t+dt>T, the simulation ends. Otherwise, jump to step 2.

[0060] The present invention proposes a real-time ground network simulation method and system for dynamic topology. In view of the fact that the geographical location of network nodes, wireless channel parameters, and network topology connectivity status in real networks change over time, a network simulation method with variable network link parameters that comprehensively considers the link signal-to-noise ratio and the number of nodes in the collision domain is proposed. 1) This method can adaptively adjust the network simulation update granularity according to the statistically obtained network link packet loss change rate; 2) It is more real-time than simulation software based on discrete event simulation such as NS-3 and OPNET; 3) Compared with the existing virtualized network simulation platform and the traditional complex channel modeling method, the present invention proposes a simple mapping relationship between a statistically obtained network link packet loss rate and a) the signal-to-noise ratio of the time-varying channel b) the ratio of the distance between communication nodes to the collision domain range c) the number of other nodes in the collision domain, which realizes the adaptive change of the dynamic topology network link packet loss rate and further improves the authenticity and efficiency of the simulation results.

[0061] The present invention proposes a network simulation method for variable network link parameters that comprehensively considers the link signal-to-noise ratio and the number of nodes in the collision domain. The method 1) can calculate the network link packet loss rate PL according to the network link packet loss rate PL. ij (t) The rate of change compared to the previous moment, adaptively adjust the network update time 2) Compared with the traditional complex network channel modeling, it can i (t)b) The signal-to-noise ratio (SNR) of the link between nodes at time t ij (t)c) The number of nodes m within the collision domain of each node at time t and the packet loss rate PL of the network link between each node ij (t) to establish a simple mapping relationship between them, reducing the simulation resource overhead 3) can be based on PL ij (t) By comparing with the preset threshold, the real-time network link is judged to achieve dynamic network topology update, which improves the authenticity and efficiency of network simulation results.

[0062] The beneficial effects of the present invention are as follows: 1) It solves the problem of insufficient authenticity of traditional network simulation based on event-driven simulation; 2) It solves the problem of changes in simulated network link parameters over time when the geographical location of network nodes, wireless channel parameters, and network topology connectivity status change over time; 3) It proposes a network simulation method that comprehensively considers the link signal-to-noise ratio and the number of nodes in the collision domain to make network link parameters variable; 4) It proposes a network simulation method that adaptively adjusts the network update time. Compared with simulation software based on discrete event simulation such as NS-3 and OPNET, it is more real-time, and compared with the existing virtualized network simulation platform, it further improves the authenticity and efficiency of simulation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a system composition diagram of a real-time ground network simulation method and system for dynamic topology according to the present invention.

[0064] Figure 2 The present invention is a system workflow diagram of a real-time ground network simulation method and system for dynamic topology.

[0065] Figure 3 The network topology at time t0 of an embodiment of a real-time ground network simulation method and system for dynamic topology of the present invention.

[0066] Figure 4 The network topology at time t0-dt of an embodiment of a real-time ground network simulation method and system for dynamic topology of the present invention. DETAILED DESCRIPTION

[0067] The present invention is further described below in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] The present invention first provides a real-time ground network simulation method and system for dynamic topology. Figure 1 The present invention is a system architecture diagram of a real-time ground network simulation method and system for dynamic topology, Figure 2 The system workflow diagram of a real-time ground network simulation method and system for dynamic topology of the present invention is as follows: Figure 3 It is a network topology diagram at time t0 of an embodiment of a real-time ground network simulation method and system for dynamic topology of the present invention; Figure 4 The present invention is a real-time ground network simulation method and system for dynamic topology, and the network topology diagram at the time t0-dt is implemented

[0069] The specific steps are as follows:

[0070] Step 1, User parameter setting

[0071] In this embodiment, the user sets 1) the total network simulation duration T = 3600s, 2) the number of simulation nodes n = 8, 3) the virtual IP addresses of each network node during simulation (192.168.31.x, where x represents the xth network node), 4) the geographical location function Loc i (t) of each network node during simulation, 5) the signal-to-noise ratio function SNR ij (t) of each network node during simulation, 6) the connectivity state matrix Topo(0) of the initial network topology, 7) the collision domain range Dmax of each network node i = 20, 8) the modulation and coding scheme function MCS i (t) of each network node during packet transmission, 9) the network simulation time granularity dt = 10ms

[0072] Please refer to Figure 3 , the network topology at time t0 is as Figure 3 shown. To ensure generality, in this embodiment, dt < t0 < T is taken, that is, t0 is not the initial moment:

[0073] Step 2, Collision domain node sensing

[0074] Take the network topology G = (V(t), E(t)) = ({Loc i (t), MCS i (t)}, {SNR ij (t), PL ij (t)}) at time t and the network topology connectivity state matrix Topo(t) from the time-varying topology VTG. Through the geographical location function Loc i (t) of each network node, obtain the geographical location information (x, y, z) of each node at time t, the collision domain range Dmax of each node i and the connectivity relationship of each node in the network topology connectivity state matrix Topo(t) at time t, and count the number m of connected nodes within the collision domain of each node.

[0075] The specific steps include:

[0076] Step 2.1, Calculation of the distance matrix D(t) of each node

[0077] The system calculates the distances between each node to obtain the node distance matrix D(t).

[0078] Taking node 1 as an example, at time t0, the nodes connected to node 1 in the network topology are node 2 and node 5. Among them, the geographical location of node 1 is set to Loc1(t0) = (0, 0, 2), the geographical location of node 2 is set to Loc2(t0) = (5, 2, 5), and the geographical location of node 5 is set to Loc5(t0) = (3, -10, 4). Then the distance d between node 1 and node 2 is 12 (t0):

[0079]

[0080] The distance d between node 1 and node 5 15 (t0):

[0081]

[0082] Similarly, the distance matrix D(t0) obtained by calculating all other nodes is:

[0083]

[0084] Step 2.2: The connected node set P in the collision domain of each node i (t) Calculation

[0085] Traverse the node distance matrix D(t) and the network topology connectivity matrix Topo(t) row by row, and determine whether each element in D(t) has a collision domain range Dmax. i and whether the value of each element in Topo(t) is 1 (the value of the element in the i-th row and j-th column in Topo(t) is 1, indicating that nodes i and j are connected), and the statistics are less than the collision domain range Dmax i And the connected node ids, get the connected node set P in the collision domain of each node i (t)(i=1,2,…,n).

[0086] In this example, at time t0, the collision domain of node 1 is shown as the dotted circle in the figure, the collision domain range Dmax1 = 20, and the network topology connectivity matrix Topo(t0):

[0087]

[0088] Among them, the element in the i-th row and j-th column is 1, which means that the network topology of node i and node j is connected.

[0089] For node 1, the nodes that are within the collision domain range Dmax1 and connected are node 2 and node 5. Therefore, for node 1, the connected node set P1(t0) in the collision domain is:

[0090] P1(t0)={2,5}

[0091] Step 3: Calculate the packet loss rate of the network link

[0092] According to the signal-to-noise ratio function matrix SNR between each network node at time t ij (t), the modulation and coding function MCS of each node i (t) and the connected node set P in the collision domain of each node i (t), calculate the packet loss rate PL between each node link ij (t).

[0093] The specific steps include:

[0094] Step 3.1: Baseline packet loss rate p between network links of each node loss (MCS i (t),SNR ij (t))Calculation

[0095] According to the preset signal-to-noise ratio function SNR between each network node ij (t) and the MCS of the sending node i (t), calculate each MCS i (t) and SNR ij The baseline packet loss rate p under (t) loss (MCS i (t),SNR ij (t)):

[0096]

[0097]

[0098]

[0099]

[0100] In this example, at time t0, node 1 uses QPSK modulation and coding, and the SNR between node 1 and node 2 is 12 (t0) is 6db, the SNR between node 1 and node 5 15 (t0) is 1db, then the baseline packet loss rate between node 1 and node 2 is p loss (QPSK,SNR 12 (t0)):

[0101]

[0102] The baseline packet loss rate p between node 1 and node 5 loss (QPSk,SNR 15 (t0)):

[0103]

[0104] Step 3.2: Correction of baseline packet loss rate between network links of each node

[0105] According to the connected node set P in the collision domain of each node at each moment obtained in step 2 i (t) the number of elements, the distance d between node i and node j ij (t) and the collision domain range Dmaxx of node i to the benchmark packet loss rate p loss (MCS i (t),SNR ij (t)) is corrected to obtain the packet loss rate PL of each link ij (t):

[0106]

[0107] Among them, m is the connected node set P in the collision domain of each node at time t i (t) is the number of elements, PLC is the packet loss rate increasing constant, Dmax i is the collision domain range of the i-th node.

[0108] In this example, PLC is set to 0.05, and the link packet loss rate PL between node 1 and node 2 at time t0 is 12 (t0):

[0109]

[0110] The packet loss rate LP of the link between node 1 and node 5 15 (t0):

[0111]

[0112] Step 4: Network topology update judgment

[0113] Determine the PL calculated in step 3 ij (t) Is it greater than the set network link connectivity threshold P? connect , if PL ij (t)≥P connect , then the link between node i and node j is considered to be disconnected, Topo(t) is updated, the element in the i-th row and j-th column and the element in the j-th row and i-th column in Topo(t) are set to 0, and the process jumps to step 2; otherwise, Topo(t) is not updated, and the process jumps to step 5.

[0114] In this example, the network link connectivity threshold P cinnect Set to 0.90,

[0115] PL 12 (t0)=0.415<0.90

[0116] PL 15 (t0)=0.790<0.90

[0117] Therefore, skip to step 5.

[0118] Step 5: Network link parameter configuration

[0119] According to the IDs of the connected nodes in the collision domain of each node at each moment obtained in step 2 and the packet loss rate of each link at each moment calculated in step 3, flow control configuration is performed on each link to achieve dynamic adjustment of the network link status.

[0120] The specific steps include:

[0121] Step 5.1: Send the configuration parameters

[0122] The network simulation system sends 1) the ID of the connected nodes in the collision domain of each node at each time and 2) the link packet loss rate between each connected node at each time calculated in step 2 to each network node.

[0123] In this embodiment, the real-time network simulation system runs the following commands:

[0124] echo "2 0.415$5 0.790%%"> / dev / udp / 192.168.31.1 / 14555

[0125] The above command combines the connected node set P1(t0) in the collision domain of node 1 at time t0 counted in step 2 and the PL calculated in step 3 12 (t0) and PL 15 (t0) is sent via UDP data packet to port 14555 of node 1 (virtual IP address is 192.168.31.1).

[0126] Step 5.2: Clear the network link configuration of each node at the last moment

[0127] Each network node clears the link configuration parameters at the last moment.

[0128] See also Figure 4 , the topological connectivity at time t0-dt is as follows Figure 4 As shown, node 1 is connected to node 2, node 3, and node 5. After node 1 receives the UDP data packet sent from port 14555 at time t0, node 1 runs the following command:

[0129] tc qdisc del dev eth0 root

[0130] The above command deletes the queueing rules configured on eth0 network card of node 1 before t0, which are connected to nodes 2, 3, and 5.

[0131] Step 5.3: Configure network link parameters between nodes

[0132] According to the connected node IDs in the collision domain and the packet loss rates between the connected nodes transmitted by the network simulation system, each network node resets the parameters of each network link.

[0133] In this embodiment, at time t0, node 1 runs the following command:

[0134] tc qdisc add dev eth0 root handle 1:cbq bandwidth 1000Mbit avpkt1000cell 8mpu 64

[0135] tc class add dev eth0 parent 1:0classid 1:1cbq bandwidth 1000Mbitrate1000Mbit maxburst 20allot 1514prio 8avpkt 1000cell 8weight 100000Kbit tcclass add dev eth0 parent 1:1classid 1:2cbq bandwidth 10Mbit rate10Mbitmaxburst 20allot 1514prio 2avpkt 1000cell 8weight 1Mbit split1:0boundedtcqdisc add dev eth0 parent 1:2handle 2:netem loss 41.5%

[0136] tc filter add dev eth0 protocol ip parent 1:0prio 1u32 match ipdst192.168.31.2 / 32flowid 1:2

[0137] tc class add dev eth0 parent 1:1classid 1:3cbq bandwidth 10Mbitrate10Mbit maxburst 20allot 1514prio 2avpkt 1000cell 8weight 1Mbit split1:0boundedtc qdisc add dev eth0 parent 1:3handle 3:netem loss 79%

[0138] tc filter add dev eth0 parent 1:0protocol ip prio 1u32 match ipdst192.168.31.3 / 32flowid 1:3

[0139] The above command configures the packet loss rate of the link sent to the IP address 192.168.31.2 (node ​​2) to be 41.5%, and the packet loss rate of the link sent to the IP address 192.168.31.3 (node ​​3) to be 79%.

[0140] Step 6: Network node data packet sending

[0141] Each sending network node sends a network data packet to the currently connected node.

[0142] In this embodiment, at time t0, after node 1 completes the configuration of network link parameters to node 2 and node 5, it runs the following command:

[0143] ping 192.168.31.2

[0144] ping 192.168.31.5

[0145] Step 7: Update node location, wireless channel parameters and simulation time granularity

[0146] Determine the packet loss rate PL of each node link calculated at the current moment ij (t) and the packet loss rate PL of each node link at the previous moment ij Rate of change of (t-dt) Is it less than the set network simulation time granularity threshold P? chanhge , if the packet loss rate change rate of the link between any two nodes satisfies Then dt=1.5×dt, otherwise, dt=10ms.

[0147] After completing the simulation time granularity dt update, the real-time simulation system changes the geographical location of the nodes in the network from Loci (t) becomes Loc i (t+dt), the signal-to-noise ratio of the wireless channel between nodes in the network is SNR ij (t) becomes SNR ij (t+dt).

[0148] In this embodiment, the network simulation time granularity threshold P change Set to 30%.

[0149] The link packet loss rate PL between node 1 and node 2 at time t0 12 (t0) and the packet loss rate PL of each node link at the previous moment 12 Rate of change of (t0-dt) Therefore, dt=10ms.

[0150] Step 8: Determine whether the total simulation time has been reached

[0151] The real-time network simulation system determines whether the current time t+dt is greater than the network simulation time T set by the user. If t+dt>T, the simulation ends. Otherwise, jump to step 2.

[0152] The above-mentioned embodiments are only preferred embodiments of the present invention, and their descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the attached claims.

Claims

1. A real-time ground network simulation system for dynamic topology, characterized in that: The system consists of two parts: (1) a number of network nodes, (2) a real-time network simulation system; a number of network nodes are connected to the real-time network simulation system; Wherein, the network node: 1) generates a real data packet and sends the data packet to the real-time network simulation system through the real network port; 2) receives the data packet from the real-time network simulation system through the real network port and parses the data packet; 3) receives the ID of the connected node in the collision domain and the real-time packet loss rate of the link between the connected nodes in the collision domain sent by the real-time network simulation system; 4) configures the network link parameters reaching each network node according to the ID of the connected node in the collision domain and the real-time packet loss rate of the link between the connected nodes in the collision domain received from the real-time network simulation system; The real-time network simulation system includes: 1) the user sets the total duration of network simulation, the number of simulation nodes, the virtual IP address of each network node during simulation, the geographical location of the network node at each time during simulation, the signal-to-noise ratio between the geographical locations at each time, the collision domain range of each network node, and the modulation coding scheme (MCS) of each network node; 2) counting the number of remaining network nodes in the collision domain of each network node; 3) calculating the real-time packet loss rate of the network link between each node and the connected nodes in its collision domain; 4) calculating the dynamic topology connectivity status of the network at each time; 5) receiving data packets sent by the network node from the real network port; 6) forwarding data packets sent by the network node to the real network port; The dynamic topology represents a time-varying topology VTG. A VTG consists of To characterize; among them, represents a VTG, V represents the node parameter set in the VTG, E represents the link parameter set in the VTG, represents the life cycle of the VTG, ρ represents the node connectivity relationship in the VTG; the network node parameter set V is composed of the geographical location Loc of the network node related to time. i (t) and the MCS when the node sends a data packet i (t) characterization, that is, V = {Loc i (t), MCSi(t)}; the link parameter set E is composed of the signal-to-noise ratio SNR between each node link related to time ij (t) and the packet loss rate PL between each connected node ij (t) characterization, that is, E = {SNR ij (t),PL ij (t)}; VTG life cycle It is represented by the total network simulation time T set by the user during initialization, that is, The node connectivity relationship ρ is represented by the time-dependent network topology connectivity state matrix Topo(t), that is, ρ=Topo(t).

2. The real-time ground network simulation system for dynamic topology according to claim 1 is characterized in that: The system process consists of eight steps: Step 1, user parameter setting; Step 2, collision domain node perception; Step 3, network link packet loss rate calculation; Step 4, network topology update judgment; Step 5, network link parameter configuration; Step 6, network node data packet sending; Step 7, node location, wireless channel parameters, simulation time granularity update; Step 8, whether to reach the set total simulation time judgment; Step 1: User parameter settings: 1) total network simulation time, T 2) number of simulation nodes n, 3) virtual IP address of each network node during simulation, 4) geographical location function of each network node during simulation, Loc i (t), 5) Signal-to-noise ratio function SNR between each network node ij (t), 6) the connectivity matrix Topo(0) of the initial network topology, 7) the collision domain range Dmax of each network node i ,8) Modulation and coding mode function MCSi(t) of each network node, 9) Initialize network simulation time granularity dt; After completing the above configuration, the time-varying topology VTG is obtained: {SNR ij (t),PL ij (t)},T,Topo(t)); Step 2: Collision domain node perception From the time-varying topology VTG, take the network topology G at time t = (V(t), E(t)) = ({Loc i (t),MCSi(t)},{SNR ij (t),PL ij (t)}) and the network topology connectivity matrix Topo(t) at time t, through the geographical location function Loc i (t) The geographical location information (x, y, z) of each node at time t and the collision domain range Dmax of each node i And the connectivity relationship of each node in the connectivity state matrix Topo(t) of the network topology at time t, and count the number of connected nodes m in the collision domain of each node; In step 3, the network link packet loss rate is calculated based on the signal-to-noise ratio function matrix SNR between each network node at time t ij (t), the modulation coding function MCSi(t) of each node, and the connected node set P in the collision domain of each node i (t), calculate the packet loss rate PL between each node link ij (t); Step 4: Network topology update judgment: judge the PL calculated in step 3 ij (t) Is it greater than the set network link connectivity threshold P? connect , if PL ij (t)≥P connect , then the link between node i and node j is considered disconnected, and Topo(t) is updated, and the element in row i and column j and the element in row j and column i in Topo(t) are set to 0, and then jump to step 2; Otherwise, do not update Topo(t) and jump to step 5; Step 5: Configure network link parameters. According to the IDs of connected nodes in the collision domain of each node at time t obtained in step 2 and the packet loss rate of each link at time t calculated in step 3, configure flow control for each link to dynamically adjust the network link status. Step 6: Network node data packet transmission. Each sending end network node sends a network data packet to the node connected at the current moment. Step 7: Update node positions, wireless channel parameters and simulation time granularity. Determine the packet loss rate PL of each node link calculated at the current moment ij (t) and the packet loss rate PL of each node link at the previous moment ij Rate of change of (t-dt) Is it less than the set network simulation time granularity threshold P? chande ,like Then dt = 1.5 × dt, otherwise, dt = 10ms; After completing the simulation time granularity dt update, the real-time simulation system changes the geographical location of the nodes in the network from Loc i (t) becomes Loc i (t+dt), the signal-to-noise ratio of the wireless channel between nodes in the network is SNR ij (t) becomes SNR ij (t+dt); Step 8: Determine whether the total simulation time has been reached. The real-time network simulation system determines whether the current time t+dt is greater than the network simulation time T set by the user. If t+dt>T, the simulation ends. Otherwise, jump to step 2.

3. The real-time ground network simulation system for dynamic topology according to claim 2 is characterized in that: Step 2 includes: Step 2.1: Calculation of the distance matrix D(t) of each node The system calculates the distance between each node and obtains the node distance matrix D(t); Assume that the geographical coordinates of two nodes i and j at time t are Loc i (t) = (x i ,y i ,z i ), Loc j (t) = (x j ,y j ,z j ), the distance between the two nodes is d ij (t): Then the distance matrix D(t) is: Step 2.2: The connected node set P in the collision domain of each node i (t) Calculation Traverse the node distance matrix D(t) and the network topology connectivity matrix Topo(t) row by row, and determine whether each element in D(t) has a collision domain range Dmax. i and whether the value of each element in Topo(t) is 1 (the value of the element in the i-th row and j-th column in Topo(t) is 1, indicating that nodes i and j are connected), and the statistics are less than the collision domain range Dmax i And the connected node ids, get the connected node set P in the collision domain of each node i (t)(i=1,2,…,n), P i (t)={jd1,id2,…,id m },i=1,2,…,n。 4. The real-time ground network simulation system for dynamic topology according to claim 2 is characterized in that: The specific steps in step 3 include: Step 3.1: Baseline packet loss rate p between network links of each node loss (MCS i (t),SNR ij (t))Calculation According to the preset signal-to-noise ratio function SNR between each network node ij (t) and the MCS of the sending node i (t), calculate each MCS i (t) and SNR ij The baseline packet loss rate p under (t) loss (MCS i (t),SNR ij (t)): Step 3.2: Correction of baseline packet loss rate between network links of each node According to the connected node set P in the collision domain of each node at time t obtained in step 2 i (t) the number of elements, the distance d between node i and node j ij (t) and the collision domain range Dmax of node i i For the baseline packet loss rate p loss (MCS i (t),SNR ij (t)) is corrected to obtain the packet loss rate PL of each link ij (t): Among them, m is the connected node set P in the collision domain of each node at time t i (t) is the number of elements, PLC is the packet loss rate increasing constant, Dmax i is the collision domain range of the i-th node.

5. The real-time ground network simulation system for dynamic topology according to claim 2 is characterized in that: Step 5 The specific steps include: Step 5.1: Send the configuration parameters The network simulation system sends 1) the ID of the connected nodes in the collision domain of each node at each time 2) the link packet loss rate between each connected node at each time calculated in step 2 to each network node; Step 5.2: Clear the network link configuration of each node at the last moment Each network node clears the link configuration parameters of the last moment; Step 5.3: Configure network link parameters between nodes According to the connected node IDs in the collision domain and the packet loss rates between the connected nodes transmitted by the network simulation system, each network node resets the parameters of each network link.

Images