Airborne network layered directional neighbor discovery method based on communication perception integration

Abstract

The invention provides an airborne network layered directional neighbor discovery method based on communication perception integration. The method comprises the following steps: node time synchronization; initializing a node neighbor list; in the initial sensing stage, nodes send ISAC signals through narrow beams in sequence, and potential neighbor distribution information is obtained based on echoes; in the adaptive handshake stage, the node adjusts the beam selection probability and the ACK probability based on the sensing information; the node decides transceiving based on a preset probability; a sending node firstly selects a beam to send a Hello message based on a wide beam probability, and selects a beam based on a narrow beam probability only when all near neighbor nodes are found; and the receiving node selects beam interception based on the wide beam probability, returns an ACK message based on the ACK probability, and does not return an ACK message frame to the discovered neighbor. The method can reduce the message collision probability, accelerates the neighbor discovery speed, and is suitable for the technical field of military aviation communication.

Description

Technical Field

[0001] The present invention belongs to the technical field of military aviation communications, and in particular relates to an airborne network hierarchical directional neighbor discovery method based on communication perception integration. Background Art

[0002] Airborne networks connect various air combat units through wireless communication, realize efficient information interaction and situation sharing between platforms, and enable air information-based collaborative operations. Driven by the trend of highly informationized operations, the degree of informationization of future aviation battlefields will continue to expand and deepen. First, with the development of new combat modes and air combat units, battlefield information is developing towards the comprehensive development of multiple services such as voice, data, images, and video information, which puts higher requirements on the transmission bandwidth of airborne networks; secondly, the battlefield electromagnetic environment is becoming increasingly complex, and the future aviation combat information confrontation intensity is high, requiring the wireless transmission of airborne networks to have good low interception / low detection performance. It is foreseeable that traditional omnidirectional communication methods can no longer meet these requirements. Therefore, equipping air combat units with directional antennas to realize the "directional transmission-directional reception" of data between nodes has become one of the development trends of future airborne networks.

[0003] Neighbor discovery is a key step before nodes in wireless communication networks are networked and communicated. Its timeliness affects subsequent networking, routing and communication delay. Due to the particularity of airborne networks, airborne networks have higher requirements for data transmission delay compared to other wireless networks. Therefore, it is crucial to ensure the rapid convergence of neighbor discovery in airborne networks. However, when directional antennas are used, rapid neighbor discovery in airborne networks faces many challenges: first, the information is limited before nodes establish communication links. Directional neighbor discovery is actually a blind intersection problem of directional narrow beams, and the efficiency of discovering neighbors is low; second, tactical airborne networks are driven by tasks, and the distribution of nodes in three-dimensional space is often uneven, that is, for some nodes, their neighbors may be densely distributed in some directions, while there are no neighbor nodes in other directions. In order to discover all neighbors, using directional narrow beams to cover all directions is inefficient and takes a long time; third, when neighbors are densely distributed in a certain beam, the collision problem of neighbor discovery handshake messages cannot be ignored. If it is not solved, it will reduce the convergence speed of neighbor discovery.

[0004] Due to the importance of directional neighbor discovery, many scholars have conducted research on this issue in other wireless ad hoc network scenarios. Recently, in order to solve the directional neighbor discovery problem in similar scenarios with limited prior information, some researchers have considered introducing Integrated Sensing and Communication (ISAC) technology to provide auxiliary information for neighbor discovery with the help of the echo of ISAC signals. However, the existing solutions cannot adapt well to the requirements of airborne networks, especially the handshake message collision problem in three-dimensional scenarios with uneven node distribution. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, the present invention provides an airborne network hierarchical directional neighbor discovery method based on integrated communication perception. It is assumed that the total number of nodes in the airborne network is N+1, and each node is identified by a unique ID; each node is equipped with a phased array antenna and has the ability to generate narrow beams and wide beams; the number of narrow beams and wide beams required for the node to traverse and scan 360 degrees is recorded as K respectively. N and K W ; The nodes number the beam directions according to a uniform rule;

[0006] Each node is within the communication range of other nodes; the near-domain range represents the communication range of a node when it uses a wide beam, and the far-domain range represents the communication range of a node when it uses a narrow beam; the neighbor nodes within the near-domain range of a node are called the near-neighbor nodes of the node, and the neighbor nodes within the far-domain range of a node are called the far-neighbor nodes of the node; the nodes in the airborne network have the ability to send ISAC signals and receive and process echo signals. The nodes obtain the number of potential neighbors and the rough relative distance in the current beam based on the received echo signals;

[0007] The method specifically comprises the following steps:

[0008] S1. All nodes synchronize their time through GPS and obtain the current time and their own location information;

[0009] S2. The neighbor discovery process of the node includes multiple time slots, each of which is divided into two sub-time slots; wherein, the working state of the node in each sub-time slot includes a sending state, a receiving state, and a silent state; in the initial perception phase of neighbor discovery, the working state of the node at the activation time is a sending state, and the working state at the non-activation time is a silent state; in the adaptive handshake phase of neighbor discovery, at the beginning of each time slot, the probability of the node entering the sending state is δ, and the probability of entering the receiving state is 1-δ; the node has the ability to quickly switch the working state in each sub-time slot. If the node sends a Hello message frame in the first sub-time slot, it switches to a receiving state in the second sub-time slot and listens for an ACK message frame on the corresponding beam; if the node is in a receiving state in the first sub-time slot, if the Hello message frame is successfully received, it switches to a sending state in the second sub-time slot and sends an ACK message frame on the corresponding beam; if the node is in a receiving state in the first sub-time slot but does not receive a Hello message frame, it switches to a silent state in the second sub-time slot;

[0010] S3, initialize the neighbor list, distance table and potential neighbor list of each node; the neighbor list records the neighbor information discovered through handshake, including neighbor node ID and location information; the distance table records the perceived rough relative distance with the neighbor node, and the rough relative distance is defined as the estimated neighbor node distance obtained after processing the echo signal;

[0011] For each node i, the potential neighbor list includes a potential distant neighbor sub-table and a potential near neighbor sub-table, which respectively record the number of potential distant neighbors sensed on each narrow beam, denoted as And the number of potential neighbors sensed on each wide beam, denoted as Among them, potential close neighbors are defined as neighbor nodes within the perceived wide beam transmission range, and potential distant neighbors are defined as neighbor nodes within the perceived narrow beam transmission range, excluding close neighbor nodes that have been counted as potential close neighbors, which can be obtained by judging the rough distance to the neighbor nodes; in addition, by converting the solid angles of the corresponding directions of the wide beam and the narrow beam, the total number of potential neighbors in each wide beam k direction can be obtained, which is recorded as Define the total number of potential near neighbors and the total number of potential far neighbors as N n and N r ,have and N is the total number of neighbors;

[0012] Definition i,k is the number of discovered neighbor nodes in the kth wide beam direction of node i, b i,l is the number of discovered distant neighbor nodes in the lth narrow beam direction of node i, n i,k is the total number of discovered distant neighbor nodes and near neighbor nodes in the k-th wide beam direction of node i; initialization, let ai,k =0, b i,l =0,n i,k =0;

[0013] S4. In the initial perception phase of neighbor discovery, each node determines its own activation time slot based on the hash function. The specific expression is: Where ID active represents the active node ID, T represents the current time slot number, K N Indicates the number of narrow beams; the narrow beam number used by the activated node in the current time slot number T is l = (Tmod(K N N))mod(K N );

[0014] After determining the activation time, node i uses a determined narrow beam to send an ISAC signal in the first sub-time slot of its activation time, receives an echo signal in the second sub-time slot, and repeats this step by switching beams in the time slot until all directions are traversed; in the second sub-time slot, the activated node i obtains the distribution and relative distance of neighbor nodes in each beam based on the echo signal, and records them in the potential neighbor list and distance table; for inactive nodes, a narrow beam is randomly selected for listening at the inactive time. If a Hello message frame is successfully received in the first sub-time slot of the current time slot, an ACK message frame is sent back in the second sub-time slot; otherwise, the node remains silent in the second sub-time slot;

[0015] S5. If the current time slot exceeds the preset number of time slots in the initial perception phase, the node exits the neighbor discovery initial perception phase and enters the neighbor discovery adaptive handshake phase; here, the preset number of time slots in the initial perception phase is set to T ini =2(N+1)K N , ensuring that each node successfully traverses K N A narrow beam;

[0016] S6. In the neighbor discovery adaptive handshake phase, each node initializes the probability of selecting each beam for transmission and reception based on the neighbor list, potential neighbor list and distance table. The node determines the beam distribution of the potential neighbors to be discovered based on the information obtained in the initial perception phase, and calculates the beam selection probability. The calculation basis is that the beam with a large number of neighbors is selected with a high probability, and the beam without neighbors will not be selected. When node i is in the sending state, the probability of selecting the kth wide beam from the wide beam set for transmission is recorded as p i,k , the probability of selecting the lth narrow beam from the narrow beam set is recorded as q i,k , respectively assign initial values: When node i is in the receiving state, the probability of selecting the kth wide beam to receive from the wide beam set is recorded as r i,k , assign initial value

[0017] In order to reduce the collision when the receiving node sends back the ACK message frame, the receiving node is designed to send back the ACK message frame in a probabilistic manner; the probability of the receiving node in the kth wide beam direction of the sending node i sending back the ACK message frame is denoted as α i,k , assign initial value: The probability that the receiving node in the lth narrow beam direction of the sending node i sends back an ACK is denoted as α i ' ,l , assign initial value: Among them, ε is the preset constant parameter of the system, and ε∈(0,1] is taken;

[0018] S7. In the neighbor discovery adaptive handshake phase, at the beginning of each time slot, each node i determines its own transmit and receive state based on the preset probability δ; if node i is in the transmitting state in the current time slot, it first determines whether all potential neighbors have been added to the neighbor list; here, if node i still has potential neighbors that have not been added to the neighbor list, it sends a wide beam probability set {p i,k ,k∈[1,K W ]}Randomly select a wide beam, send a Hello message frame with the selected wide beam in the first sub-slot of the current time slot, and then switch to the receiving state in the second sub-slot; otherwise, based on the probability set {q i,l ,l∈[1,K N ]} randomly selects a narrow beam l, sends a Hello message frame using the selected narrow beam l in the first sub-slot of the current time slot, and then switches to the receiving state in the second sub-slot; in addition, if node i is in the receiving state in the current time slot, then based on the receiving wide beam probability set {r i,k ,k∈[1,K W ]}Randomly select a wide beam and listen to the Hello message frame in the direction of the selected wide beam in the first sub-time slot of the current time slot;

[0019] If the two nodes are aligned in the beam direction selected in the current time slot, and one is in the sending state and the other is in the receiving state, and the receiving node is within the transmission range of the beam selected by the sending node, and there is no collision of message frames, then the receiving node can successfully receive the Hello message frame sent by the sending node in the first sub-time slot and send an ACK message frame back to the sending node in the second sub-time slot; otherwise, the receiving node remains silent in the second sub-time slot until the end of the time slot;

[0020] S8. In the neighbor discovery adaptive handshake phase, if the sending node i successfully receives the ACK message frame in the second sub-time slot of the current time slot, it is considered that the neighbor node is successfully discovered; if the neighbor node discovered this time is a close neighbor node, let a i,k =a i,k +1,ni,k =n i,k +1; if it is a distant neighbor node, let b i,l =b i,l +1,n i,k′ =n i,k′ +1, k′ indicates the corresponding number of the wide beam direction including the narrow beam l direction;

[0021] Node i extracts the ID information and location information in the ACK message frame, updates its neighbor list, and updates the beam selection probability p i,k ,q i,l , r i,k , and ACK probability α i,k , α i ' ,l ; Among them, the probability update rules are:

[0022] If the receiving node j successfully receives the Hello message frame in the first sub-time slot of the current time slot, it extracts the ID information and location information, and first determines whether the neighbor node has been added to the neighbor list; if the current neighbor node has not been added to the neighbor list, it updates its own neighbor list; if it is a close neighbor node, it sets a j,k =a j,k +1,n j,k =n j,k +1, if it is a distant neighbor node, then let b j,l =b j,l +1,n j,k′ =n j,k′ +1; then, node j updates the beam selection probability and ACK probability, and the update rules are the same as above; if the current neighbor has been added to the neighbor list, node j will not send back the ACK message frame and switch to the silent state in the second sub-time slot;

[0023] S9. If the node's neighbor list records information of all N neighbors, the node's neighbor discovery ends, the node enters silence, and no longer participates in handshake interactions with other nodes; otherwise, the node enters the next time slot and uses the next time slot as the current time slot, and returns to step S7;

[0024] S10. If neighbor discovery of all nodes is completed, neighbor discovery of the entire network is completed; if the total number of time slots for neighbor discovery exceeds the number of time slots preset by the network, neighbor discovery of the entire network is completed.

[0025] In one embodiment of the present invention, a Hello message frame structure is defined, including:

[0026] "Type", message frame type, used to indicate that the message frame is a Hello message frame. 0 indicates a Hello message frame, and 1 indicates an ACK message frame;

[0027] "Node_ID", used to indicate the ID of the sending node;

[0028] "NL" is used to indicate the neighbor list of the sending node, which records the ID and location information of the neighbor nodes that the current sending node has discovered;

[0029] "Beam_Type" is used to indicate whether the beam type used by the transmitting node is a wide beam or a narrow beam. 0 indicates a wide beam and 1 indicates a narrow beam.

[0030] "Beam_ID", used to indicate the beam number used by the sending node;

[0031] "Location" is used to indicate the current location of the sending node;

[0032] "ACK_Pro" is used to indicate the probability of the receiving node sending back an ACK message frame.

[0033] The method of the present invention includes two stages: an initial perception stage and an adaptive handshake stage. In the initial perception stage, each node traverses a narrow beam to send an ISAC signal to achieve preliminary perception of neighbors in the network and obtain the distribution and coarse distance information of all potential neighbors; then, the node divides the potential neighbors into close-distance nodes and distant neighbor nodes according to the obtained coarse distance information, and communicates with them using wide beams and narrow beams respectively in the subsequent adaptive handshake stage. In the adaptive handshake stage, the node adjusts the probability of selecting wide and narrow beams based on the potential neighbor distribution information obtained in the initial perception stage, realizes rapid alignment of beams between nodes in the network, and thus speeds up neighbor discovery.

[0034] The advantages of the present invention are:

[0035] 1. The present invention considers the problem of neighbor discovery handshake message collision in densely distributed node areas in airborne networks, and designs a hierarchical neighbor discovery scheme, which uses wide beams and narrow beams to shake hands with neighbor nodes in the near domain and far domain, respectively, reducing the probability of nodes colliding with nodes in the same direction;

[0036] 2. Aiming at the problem of limited prior information before initial networking and link establishment in airborne networks, the present invention introduces ISAC technology to design a neighbor discovery algorithm. By sending ISAC signals in the initialization perception phase of neighbor discovery, the distribution of potential neighbor nodes is preliminarily perceived and the rough distance is estimated, providing information assistance for subsequent handshakes with neighbor nodes, effectively improving the efficiency of neighbor discovery;

[0037] 3. The present invention is based on the potential neighbor distribution and estimated coarse distance information obtained in the initialization perception stage, and combines the neighbor discovery progress on each beam to adjust the probability of selecting each beam, so that beams with dense potential neighbor distribution are selected with a higher probability, thereby achieving rapid alignment of beams between nodes; in particular, beams without potential neighbor nodes will not be selected, which can further reduce the delay of neighbor discovery. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flowchart of the actual application of the neighbor discovery mechanism described in the present invention;

[0039] Figure 2 This is a schematic diagram of an implementation scenario of the present invention;

[0040] Figure 3 This is a schematic diagram of the time slot structure in the initialization phase of the neighbor discovery mechanism of the present invention;

[0041] Figure 4 This is a message frame structure diagram used in the present invention. DETAILED DESCRIPTION

[0042] The present invention will be described in detail below with reference to the accompanying drawings.

[0043] The present invention provides an ISAC technology-assisted hierarchical neighbor discovery method for airborne networks, such as Figure 1 As shown. The neighbor discovery process consists of two stages: the initial perception stage and the adaptive handshake stage. In the initial perception stage, the node sends the ISAC signal to achieve preliminary perception of the neighbors in the network, providing prior information for the subsequent neighbor discovery handshake stage; in addition, a wide and narrow beam multi-layer beam scanning algorithm is designed to avoid message frame collisions between neighbor nodes at different distances in the same direction. Among them, the message frame collision in the neighbor discovery process includes two situations: In the first situation, the receiving node receives two or more Hello message frames from different sending nodes in the first sub-time slot; in the second situation, the sending node receives two or more ACK message frames from different receiving nodes in the second sub-time slot. In the adaptive handshake stage, the node adjusts the probability of selecting wide and narrow beams based on the prior information obtained in the initial perception stage, reducing the collision of handshake message frames while achieving rapid convergence of neighbor discovery.

[0044] The total number of nodes in the airborne network is N+1, and each node is identified by a unique ID. Each node is equipped with a phased array antenna and has the ability to generate narrow beams and wide beams; the number of narrow beams and wide beams required for the node to traverse and scan 360° is recorded as K N and K W ; The nodes number the beam directions according to a unified rule.

[0045] Figure 2Draw an example of the communication range of a node in an airborne network, where each node is within the communication range of other nodes; Figure 2 The near-domain range in the communication range represents the communication range of a node when a wide beam is used, and the far-domain range represents the communication range of a node when a narrow beam is used; in particular, the neighbor nodes within the near-domain range of a node are called the near-neighbor nodes of the node, and the neighbor nodes within the far-domain range of a node are called the far-neighbor nodes of the node.

[0046] In addition, the nodes in the airborne network have the ability to send ISAC signals and receive and process echo signals. The nodes obtain the number of potential neighbors and the rough relative distance in the current beam based on the received echo signals.

[0047] The present invention specifically comprises the following steps:

[0048] S1. All nodes synchronize their time through GPS and obtain the current time and their own location information.

[0049] S2. The neighbor discovery process of the node includes multiple time slots, each of which is divided into two sub-time slots; wherein, the working state of the node in each sub-time slot includes the sending state, the receiving state, and the silent state; in particular, in the initial perception phase of neighbor discovery, the working state of the node at the activation time is the sending state, and the working state at the non-activation time is the silent state; in the adaptive handshake phase of neighbor discovery, at the beginning of each time slot, the probability of the node entering the sending state is δ, and the probability of entering the receiving state is 1-δ. In addition, the node has the ability to quickly switch the working state in each sub-time slot. If the node sends a Hello message frame in the first sub-time slot, it switches to the receiving state in the second sub-time slot and listens to the ACK message frame on the corresponding beam; if the node is in the receiving state in the first sub-time slot, if the Hello message frame is successfully received, it switches to the sending state in the second sub-time slot and sends the ACK message frame on the corresponding beam; if the node is in the receiving state in the first sub-time slot but does not receive the Hello message frame, it switches to the silent state in the second sub-time slot.

[0050] S3. Initialize the neighbor list, distance table and potential neighbor list of each node; the neighbor list records the neighbor information discovered through handshake, including neighbor node ID and location information; the distance table records the perceived coarse relative distance to the neighbor node, and the coarse relative distance is defined as the estimated neighbor node distance obtained after processing the echo signal.

[0051] For each node i, the potential neighbor list includes a potential distant neighbor sub-table and a potential near neighbor sub-table, which respectively record the number of potential distant neighbors sensed on each narrow beam, denoted as And the number of potential neighbors sensed on each wide beam, denoted as Among them, potential close neighbors are defined as neighbor nodes within the perceived wide beam transmission range, and potential distant neighbors are defined as neighbor nodes within the perceived narrow beam transmission range, excluding close neighbor nodes that have been counted as potential close neighbors, which can be calculated by judging the rough distance to the neighbor nodes. In addition, by converting the solid angles of the corresponding directions of the wide beam and the narrow beam, the total number of potential neighbors in each wide beam k direction can be obtained, which is recorded as Define the total number of potential near neighbors and the total number of potential far neighbors as N n and N r ,have and N is the total number of neighbors (the total number of nodes in the network is N+1, and it is assumed that all are within the communication range of each other, so any node has N neighbors, and the total number of neighbors of each node is N).

[0052] Definition i,k is the number of discovered neighbor nodes in the kth wide beam direction of node i, b i,l is the number of discovered distant neighbor nodes in the lth narrow beam direction of node i, n i,k is the total number of discovered distant neighbor nodes and near neighbor nodes in the k-th wide beam direction of node i; initialization, let a i,k =0, b i,l =0,n i,k =0.

[0053] S4. In the initial perception phase of neighbor discovery, each node determines its own activation time slot based on the hash function. The specific expression is: Where ID active represents the active node ID, T represents the current time slot number, K N Indicates the number of narrow beams; the narrow beam number used by the activated node in the current time slot number T is l = (Tmod(K N N))mod(K N ).

[0054] After determining the activation time, node i uses a determined narrow beam to send an ISAC signal in the first sub-time slot of its activation time, receives an echo signal in the second sub-time slot, and repeats this step by switching beams in each time slot until all directions are traversed. In the second sub-time slot, the activated node i obtains the distribution of neighbor nodes on each beam based on the echo signal ( and ) and relative distance, and record them in the potential neighbor list and distance table. For inactive nodes, a narrow beam is randomly selected for listening at the inactive time. If a Hello message frame is successfully received in the first sub-time slot of the current time slot, an ACK message frame is sent back in the second sub-time slot; otherwise, the node remains silent in the second sub-time slot. Figure 3 shown.

[0055] S5. If the current time slot exceeds the preset number of time slots in the initial perception phase, the node exits the neighbor discovery initial perception phase and enters the neighbor discovery adaptive handshake phase. Here, the preset number of time slots in the initial perception phase is set to T ini =2(N+1)K N , ensuring that each node successfully traverses K N A narrow beam.

[0056] S6. In the neighbor discovery adaptive handshake phase, each node initializes the probability of selecting each beam for transmission and reception based on the neighbor list, potential neighbor list and distance table. The node determines the beam distribution of the potential neighbors to be discovered based on the information obtained in the initial perception phase, and calculates the beam selection probability. The calculation basis is that the beam with a large number of neighbors is selected with a high probability, and the beam without neighbors will not be selected. Specifically, when node i is in the sending state, the probability of selecting the kth wide beam from the wide beam set for transmission is recorded as p i,k , the probability of selecting the lth narrow beam from the narrow beam set is recorded as q i,k , respectively assign initial values: When node i is in the receiving state, the probability of selecting the kth wide beam to receive from the wide beam set is recorded as r i,k , assign initial value

[0057] In addition, in order to reduce the collision when the receiving node sends back the ACK message frame, the receiving node is designed to send back the ACK message frame in a probabilistic manner; specifically, the probability of the receiving node in the k-th wide beam direction of the sending node i sending back the ACK message frame is denoted as α i,k , assign initial value: The probability that the receiving node in the lth narrow beam direction of the sending node i sends back an ACK is denoted as α i ' ,l , assign initial value: Among them, ε is a preset constant parameter of the system, and ε∈(0,1].

[0058] S7. In the neighbor discovery adaptive handshake phase, at the beginning of each time slot, each node i determines its own transmit and receive state based on the preset probability δ; if node i is in the transmitting state in the current time slot, it first determines whether all potential neighbors have been added to the neighbor list; here, if node i still has potential neighbors that have not been added to the neighbor list, it sends a wide beam probability set {p i,k ,k∈[1,K W ]}Randomly select a wide beam, send a Hello message frame with the selected wide beam in the first sub-slot of the current time slot, and then switch to the receiving state in the second sub-slot; otherwise, based on the probability set {q i,l,l∈[1,K N ]} randomly selects a narrow beam l, sends a Hello message frame using the selected narrow beam l in the first sub-slot of the current time slot, and then switches to the receiving state in the second sub-slot; in addition, if node i is in the receiving state in the current time slot, then based on the receiving wide beam probability set {r i,k ,k∈[1,K W ]}Randomly select a wide beam and listen to the Hello message frame in the selected wide beam direction in the first sub-slot of the current time slot.

[0059] In particular, the present invention designs a Hello message frame structure, such as Figure 4 As shown, it mainly includes:

[0060] "Type", message frame type, used to indicate that the message frame is a Hello message frame. 0 indicates a Hello message frame, and 1 indicates an ACK message frame;

[0061] "Node_ID", used to indicate the ID of the sending node;

[0062] "NL" is used to indicate the neighbor list of the sending node, which records the ID and location information of the neighbor nodes that the current sending node has discovered;

[0063] "Beam_Type" is used to indicate whether the beam type used by the transmitting node is a wide beam or a narrow beam. 0 indicates a wide beam and 1 indicates a narrow beam.

[0064] "Beam_ID", used to indicate the beam number used by the sending node;

[0065] "Location" is used to indicate the current location of the sending node;

[0066] "ACK_Pro" is used to indicate the probability of the receiving node sending back an ACK message frame.

[0067] If the two nodes are aligned in the beam direction selected in the current time slot, and one is in the sending state and the other is in the receiving state, and the receiving node is within the transmission range of the beam selected by the sending node, and there is no collision of message frames, then the receiving node can successfully receive the Hello message frame sent by the sending node in the first sub-time slot and send an ACK message frame back to the sending node in the second sub-time slot; otherwise, the receiving node remains silent in the second sub-time slot until the end of the time slot.

[0068] S8. In the neighbor discovery adaptive handshake phase, if the sending node i successfully receives the ACK message frame in the second sub-time slot of the current time slot, it is considered that the neighbor node is successfully discovered. If the neighbor node discovered this time is a close neighbor node, let a i,k =ai,k +1,n i,k =n i,k +1; if it is a distant neighbor node, let b i,l =b i,l +1,n i,k′ =n i,k′ +1, k′ represents the corresponding number of the wide beam direction including the narrow beam l direction.

[0069] Node i extracts the ID information and location information in the ACK message frame, updates its neighbor list, and updates the beam selection probability p i,k ,q i,l , r i,k , and ACK probability α i,k , α i ' ,l ; Among them, the probability update rules are:

[0070] If the receiving node j successfully receives the Hello message frame in the first sub-slot of the current time slot, it extracts the ID information and location information, and first determines whether the neighbor node has been added to the neighbor list. If the current neighbor node has not been added to the neighbor list, it updates its own neighbor list. Similarly, if it is a close neighbor node, let a j,k =a j,k +1,n j,k =n j,k +1, if it is a distant neighbor node, then let b j,l =b j,l +1,n j,k′ =n j,k′ +1. Then, node j updates the beam selection probability and ACK probability, and the update rules are the same as above; if the current neighbor has been added to the neighbor list, node j will not send back the ACK message frame and switch to the silent state in the second sub-time slot;

[0071] S9. If the node's neighbor list records information of all N neighbors, the node's neighbor discovery ends, the node enters silence, and no longer participates in handshake interactions with other nodes; otherwise, the node enters the next time slot and uses the next time slot as the current time slot, and returns to step S7.

[0072] S10. If neighbor discovery of all nodes is completed, neighbor discovery of the entire network is completed; if the total number of time slots for neighbor discovery exceeds the number of time slots preset by the network, neighbor discovery of the entire network is completed.

Claims

1. A hierarchical directional neighbor discovery method for airborne networks based on integrated communication and perception. It is assumed that the total number of nodes in the airborne network is N+1, and each node is identified by a unique ID. Each node is equipped with a phased array antenna and has the ability to generate narrow and wide beams. The number of narrow beams and wide beams required for a node to traverse and scan 360° is denoted as K N and K W ; The nodes number the beam directions according to a uniform rule; Each node is within the communication range of other nodes; the near-domain range represents the communication range of a node when it uses a wide beam, and the far-domain range represents the communication range of a node when it uses a narrow beam; the neighbor nodes within the near-domain range of a node are called the near-neighbor nodes of the node, and the neighbor nodes within the far-domain range of a node are called the far-neighbor nodes of the node; the nodes in the airborne network have the ability to send ISAC signals and receive and process echo signals. The nodes obtain the number of potential neighbors and the rough relative distance in the current beam based on the received echo signals; It is characterized in that The method specifically comprises the following steps: S1. All nodes synchronize their time through GPS and obtain the current time and their own location information; S2. The neighbor discovery process of the node includes multiple time slots, each of which is divided into two sub-time slots; wherein, the working state of the node in each sub-time slot includes a sending state, a receiving state, and a silent state; in the initial perception phase of neighbor discovery, the working state of the node at the activation time is a sending state, and the working state at the non-activation time is a silent state; in the adaptive handshake phase of neighbor discovery, at the beginning of each time slot, the probability of the node entering the sending state is δ, and the probability of entering the receiving state is 1-δ; the node has the ability to quickly switch the working state in each sub-time slot. If the node sends a Hello message frame in the first sub-time slot, it switches to a receiving state in the second sub-time slot and listens for an ACK message frame on the corresponding beam; if the node is in a receiving state in the first sub-time slot, if the Hello message frame is successfully received, it switches to a sending state in the second sub-time slot and sends an ACK message frame on the corresponding beam; if the node is in a receiving state in the first sub-time slot but does not receive a Hello message frame, it switches to a silent state in the second sub-time slot; S3, initialize the neighbor list, distance table and potential neighbor list of each node; the neighbor list records the neighbor information discovered through handshake, including neighbor node ID and location information; the distance table records the perceived coarse relative distance to the neighbor node, and the coarse relative distance is defined as the estimated neighbor node distance obtained after processing the echo signal; For each node i, the potential neighbor list includes a potential distant neighbor sub-table and a potential near neighbor sub-table, which respectively record the number of potential distant neighbors sensed on each narrow beam, denoted as And the number of potential neighbors sensed on each wide beam, denoted as Among them, potential close neighbors are defined as neighbor nodes within the perceived wide beam transmission range, and potential distant neighbors are defined as neighbor nodes within the perceived narrow beam transmission range, excluding close neighbor nodes that have been counted as potential close neighbors, which can be obtained by judging the rough distance to the neighbor nodes; in addition, by converting the solid angles of the corresponding directions of the wide beam and the narrow beam, the total number of potential neighbors in each wide beam k direction can be obtained, which is recorded as Define the total number of potential near neighbors and the total number of potential far neighbors as N n and N r ,have and N is the total number of neighbors; Definition i,k is the number of discovered neighbor nodes in the kth wide beam direction of node i, b i,l is the number of discovered distant neighbor nodes in the lth narrow beam direction of node i, n i,k is the total number of discovered distant neighbor nodes and near neighbor nodes in the k-th wide beam direction of node i; initialization, let a i,k =0, b i,l =0,n i,k =0; S4. In the initial perception phase of neighbor discovery, each node determines its own activation time slot based on the hash function. The specific expression is: Where ID active represents the active node ID, T represents the current time slot number, K N Indicates the number of narrow beams; the narrow beam number used by the activated node in the current time slot number T is l = (Tmod(K N N))mod(K N ); After determining the activation time, node i uses a determined narrow beam to send an ISAC signal in the first sub-time slot of its activation time, receives an echo signal in the second sub-time slot, and repeats this step by switching beams in the time slot until all directions are traversed; in the second sub-time slot, the activated node i obtains the distribution and relative distance of neighbor nodes in each beam based on the echo signal, and records them in the potential neighbor list and distance table; for inactive nodes, a narrow beam is randomly selected for listening at the inactive time. If a Hello message frame is successfully received in the first sub-time slot of the current time slot, an ACK message frame is sent back in the second sub-time slot; otherwise, the node remains silent in the second sub-time slot; S5. If the current time slot exceeds the preset number of time slots in the initial perception phase, the node exits the neighbor discovery initial perception phase and enters the neighbor discovery adaptive handshake phase; here, the preset number of time slots in the initial perception phase is set to T ini =2(N+1)K N , ensuring that each node successfully traverses K N A narrow beam; S6. In the neighbor discovery adaptive handshake phase, each node initializes the probability of selecting each beam for transmission and reception based on the neighbor list, potential neighbor list and distance table. The node determines the beam distribution of the potential neighbors to be discovered based on the information obtained in the initial perception phase, and calculates the beam selection probability. The calculation basis is that the beam with a large number of neighbors is selected with a high probability, and the beam without neighbors will not be selected. When node i is in the sending state, the probability of selecting the kth wide beam from the wide beam set for transmission is recorded as p i,k , the probability of selecting the lth narrow beam from the narrow beam set to send is denoted as q i,k , respectively assign initial values: When node i is in the receiving state, the probability of selecting the kth wide beam to receive from the wide beam set is denoted as r i,k , assign initial value In order to reduce the collision when the receiving node sends back the ACK message frame, the receiving node is designed to send back the ACK message frame in a probabilistic manner; the probability of the receiving node in the kth wide beam direction of the sending node i sending back the ACK message frame is denoted as α i,k , assign initial value: The probability that the receiving node in the lth narrow beam direction of the sending node i sends back an ACK is denoted as α i ' ,l , assign initial value: Among them, ε is the preset constant parameter of the system, and ε∈(0,1] is taken; S7. In the neighbor discovery adaptive handshake phase, at the beginning of each time slot, each node i determines its own transmit and receive state based on the preset probability δ; if node i is in the transmitting state in the current time slot, it first determines whether all potential neighbors have been added to the neighbor list; here, if node i still has potential neighbors that have not been added to the neighbor list, it sends a wide beam probability set {p i,k ,k∈[1,K W ]}Randomly select a wide beam, send a Hello message frame with the selected wide beam in the first sub-slot of the current time slot, and then switch to the receiving state in the second sub-slot; otherwise, based on the probability set {q i,l ,l∈[1,K N ]} randomly selects a narrow beam l, sends a Hello message frame using the selected narrow beam l in the first sub-slot of the current time slot, and then switches to the receiving state in the second sub-slot; in addition, if node i is in the receiving state in the current time slot, then based on the receiving wide beam probability set {r i,k ,k∈[1,K W ]}Randomly select a wide beam and listen to the Hello message frame in the direction of the selected wide beam in the first sub-time slot of the current time slot; If the two nodes are aligned in the beam direction selected in the current time slot, and one is in the sending state and the other is in the receiving state, and the receiving node is within the transmission range of the beam selected by the sending node, and there is no collision of message frames, then the receiving node can successfully receive the Hello message frame sent by the sending node in the first sub-time slot and send an ACK message frame back to the sending node in the second sub-time slot; otherwise, the receiving node remains silent in the second sub-time slot until the end of the time slot; S8. In the neighbor discovery adaptive handshake phase, if the sending node i successfully receives the ACK message frame in the second sub-time slot of the current time slot, it is considered that the neighbor node is successfully discovered; if the neighbor node discovered this time is a close neighbor node, let a i,k =a i,k +1,n i,k =n i,k +1; if it is a distant neighbor node, let b i,l =b i,l +1,n i,k′ =n i,k′ +1, k′ indicates the corresponding number of the wide beam direction including the narrow beam l direction; Node i extracts the ID information and location information in the ACK message frame, updates its neighbor list, and updates the beam selection probability p i,k ,q i,l , r i,k , and ACK probability α i,k , α i ' ,l ; Among them, the probability update rules are: If the receiving node j successfully receives the Hello message frame in the first sub-time slot of the current time slot, it extracts the ID information and location information, and first determines whether the neighbor node has been added to the neighbor list; if the current neighbor node has not been added to the neighbor list, it updates its own neighbor list; if it is a close neighbor node, it sets a j,k =a j,k +1,n j,k =n j,k +1, if it is a distant neighbor node, then let b j,l =b j,l +1,n j,k′ =n j,k′ +1; then, node j updates the beam selection probability and ACK probability, and the update rules are the same as above; if the current neighbor has been added to the neighbor list, node j will not send back the ACK message frame and switch to the silent state in the second sub-time slot; S9. If the node's neighbor list records information of all N neighbors, the node's neighbor discovery ends, the node enters silence, and no longer participates in handshake interactions with other nodes; otherwise, the node enters the next time slot and uses the next time slot as the current time slot, and returns to step S7; S10. If neighbor discovery of all nodes is completed, neighbor discovery of the entire network is completed; if the total number of time slots for neighbor discovery exceeds the number of time slots preset by the network, neighbor discovery of the entire network is completed.

2. The method for hierarchical directional neighbor discovery in an airborne network based on integrated communication and perception according to claim 1, characterized in that: Define a Hello message frame structure, including: "Type", message frame type, used to indicate that the message frame is a Hello message frame. 0 indicates a Hello message frame, and 1 indicates an ACK message frame. "Node_ID", used to indicate the ID of the sending node; "NL" is used to indicate the neighbor list of the sending node, recording the ID and location information of the neighbor nodes that the current sending node has discovered; "Beam_Type" is used to indicate whether the beam type used by the transmitting node is wide beam or narrow beam. 0 indicates wide beam and 1 indicates narrow beam. "Beam_ID", used to indicate the beam number used by the sending node; "Location", used to indicate the current location of the sending node; "ACK_Pro" is used to indicate the probability of the receiving node sending back an ACK message frame.

Images