A routing selection method based on energy consumption and link quality

Abstract

The present disclosure is about a routing selection method based on energy consumption and link quality, the method comprising: establishing a software-defined flight ad hoc network model, the model comprising a control node and multiple transmission nodes; multiple transmission nodes respectively generate their own attribute data packets within a broadcast interval, and upload their own attribute data packets to the control node respectively; the control node generates a network topology based on all received attribute data packets, and according to the generated network topology, uses an ant colony optimization algorithm to respectively calculate the selection probability of each transmission node, and finally determines the best route for data transmission through the selection probability. The present disclosure can not only ensure low network latency and network overhead, while improving network throughput, but also ensure the balance of energy consumption of all transmission nodes, so that the comprehensive performance of each transmission node is excellent.

Description

Technical Field

[0001] The present disclosure relates to the technical field of wireless network, and in particular to a routing selection method based on energy consumption and link quality. Background Art

[0002] Currently, drones have been widely used in the field of communication. The Flying Ad-Hoc Network (FANET) based on the ad hoc network can provide reliable and real-time network communication between multiple drones, becoming a good solution to the problem of multi-drone collaboration.

[0003] However, as the scale and complexity of wireless networks gradually increase, many problems have arisen in the FANET network architecture. For example, data transmission needs to be forwarded through multiple layers of protocol stacks, which increases network latency; the functional modules in the FANET network architecture are scattered, and repeated calls cause unnecessary waste of resources; limited storage capacity leads to limited routing table capacity and inflexible control. Therefore, in view of the various problems existing in the current FANET routing, it is necessary to improve one or more of the problems existing in the above-mentioned related technical solutions to determine the optimal routing path for data transmission.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the embodiments of the present disclosure is to provide a routing selection method based on energy consumption and link quality to determine an optimal routing path for data transmission.

[0006] According to a first aspect of an embodiment of the present disclosure, a route selection method based on energy consumption and route quality is provided, the method comprising the following steps:

[0007] Establish a software-defined flight ad hoc network model, which includes a control node and multiple transmission nodes;

[0008] The plurality of transmission nodes respectively generate their own attribute data packets within a broadcast interval, and respectively upload their own attribute data packets to the control node;

[0009] The control node generates a network topology according to all the received attribute data packets;

[0010] The control node calculates the selection probability of each transmission node respectively by using an ant colony optimization algorithm according to the network topology, and determines the best route for data transmission through the selection probability;

[0011] The plurality of transmission nodes repeat the above steps at the next broadcast interval, and the control node regenerates the network topology using the updated attribute data packet of each transmission node, and re-determines the best route for data transmission.

[0012] In an exemplary embodiment of the present disclosure, the step of establishing the software-defined flight ad hoc network model includes: modeling the software-defined flight ad hoc network including n transmission nodes as an undirected graph G=<V,E> , where E represents the link set; V = {C, TN} represents the set of all nodes in the software-defined flight ad hoc network, where C represents the control node; TN = {v1, v2, …, v n} represents the set of transmission nodes, v n Indicates the nth transmission node.

[0013] In an exemplary embodiment of the present disclosure, in the software-defined flight ad hoc network model, all the transmission nodes constitute a homogeneous network, and the initial states of all the transmission nodes are the same; all the transmission nodes fly at the same altitude, and any two transmission nodes are not at the same position; the communication range of all the transmission nodes is a circular area with a radius of R;

[0014] The control node is able to maintain centralized control over all of the transmission nodes.

[0015] In an exemplary embodiment of the present disclosure, the plurality of transmission nodes generating respective attribute data comprises the following steps:

[0016] Within a broadcast interval of the transmission node, all the transmission nodes broadcast the sampling data packets to the outside respectively, and simultaneously receive the sampling data packets from the neighboring transmission nodes respectively;

[0017] All the transmission nodes calculate the metric parameters of the current link quality according to the received sampled data packets, and the metric parameters include the average values ​​of the packet receiving rate PRR, the received signal strength RSSI, and the signal-to-noise ratio SNR of the corresponding channels. and the average value of the link quality indicator LQI

[0018] All the transmission nodes generate their own attribute data packets according to the metric parameters of their own current links and the current node residual energy, and upload the attribute data packets to the control node through the control channel.

[0019] In an exemplary embodiment of the present disclosure, the transmission node v i The formula for the broadcast interval includes:

[0020]

[0021] Among them, e ij represents the link between transmission nodes v i and transmission node v j , where v i ∈v n , v j ∈v n ; NH i represents the set of neighbor transmission nodes of transmission node v i ; T(e ij ) represents the link duration between transmission nodes v i and transmission node v j , and the formula for the link duration includes:

[0022]

[0023] Among them, d ij represents the Euclidean distance between transmission nodes v i and transmission node v j ; and respectively represent the velocity vectors of transmission nodes v i and transmission node v j ;

[0024] Within the circular area with radius R, when the average distance between the two transmission nodes is 0.9054R, the formula for the link duration becomes:

[0025]

[0026] In an exemplary embodiment of the present disclosure, the calculation formula of the packet reception rate PRR includes:

[0027]

[0028] Among them, 0 < m ≤ M, m represents the number of sampled data packets successfully received by the transmission node; M represents the total number of sampled data packets;

[0029] The packet reception rate PRR corresponds to the corresponding channel quality score LQS in the control node;

[0030] The calculation processes of the received signal strength RSSI, signal-to-noise ratio SNR, and link quality indicator LQI include:

[0031] When the transmission node receives the i-th sampled data packet, it measures the received signal power P Ri , and calculates the value SNR Ri of the signal-to-noise ratio SNR and the value RS i of RSSI based on the received signal power P i, where the SNR i is directly obtained by the receiving end of the transmission node. The calculation formula of RSSI includes:

[0032] RS i = 10lg P Ri (5)

[0033] where, lgP Ri represents the logarithm of P Ri to the base 10;

[0034] Record the LQI value of the i-th sampled data packet as l i , and its calculation formula includes:

[0035] l i = 255×(RS i + 81) / 91 (6)

[0036] Within a broadcast interval, the transmission node successfully receives a total of m sampled data packets, 0 < m ≤ M. Then the average value of the signal-to-noise ratio SNR and the average value of the link quality indicator LQI are respectively:

[0037]

[0038] According to the average value of SNR and the average value of the link quality indicator LQI Calculate the Triangle value of the corresponding channel. The calculation formula of the Triangle value includes:

[0039]

[0040] The Triangle value corresponds to the corresponding channel quality score LQS in the control node;

[0041] When the channel quality scores LQS corresponding to the Triangle value and the PRR value are both high, it is considered that the overall quality of the channel is high. When the calculated Triangle value is inconsistent with the link quality score LQS corresponding to the PRR value, take the worse one of the two as the final determination result of the link quality; The formula for link quality estimation includes:

[0042] TM' = min(LQS(Triangle), LQS(R W )) (10)

[0043] The process of calculating the remaining energy of the transmission node includes:

[0044] When the transmission node v iSend data xbit to transmission node v j When the transmission node v i Send data and transfer node v j The energy consumed by receiving data is:

[0045]

[0046] E r (x) = E e x (12)

[0047] Among them, E e Represents the transmission node v i The energy consumed to process 1 bit of data, E f represents the energy loss coefficient; R represents the radius of the communication range of the transmission node; the energy E currently consumed by the transmission node is obtained according to equations (11) and (12): lost , assuming that the initial energy of each transmission node is E init , then the remaining energy of the transmission node is:

[0048] E rest =E init -E lost (13).

[0049] In an exemplary embodiment of the present disclosure, the step of generating a network topology by the control node includes: the source node address and the destination node address of the attribute data packet both use MAC addresses, the message type of the attribute data packet is used to distinguish the category to which the information belongs, and when the message type is a sampling data packet, the destination node address is a broadcast address;

[0050] When the control node receives the location information from the transmission node, it generates a network topology, which is represented by an adjacency matrix as an undirected graph G including:

[0051]

[0052] Among them, e ij Represents the transmission node v i and v j The link between ij ∈e nn , whether the link is reachable, the expression methods include:

[0053]

[0054] Among them, d ij Represents the transmission node v i and v j The Euclidean distance between them is calculated by the following formula:

[0055]

[0056] In an exemplary embodiment of the present disclosure, the process of the control node determining the best route for data transmission includes two steps: route discovery and selection and route establishment, wherein:

[0057] The process of discovering and selecting the route includes: when the transmission node v i To the transmission node v j When sending data, the transmission node v i First, a routing request is sent to the control node through a control channel. After the control node obtains the routing request, it designs an heuristic factor function based on the available resources of the transmission node, designs a corresponding path pheromone concentration function based on the link quality, and calculates the probability of each transmission node selecting its neighbor node;

[0058] The process of establishing the route includes: after the control node determines the best route, the route information is sent to the software-defined flight ad hoc network in a directional manner, so that the relevant transmission nodes can transmit data along the best route.

[0059] In an exemplary embodiment of the present disclosure, the process of discovering and selecting the route includes:

[0060] When the transmission node v i The neighbor transmission node set NH i When there are multiple transmission nodes, first determine the destination node v d Whether it belongs to the neighbor transmission node set NH i , if v d ∈NH i , then directly select v d As the next hop node; otherwise, the path probability is calculated according to the ant colony algorithm to select the next hop node;

[0061] Choose NH i The transmission node v in j The expected degree of the next hop node is determined by the heuristic factor η ij Expressions include:

[0062]

[0063] Among them, ξ1 and ξ2 are weight coefficients, representing and The degree of influence on the inspiration factor, ξ1+ξ2=1; Represents the transmission node v j The current remaining energy, Represents the transmission node v j The current available load factor, which is expressed as:

[0064]

[0065] Among them, L j For the transmission node v j Current MAC layer interface queue length, L max is the maximum load that the transmission node can bear. In the software-defined flight ad hoc network model, all the transmission nodes constitute a homogeneous network, and the initial states of all the transmission nodes are the same, but the units are inconsistent. The heuristic factor is normalized, and the formula (17) is rewritten as:

[0066]

[0067] At the transmission node v j The neighbor node set NH i In the , select the transmission node v j The probability expressions of being the next hop node include:

[0068]

[0069]

[0070] Among them, τ ij For path e ij The pheromone concentration on ij =e TM' -1; α and β are system parameters, which respectively represent the influence of path pheromone and heuristic factor on the selection of next hop node; Formula (21) represents the constraint condition, which means that the probability of each path is between 0 and 1, and the selection probability of all reachable paths is 1;

[0071] The control node uses the depth-first selection DES algorithm to find the source node v s To the destination node v d The collection of all paths Path sd , when the set Path sd When there are multiple available paths in , the formula for calculating the selection probability of each path includes:

[0072]

[0073] Among them, P ij is the selection probability of each link in each path; H is the number of routing hops;

[0074] Select the path with the highest probability as the best route, and the data packet completes data transmission along this route.

[0075] In an exemplary embodiment of the present disclosure, the process of establishing the route includes: the control node sends the route information to the software defined flight ad hoc network in a directional manner, that is, when the best route is p sd = <v s ,v a ,v b ,v c ,v d >, the control node only transmits to the transmission node v through the corresponding control channel s 、v a 、v b 、v c and v d A routing information data packet is sent, the previous hop address of the routing information data packet informs the transmission node of the source of the routing information data packet that needs to be received and processed, and the next hop address indicates to which transmission node the transmission node should forward the routing information data packet; the previous hop address in the routing information received by the source node and the next hop address in the routing information received by the destination node are both empty.

[0076] The technical solution provided by the present disclosure may have the following beneficial effects:

[0077] In the disclosed embodiment, an energy consumption model and a link quality estimation model are established in the data transmission method, which can not only improve energy utilization efficiency, but also avoid link congestion and ensure link transmission stability; when establishing a route, the geographical location information of the transmission node is used as the basis, combined with the current state of the transmission node and the link state information, to select the best route for data transmission. The routing protocol can ensure balanced network energy consumption, while obtaining lower network latency and network overhead, and higher throughput, which can meet the various requirements of the drone network to perform tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0079] Figure 1 A schematic diagram showing steps based on a data transmission method in an exemplary embodiment of the present disclosure is shown;

[0080] Figure 2 A schematic diagram showing a change in network load when a transmission node periodically broadcasts externally in an exemplary embodiment of the present disclosure;

[0081] Figure 3A comparison diagram of network energy consumption balance in a simulation experiment in an exemplary embodiment of the present disclosure is shown;

[0082] Figure 4 A comparison diagram of average network delays in simulation experiments in exemplary embodiments of the present disclosure is shown;

[0083] Figure 5 A comparison diagram showing the average number of network routing hops in a simulation experiment in an exemplary embodiment of the present disclosure;

[0084] Figure 6 A comparison diagram of average network throughputs of simulation experiments in exemplary embodiments of the present disclosure is shown;

[0085] Figure 7 A diagram showing a comparison of network routing costs in a simulation experiment in an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0086] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0087] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0088] In this example implementation, a routing protocol based on energy consumption and link quality in a software-defined flight ad hoc network is first provided. Figure 1 As shown in , the routing protocol may include the following steps:

[0089] Step S101: Establish a software-defined flight ad hoc network model, which includes a control node and multiple transmission nodes;

[0090] Step S102: the plurality of transmission nodes respectively generate their own attribute data packets within a broadcast interval, and respectively upload their own attribute data packets to the control node;

[0091] Step S103: the control node generates a network topology according to all the received attribute data packets;

[0092] Step S104: the control node calculates the selection probability of each transmission node respectively by using the ant colony optimization algorithm according to the network topology, and determines the best route for data transmission according to the selection probability;

[0093] Step S105: the plurality of transmission nodes repeat the above steps at the next broadcast interval, and the control node regenerates the network topology using the updated attribute data packet of each transmission node, and re-determines the best route for data transmission.

[0094] In the disclosed embodiment, an energy consumption model and a link quality estimation model are established in the data transmission method, which can not only improve energy utilization efficiency, but also avoid link congestion and ensure link transmission stability; when establishing a route, the geographical location information of the transmission node is used as the basis, combined with the current state of the transmission node and the link state information, to select the best route for data transmission. The routing protocol can ensure balanced network energy consumption, while obtaining lower network latency and network overhead, and higher throughput, which can basically meet the various requirements of the drone network for executing tasks.

[0095] Below, each step of the above method in this exemplary embodiment will be described in more detail.

[0096] Step S101 includes modeling a software-defined flight ad hoc network including n transmission nodes as an undirected graph G=<V,E> , where E represents the link set; V = {C, TN} represents the set of all nodes in the software-defined flight ad hoc network, where C represents the control node; TN = {v1, v2, …, v n} represents the set of transmission nodes, v n Indicates the nth transmission node.

[0097] The software-defined flight autonomous network includes a control node and several transmission nodes, among which the transmission node can sense and collect surrounding environment information, share relevant information with other transmission nodes, and upload their own attribute data packets (Node Attribute, NA message) to the control node through the control channel. The control node is used to receive the NA message transmitted by each transmission node, process it and give feedback, that is, to complete the functions of network topology generation, update and routing calculation.

[0098] In this example, in the software-defined flight ad hoc network, all transmission nodes fly at the same altitude, and no two transmission nodes are in the same position; the communication range of each transmission node is a circular area with a radius of R; all transmission nodes form a homogeneous network, and the initial state of all transmission nodes is the same; the control node has sufficient capabilities (including computing, storage, and energy, etc.) to maintain centralized control of all transmission nodes.

[0099] Step S102 includes the following steps:

[0100] Step S1021: within a broadcast interval of the transmission node, all the transmission nodes broadcast sample data packets to the outside respectively, and simultaneously receive sample data packets from neighbor transmission nodes respectively;

[0101] See attached Figure 1 and attached Figure 2 As shown in the figure, the transmission node obtains neighbor node information and related link information by periodically broadcasting sampled data packets and receiving sampled data packets from neighbor nodes. However, periodic broadcasting of data packets will cause the network to send many data packets at the same time in a certain period of time, resulting in broadcast storms and network congestion. Figure 2 As shown, after each cycle CT, there will be a detection time T W A large number of data packets appear within T W During other time periods, the network load is relatively small, and a large amount of network resources cannot be used for data transmission. Therefore, periodic external broadcasting not only poses a risk of network congestion, but also results in low utilization of the entire network resources.

[0102] In this example, in order to solve the above problems, a periodic broadcast method is no longer used. Instead, an adaptive broadcast method is proposed, which can obtain the optimal broadcast interval according to the network topology changes and enhance the adaptability of the network to the topology changes. The specific adaptive broadcast method is:

[0103] Transmission Node v i The formula for the broadcast interval includes:

[0104]

[0105] Among them, e ij Represents the transmission node v i and v j The link between i ∈v n , v j ∈v n NH i Represents the transmission node v i The set of neighbor transmission nodes; T(eij ) represents the link duration between the transmission node vi and the transmission node v j . The formula for the link duration includes:

[0106]

[0107] where d ij represents the Euclidean distance between the transmission node v i and the transmission node v j ; and respectively represent the velocity vectors of the transmission node v i and the transmission node v j ;

[0108] Since within a circular area with a radius of R, the average distance between two said transmission nodes is 0.9054R; then the formula for the link duration is:

[0109]

[0110] Step S1022: All said transmission nodes respectively calculate the metric parameters of the current link quality according to the received sampled data packets. The metric parameters include the average values of the packet reception rate PRR, the received signal strength RSSI, and the signal-to-noise ratio SNR of their respective corresponding channels and the average value of the link quality indicator LQI

[0111] For all the above calculation results, each transmission node does not forward after processing, but directly destroys them from its respective memory.

[0112] The packet reception rate (PRR) is a statistic for packet reception and retransmission at the data link layer, which can directly reflect the quality of the network performance and can be used as a metric indicator to measure whether the link transmission is stable.

[0113] In this example, the formula for the packet reception rate PRR includes:

[0114]

[0115] where 0 < m ≤ M, m represents the number of sampled data packets received by the transmission node; M represents the total number of sampled data packets;

[0116] The description of the PRR value for the channel quality can quantitatively give the corresponding channel quality score (Link Quality Score, LQS), as shown in Table 1:

[0117] Table 1: Channel quality scores corresponding to PRR

[0118]

[0119]

[0120] The PRR can only judge the accuracy of the short - term link quality change, and it is less sensitive to the link change. Therefore, in order to more accurately characterize the quality of the link, the concept of link transmission quality is introduced here. Using this link transmission quality and combining with the PRR to jointly estimate the link quality.

[0121] The current metric parameters for measuring the link transmission quality include the received signal strength (Radio Signal Strength Indicator, RSSI), the signal - to - noise ratio (Signal to Noise Ratio, SNR), and the link quality indicator (Link Quality Indicator, LQI), etc. Since these parameters are received and measured by the physical - layer hardware, they are easily affected by external noise, multipath fading, and interference. In order to overcome the problem that a single metric parameter cannot characterize the link transmission quality, the Triangle model is introduced here. The estimation process of this link transmission quality is as follows:

[0122] When the transmitting node receives the i - th sampled data packet, it measures the received signal power P Ri , according to this received signal power P Ri calculate the value of the signal - to - noise ratio SNR, SNR i and the value of RSSI, RS i , where SNR i is directly obtained from the receiving end of the transmitting node, and the calculation formula of RSSI includes:

[0123] RS i = 10lgP Ri (5)

[0124] where, lgP Ri represents the logarithm of P Ri to the base 10;

[0125] Record the value of LQI of the i - th sampled data packet as:

[0126] l i = 255×(RS i + 81) / 91 (6)

[0127] Within a broadcast interval, the transmitting node has successfully received m sampled data packets in total, 0 < m ≤ M. Then the average value of the signal - to - noise ratio SNR and the average value of the link quality indicator LQI are respectively:

[0128]

[0129] The average SNR calculated by and the average value of the link quality indicator LQI Calculate the Triangle value of the corresponding channel. The calculation formula of the Triangle value is:

[0130]

[0131] According to the Triangle value, the channel quality can also be divided according to the threshold and the corresponding LQS is given, as shown in Table 2:

[0132] Table 2: Channel quality scores corresponding to Triangle values

[0133]

[0134] In actual network communications, not only is high channel transmission quality required, but also channel transmission stability is expected to be guaranteed. That is, only when the channel quality scores measured by the Triangle and PRR values ​​are relatively high, the overall quality of the channel is considered to be high. When the calculated channel transmission quality and the channel quality characterized by stability are inconsistent, the worse one of the two is taken as the final judgment result of the channel quality. For example, when the Triangle value is 150 and the PRR value is only 0.3, it can only be said that the channel transmission quality is good, but the stability is very poor, so the channel quality is judged to be poor. Therefore, the expression for link quality estimation is defined as:

[0135] TM'=min(LQS(Triangle),LQS(R W )) (10).

[0136] Step S1023: All the transmission nodes generate their own attribute data packets according to the metric parameters of their current links and the current node residual energy, and upload the attribute data packets to the control node through the control channel.

[0137] In this example, the communication energy consumption of the transmission node is mainly considered, that is, the energy consumed by the wireless module of the transmission node for data transmission. This part of energy is the total energy consumed by the transmission node when sending and receiving data.

[0138] When the transmission node v i Send data xbit to transmission node v j When the transmission node v i Send data and transfer node v j The energy consumed by receiving data is:

[0139]

[0140] E r (x) = E e x (12)

[0141] Among them, E e Represents the transmission node v i The energy consumed by internal circuit components to process 1 bit of data, E f represents the energy loss coefficient; R represents the radius of the communication range of the transmission node; the energy E currently consumed by the transmission node is obtained according to equations (11) and (12): lost , assuming that the initial energy of each transmission node is E init , then the remaining energy of the transmission node is:

[0142] E rest =E init -E lost (13).

[0143] In step S103, the control node generates a network topology according to the NA information of each transmission node in the software-defined flight ad hoc network, which is represented by an adjacency matrix, as shown in Table 3:

[0144] Table 3: NA information format representing the generated network topology

[0145]

[0146] Among them, the source node address and the destination node address of the attribute data packet are both MAC addresses, and the message type of the attribute data packet is used to distinguish the category to which the information belongs. When the message type is an attribute data packet, the destination node address is a broadcast address;

[0147] For convenience, a node refers to a transmission node, a source node refers to a source transmission node, and a destination node refers to a destination transmission node.

[0148] In this example, when the control node receives the location information from the transmission node, a network topology is generated. The network topology is represented by an undirected graph G represented by an adjacency matrix and includes:

[0149]

[0150] Among them, e ij Represents the transmission node v i and v j The link between ij ∈e nn , whether the link is reachable, the expression methods include:

[0151]

[0152] Among them, d ij Represents the transmission node v i and v j Since in this network model, all transmission nodes fly at the same altitude and no two transmission nodes are in the same position, the Euclidean distance between any two transmission nodes is calculated as:

[0153]

[0154] Step S104 includes the following steps:

[0155] Step S1041: Route discovery and selection: When the transmission node v i To the transmission node v j When sending data, the transmission node v i First, a routing request is sent to the control node through the control channel. After the control node obtains the routing request, this example implements the routing discovery mechanism based on the ant colony algorithm. The core part of the ant colony algorithm lies in the path pheromone concentration and the heuristic factor. In this example, the heuristic factor function is designed based on the available resources of the transmission node; according to the relevant theory of the ant colony algorithm, the corresponding path pheromone concentration function is designed based on the link quality to improve the data transmission performance and node resource utilization; and the probability of each transmission node selecting its neighbor node is calculated;

[0156] In this example, when the transmission node v i The neighbor node set NH i When there are multiple nodes, first determine the destination node v d Whether it belongs to the neighbor node set NH i , if v d ∈NH i , then directly select v d As the next hop node; otherwise, the path probability is calculated according to the ant colony algorithm to select the next hop node;

[0157] Choose NH i Node v in j The expected degree of the next hop node is determined by the heuristic factor η ij express:

[0158]

[0159] Among them, ξ1 and ξ2 are weight coefficients, representing and The degree of influence on the inspiration factor, ξ1+ξ2=1; Represents the transmission node v j The current remaining energy, Represents the transmission node vj The current available load rate is expressed as:

[0160]

[0161] Among them, L j For the transmission node v j Current MAC layer interface queue length, L max is the maximum load that the transmission node can bear. In the software-defined flight ad hoc network model, all the transmission nodes constitute a homogeneous network, and the initial states of all the transmission nodes are the same, but the units are inconsistent. Therefore, the heuristic factor is normalized, and the formula (17) is rewritten as:

[0162]

[0163] According to formula (19), the greater the remaining energy and available load rate of the transmission node, the higher the probability that the transmission node is selected as the next hop node. The value of the weight coefficient can be adjusted according to the different requirements of different task types on network service quality.

[0164] At the transmission node v j The neighbor node set NH i In the , select the transmission node v j The probability expressions of being the next hop node include:

[0165]

[0166] Among them, τ ij For path e ij The pheromone concentration on ij =e TM' -1; α and β are system parameters, which respectively represent the influence of path pheromone and heuristic factor on the selection of next hop node; Formula (21) represents the constraint condition, which means that the probability of each path is between 0 and 1, and the selection probability of all reachable paths is 1;

[0167] In this example, the path pheromone is defined according to the link quality estimation model. The better the link quality is, the higher the probability of data forwarding on the path is.

[0168] The control node generates the network topology by receiving the attribute data of each transmission node and calculates the probability of each node selecting its neighbor node. After receiving the routing request of the source node, the destination node address is obtained. In this example, the control node uses the Depth First Search (DES) algorithm to find the source node v s To the destination node v d A collection of all available paths Path sd, when the set Path sd When there are multiple available paths in , the formulas for calculating the selection probability of each path include:

[0169]

[0170] Among them, P ij is the selection probability of each link in each path; H is the number of routing hops.

[0171] Select the path with the highest probability as the best route, and the data packet completes data transmission along this route.

[0172] Step S1042: Route establishment: After the control node determines the best route, it needs to send the selected route information to the software-defined flight ad hoc network to facilitate the relevant transmission nodes to process the data packets accordingly and transmit data along the best route.

[0173] In this example, in order to prevent network congestion and reduce unnecessary control overhead, the control node adopts the method of sending routing information in a directional manner, that is, when the best route is p sd = <v s ,v a ,v b ,v c ,v d >, the control node only sends a signal to the transmission node v through the corresponding control channel. s 、v a 、v b 、v c and v d Send a routing information (RI) data packet, the format of which is shown in Table 4:

[0174] Table 4: RI packet format

[0175]

[0176] Among them, the previous hop address informs the transmission node of the source of the RI data packet that needs to be received and processed, and the next hop address indicates to which transmission node the transmission node should forward the RI data packet; the previous hop address in the RI information received by the source node and the next hop address in the RI information received by the destination node are both empty.

[0177] The routing protocol based on energy consumption and link quality (RELQ) disclosed in the present invention

[0178] The RELQ algorithm disclosed in the present invention is analyzed and compared with the traditional GPSR protocol, OLSR protocol and AODV protocol by simulation experiments, and the simulation software used is: NS-2 (Network Simulator v2.35). In order to reduce the random error of the simulation experiment, all experimental results are the average value of 20 experiments. The simulation parameters are shown in Table 5:

[0179] Table 5: Simulation parameters

[0180]

[0181] First, a simulation experiment of network energy consumption is carried out.

[0182] Traditional routing algorithms do not consider the energy consumption of transmission nodes, which will cause the communication volume of some transmission nodes to be too large, resulting in the complete consumption of energy and inability to continue working, thereby increasing network delay and packet loss rate. The RELQ algorithm disclosed in this paper incorporates the energy consumption of transmission nodes into the routing selection algorithm and adopts the energy balance factor F EB Measures the balance of network communication energy consumption. EB The expression is as follows:

[0183]

[0184] Where n is the number of transmission nodes, For the transmission node v i The remaining energy, is the average remaining energy of all transmission nodes.

[0185] See attached Figure 3 Shown: Attached Figure 3 The energy balance factors of different algorithms under the same simulation environment are given in . It can be seen that no matter how the speed of the transmission node changes, the energy balance factor under the RELQ protocol is in a relatively stable state, while under the other three protocols, the energy balance factor decreases with the increase of speed. This is because the energy consumption of the transmission node is established in the RELQ protocol, so the speed change has little effect on it. As the speed of the transmission node increases, the network topology of other algorithms changes frequently, resulting in frequent changes in routing. Therefore, more nodes participate in data transmission, and the network energy consumption is more balanced. Frequent changes in routing will increase the control overhead, thereby consuming more energy. Therefore, the RELQ protocol has a better performance in energy saving.

[0186] Secondly, conduct a simulation experiment on network delay

[0187] Network latency refers to the end-to-end latency of data transmission in the network. This indicator reflects the timeliness of data packet transmission and is a very important evaluation indicator for latency-sensitive tasks. The latency referred to in this disclosure refers to the average latency of all data packet transmissions within the simulation time.

[0188] See attached Figure 4 As shown, attached Figure 4 This is a comparison chart of network delays under different routing protocols. Figure 4 In the experiment, the network delay under the RELQ protocol is lower than that of the AODV protocol and the GPSR protocol, but higher than that of the OLSR protocol. In addition, the network delay fluctuates slightly with the speed of the transmission node, or even remains basically unchanged, and remains at a relatively stable level. This shows that the RELQ protocol can not only ensure low network delay, but also adapt to highly dynamic networks.

[0189] See attached Figure 5 As shown, attached Figure 5 is the average number of hops for routing under different routing protocols. Figure 5 It can be seen that the average hop count of the RELQ protocol is the highest, which is due to the balancing mechanism in the RELQ protocol. The protocol does not optimize for a certain indicator, but selects the best route based on the overall state of the network. Therefore, over time, the optimal route will no longer be the shortest route, but the route with the best overall performance, so the average routing hop count of the RELQ protocol is higher than that of the other three routing protocols.

[0190] Increasing the number of routing hops will theoretically increase network delay, but Figure 4 It can be seen that the network delay under the RELQ protocol has no obvious increasing trend. This is because the increase in the number of routing hops only increases the network propagation delay, while the RELQ protocol takes into account the link status and the current load status of the transmission node when calculating the route, thus reducing the transmission delay of the link, the queuing delay of the transmission node, and the processing delay. In addition, as the speed of the transmission node increases, the routing stability under the other three protocols is relatively poor, which will also lead to an increase in network delay.

[0191] Therefore, the network delay performance under the RELQ protocol is relatively good.

[0192] Again, conduct a simulation experiment on network throughput

[0193] Throughput refers to the amount of data processed by the network per unit time. Throughput is an important indicator for the network. The higher the average network throughput, the more data can be transmitted and processed, and the better the network performance.

[0194] See attached Figure 6 As shown, attached Figure 6 is a comparison of the average network throughput under different routing protocols. Figure 6 It can be seen that the average throughput under the RELQ protocol is greatly improved compared with other comparison protocols. Under the same environment and conditions, it can transmit and process more data, which can greatly improve network performance and network efficiency.

[0195] In the RELQ protocol, the calculation and selection of routes are completed by the control node. When calculating the route, the control node considers both the link status and the current status of the transmission node, selects the best route after comprehensive evaluation, and adaptively adjusts the broadcast cycle according to the network status. Therefore, the route selected by this protocol is relatively more stable and is not easily interrupted during data transmission. Therefore, compared with the other three protocols, the RELQ protocol can complete data transmission on a more stable route, thereby obtaining a higher average throughput.

[0196] Finally, perform network routing overhead analysis

[0197] Routing overhead refers to the number of control packets that need to be sent to establish a route during the operation of the network. The larger the routing overhead, the more frequent the route changes and the more unstable the route; more resources will be used to transmit control packets. For drone networks with limited resources such as bandwidth and energy, unnecessary energy consumption will be generated, thereby reducing network efficiency. For different routing protocols, the format of the data packets sent to establish a route is different, and the size of each data packet is also different. In order to describe the routing overhead more objectively, the present disclosure does not define the routing overhead based on the number of control packets, but defines it based on the data volume of the control packets.

[0198] See attached Figure 7 As shown, attached Figure 7 The following is a comparison of routing costs under different routing protocols. Figure 7 It can be seen that the routing overhead of the RELQ protocol is significantly reduced compared to the OLSR protocol and the AODV protocol, but the overhead is increased compared to the GPSR protocol. Both the OLSR protocol and the AODV protocol need to maintain the routing table of the transmission node itself, which requires more control packets to ensure the real-time nature of the routing table, and both adopt a flooding mechanism in the route discovery process, so the routing overhead is relatively large; while the RELQ protocol and the GPSR protocol are based on geographic location information selection reasons, their transmission nodes themselves do not need to maintain routing tables, and only need to know the geographic location information of the corresponding node when selecting a route, and the number of control packets required is less; the RELQ protocol adds the communication process with the control node, and the number of control packets increases compared to the GPSR protocol, so the routing overhead of the RELQ protocol is higher than that of the GPSR protocol.

[0199] In summary, the RELQ protocol has good performance in terms of network energy consumption, average network delay, average throughput and network overhead. Although the protocol does not perform best in all aspects, considering all indicators, the protocol has excellent overall performance and can provide reliable communication guarantees for software-defined drone network execution tasks in actual use environments.

[0200] It should be noted that, although the steps of the method in the present disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc. In addition, it is also easy to understand that these steps may be, for example, executed synchronously or asynchronously in multiple modules / processes / threads.

[0201] In view of the problems of high end-to-end delay, large routing overhead and low throughput in the routing protocol of the traditional flying ad hoc network architecture, the present invention proposes a performance-balanced routing protocol based on ant colony optimization for software-defined flying ad hoc networks. The protocol establishes an energy consumption model and a link quality estimation model, which can not only improve energy utilization efficiency, but also avoid link congestion and ensure link transmission stability; when establishing a route, the geographical location of the transmission node is used as the basis, and the information of the current state of the transmission node and the link state is combined to select the best route for data transmission. By comparing the simulation performance of the RELQ protocol with the GPSR protocol, the OLSR protocol and the AODV protocol, the results show that the RELQ protocol can ensure balanced network energy consumption, while obtaining lower network delay and network overhead, and higher throughput, which can basically meet the various requirements of the UAV network to perform tasks.

[0202] It should be noted that although several units of the system for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiment of the present disclosure, the features and functions of two or more units described above can be concretized in one unit. Conversely, the features and functions of a unit described above can be further divided into multiple units for concretization. Some or all of the units can be selected according to actual needs to achieve the purpose of the disclosed solution. Those of ordinary skill in the art can understand and implement it without paying creative work.

[0203] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

Claims

1. A routing selection method based on energy consumption and link quality, characterized in that: Including the following steps: Establish a software-defined flying ad hoc network model, which includes a control node and multiple transmission nodes; Within a broadcast interval, multiple said transmission nodes respectively generate their own attribute data packets and upload their respective attribute data packets to the control node; The control node generates a network topology based on all the received attribute data packets; Based on the network topology, the control node uses the ant colony optimization algorithm to calculate the selection probability of each transmission node respectively, and determines the best route for data transmission through this selection probability; In the next broadcast interval, multiple said transmission nodes repeat the above steps, and the control node regenerates the network topology using the updated attribute data packets of each transmission node and re-determines the best route for data transmission; Among them, the control node uses the ant colony optimization algorithm to calculate the selection probability of each transmission node respectively based on the network topology, and determines the best route for data transmission through this selection probability, including: Route discovery and selection: When the transmission node v i Transmit node v j When sending data, the transmission node v i First, a routing request is sent to the control node through a control channel. After the control node obtains the routing request, a routing discovery mechanism is implemented based on an ant colony algorithm. The core part of the ant colony algorithm includes a path pheromone concentration and an inspiration factor. The inspiration factor function is designed based on the available resources of the transmission node. According to the relevant theory of the ant colony algorithm, a corresponding path pheromone concentration function is designed based on the link quality, and the probability of each transmission node selecting its neighbor node is calculated. When the transmission node v i The neighbor node set NH i When there are multiple nodes, first determine the destination node v d Whether it belongs to the neighbor node set NH i , if v d ∈NH i , then directly select the destination node v d As the next hop node; otherwise, the path probability is calculated according to the ant colony algorithm to select the next hop node; Choose NH i The transmission node v in j The expected degree of the next hop node is determined by the heuristic factor η ij express: Among them, ξ1 and ξ2 are weight coefficients, representing and The degree of influence on the inspiration factor, ξ1+ξ2=1; Represents the transmission node v j The current remaining energy, Represents the transmission node v j The current available load factor; At the transmission node v j The neighbor node set NH i In the , select the transmission node v j The probability expressions of being the next hop node include: Among them, τ ij For path e ij The pheromone concentration on ij =e TM′ -1, e TM′ where e represents the natural base, TM′ represents the link quality estimation expression; α and β are system parameters, which respectively represent the influence of path pheromone and heuristic factor on the selection of next hop node; Indicates path e ik The pheromone concentration on Indicates path e ij The inspiration factor on Indicates path e ik The heuristic factor on,Formula (21) represents the constraint condition, which means that the probability of each path is between 0 and 1, and the selection probability of all reachable paths is 1; The control node generates a network topology by receiving the attribute data of each transmission node and calculates the probability of each node selecting its neighbor node; Upon receiving the source node v s After the routing request, the destination node v is obtained d The address of the source node v s To the destination node v d A collection of all available paths Path sd , when the set Path sd When there are multiple available paths in , the selection probability of each path is calculated separately: Among them, P ij is the selection probability of each link in each path; H is the number of routing hops; the path with the highest selection probability is the best route; Route establishment: After the control node determines the best route, it needs to send the selected route information to the software-defined flying ad hoc network, so that the relevant transmission nodes can process the data packets, enabling the data packets to be transmitted along the best route.

2. The routing selection method according to claim 1, characterized in that: The steps of establishing the software-defined flight ad hoc network model include: modeling the software-defined flight ad hoc network including n transmission nodes as an undirected graph G=<V,E> , where E represents the link set; V = {C, TN} represents the set of all nodes in the software-defined flight ad hoc network, where C represents the control node; TN = {v1, v2, …, v n } represents the set of transmission nodes, v n Indicates the nth transmission node.

3. The routing selection method according to claim 2, characterized in that: In the software-defined flying ad hoc network model, All said transmission nodes form a homogeneous network, and all said transmission nodes have the same initial state; all said transmission nodes fly at the same altitude, and no two said transmission nodes are at the same position; the communication range of all said transmission nodes is a circular area with a radius of R; The control node can maintain centralized control over all said transmission nodes.

4. The routing selection method according to claim 3, characterized in that: The steps for multiple said transmission nodes to generate their own attribute data packets include: Within a broadcast interval of the transmission node, all said transmission nodes respectively broadcast sampling data packets and simultaneously receive sampling data packets from neighbor transmission nodes; All the transmission nodes calculate the metric parameters of the current link quality according to the received sampled data packets, and the metric parameters include the average values ​​of the packet receiving rate PRR, the received signal strength RSSI, and the signal-to-noise ratio SNR of the corresponding channels. and the average value of the link quality indicator LQI All said transmission nodes respectively generate their own attribute data packets based on the metric parameters of their respective current links and the remaining energy of the current node, and upload the attribute data packets to the control node through the control channel respectively.

5. The routing selection method according to claim 4, characterized in that: Transmission Node v i The formula for the broadcast interval includes: Among them, e ij Represents the transmission node v i and the transport node v j The link between i ∈v n , v j ∈v n NH i Represents the transmission node v i The set of neighbor transmission nodes; T(e ij ) represents the transmission node v i and the transport node v j The link duration between the two ends is given by: Among them, d ij Represents the transmission node v i and the transport node v j The Euclidean distance between and Represent the transmission node v i and the transport node v j The velocity vector of Within the circular area with a radius of R, when the average distance between two said transmission nodes is 0.9054R, the formula for the link duration becomes:

6. The routing selection method according to claim 5, characterized in that: The calculation formula of the packet reception rate PRR includes: Among them, 0 < m ≤ M, m represents the number of sampling data packets successfully received by this transmission node; M represents the total number of sampling data packets; The packet reception rate PRR corresponds to the corresponding channel quality score LQS in the control node; The calculation processes of the received signal strength RSSI, signal-to-noise ratio SNR, and link quality indicator LQI include: When the transmission node receives the i-th sampling data packet, it measures the received signal power P Ri , according to the received signal power P Ri Calculate the value of signal-to-noise ratio SNR i and RSSI value RS i , where SNR i The RSSI is directly obtained by the receiving end of the transmission node. The calculation formula includes: RS i =10lgP Ri (5) Among them, lgP Ri Indicates P Ri Logarithm to base 10; The LQI value of the i-th sample data packet is recorded as l i , and its calculation formula includes: l i =255×(RS i +81) / 91 (6) During a broadcast interval, the transmission node successfully receives a total of m sampled data packets, where 0 < m ≤ M. Then, the average value of the signal-to-noise ratio (SNR) and the average value of the link quality indicator (LQI) are respectively: According to the average SNR and the average value of the link quality indicator LQI Calculate the Triangle value of the corresponding channel, the calculation formula of the Triangle value includes: The Triangle value corresponds to the corresponding channel quality score LQS in the control node; When the channel quality scores LQS corresponding to the Triangle value and the PRR value are both high, the overall quality of the channel is considered to be high. When the calculated Triangle value and the link quality score LQS corresponding to the PRR value are inconsistent, the worse one between the two is taken as the final determination result of the link quality. The formula for link quality estimation includes: TM'=min(LQS(Triangle),LQS(R W )) (10) The process of calculating the remaining energy of the transmission node includes: When the transmission node v i Send data xbit to transmission node v j When the transmission node v i Send data and transfer node v j The energy consumed by receiving data is: AND r (x)=And e x (12) Among them, E e Represents the transmission node v i The energy consumed to process 1 bit of data, E f represents the energy loss coefficient; R represents the radius of the communication range of the transmission node; the energy E currently consumed by the transmission node is obtained according to equations (11) and (12): lost , assuming that the initial energy of each transmission node is E init , then the remaining energy of the transmission node is: AND rest =And init -AND lost (13)。 7. The routing selection method according to claim 6, characterized in that: The step of the control node generating the network topology includes: the source node address and the destination node address of the attribute data packet are both MAC addresses, the message type of the attribute data packet is used to distinguish the category to which the information belongs, and when the message type is a sampling data packet, the destination node address is a broadcast address; When the control node receives the location information from the transmission node, it generates a network topology, which is represented by an adjacency matrix as an undirected graph G including: Among them, e ij Represents the transmission node v i and v j The link between ij ∈E, E represents the link set. Whether the link is reachable can be expressed by: Among them, d ij Represents the transmission node v i and v j The Euclidean distance between them is calculated by the following formula:

8. The routing selection method according to claim 7, characterized in that: The process of the control node determining the best route for data transmission includes two steps: route discovery and selection and route establishment, wherein: The process of discovering and selecting the route includes: when the transmission node v i Transmit node v j When sending data, the transmission node v i First, a routing request is sent to the control node through a control channel. After the control node obtains the routing request, it designs an heuristic factor function based on the available resources of the transmission node, designs a corresponding path pheromone concentration function based on the link quality, and calculates the probability of each transmission node selecting its neighbor node; The process of establishing the route includes: after the control node determines the best route, the route information is sent to the software-defined flight ad hoc network in a directional manner, so that the relevant transmission nodes can transmit data along the best route.

9. The routing selection method according to claim 8, characterized in that: The process of discovering and selecting the route includes: When the transmission node v i The neighbor transmission node set NH i When there are multiple transmission nodes, first determine the destination transmission node v d Whether it belongs to the neighbor transmission node set NH i , if v d ∈NH i , then directly select v d As the next hop node; otherwise, the path probability is calculated according to the ant colony algorithm to select the next hop node; Choose NH i The transmission node v in j The expected degree of the next hop node is determined by the heuristic factor η ij Expressions include: Among them, ξ1 and ξ2 are weight coefficients, representing and The degree of influence on the inspiration factor, ξ1+ξ2=1; Represents the transmission node v j The current remaining energy, Represents the transmission node v j The current available load factor, which is expressed as: Among them, L j For the transmission node v j Current MAC layer interface queue length, L max is the maximum load that the transmission node can bear. In the software-defined flight ad hoc network model, all the transmission nodes constitute a homogeneous network, and the initial states of all the transmission nodes are the same, but the units are inconsistent. The heuristic factor is normalized, and the formula (17) is rewritten as: At the transmission node v j The neighbor node set NH i In the , select the transmission node v j The probability expressions of being the next hop node include: Among them, τ ij For path e ij The pheromone concentration on ij =e TM' -1; α and β are system parameters, which respectively represent the influence of path pheromone and heuristic factor on the selection of next hop node; Formula (21) represents the constraint condition, which means that the probability of each path is between 0 and 1, and the selection probability of all reachable paths is 1; The control node uses the depth-first selection DES algorithm to find the source transmission node v s To the destination transmission node v d The collection of all paths Path sd , when the set Path sd When there are multiple available paths in , the formula for calculating the selection probability of each path includes: Among them, P ij is the selection probability of each link in each path; H is the number of routing hops; Select the path with the highest probability as the best route, and the data packet completes data transmission along this route.

10. The routing selection method according to claim 9, characterized in that: The process of establishing the route includes: the control node sends the route information to the software defined flight ad hoc network in a directional manner, that is, when the best route is p sd = <v s ,v a ,v b ,v c ,v d >, the control node only transmits to the transmission node v through the corresponding control channel s 、v a 、v b 、v c and v d A routing information data packet is sent, the previous hop address of the routing information data packet informs the transmission node of the source of the routing information data packet that needs to be received and processed, and the next hop address indicates to which transmission node the transmission node should forward the routing information data packet; the previous hop address in the routing information received by the source node and the next hop address in the routing information received by the destination node are both empty.

Images