Free space optical fiber communication network routing method under delay tolerant network

Abstract

The invention discloses a free space optical fiber communication network routing method under a delay tolerant network, and relates to the technical field of wireless communication. The method comprises the following steps: sensing the signal quality of a link between a node and a neighbor node in real time by the node, and determining a discrete turbulence state identifier based on the signal quality; dynamically correcting a pre-stored historical connection probability value according to the identifier, and generating a connection evaluation value subjected to channel state correction; in combination with the resource state of the neighbor node, calculating a comprehensive forwarding priority, and deciding message forwarding or local caching according to a priority threshold; and establishing a forwarding mechanism triggered by channel state improvement for the cache message, and adaptively updating the correction parameters. The invention solves the technical problems of inaccurate path selection, low data transmission success rate and low resource utilization efficiency caused by the fact that the real-time atmospheric turbulence state is not fully considered in the existing routing method.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and more specifically, to a routing method for free-space optical fiber communication networks under a delay-tolerant network. Background Technology

[0002] Free-space optical communication (FSO) technology has potential applications in latency-tolerant networks in specific scenarios due to its advantages such as high bandwidth and flexible deployment. However, FSO link performance is highly susceptible to atmospheric turbulence, leading to signal attenuation, jitter, and even outages, resulting in a highly dynamic and intermittently connected network topology. This poses a challenge to traditional routing protocols based on assumptions of fixed topology or continuous connectivity.

[0003] Delay-tolerant networks (DTNs) address network outages through a "store-carry-forward" mechanism, and their routing algorithms largely rely on historical encounter information for probabilistic prediction. For example, Chinese patent CN120455354A discloses a routing method that calculates expected transmission values ​​by constructing a node link periodic table. This method primarily relies on statistical predictions based on historical connection and disconnection durations. While it can adapt to certain network dynamics, it does not consider real-time physical layer conditions (such as instantaneous turbulence intensity) that affect FSO link quality as decision variables. When atmospheric conditions change drastically, relying solely on historical statistical information cannot accurately reflect the link's immediate transmission capacity, easily leading to the selection of low-quality links for invalid attempts, reducing routing success rate and increasing network overhead.

[0004] On the other hand, routing optimization research for FSO networks typically considers channel models. For example, Chinese patent CN112688733A discloses a routing method combining an atmospheric channel model and hierarchical topology. This method optimizes paths by modeling atmospheric turbulence and employing an ant colony algorithm. However, such methods often focus on offline, globally information-based path calculation or static optimization, making it difficult to adapt to the characteristics of node movement and transient connection opportunities in delay-tolerant networks. They lack rapid response to instantaneous link states and distributed adaptive decision-making capabilities.

[0005] Therefore, in a network environment where FSO and DTN are integrated, there is an urgent need for a method that can perceive the physical channel status in real time and dynamically adjust the routing strategy accordingly, so as to make more accurate and efficient forwarding decisions when link conditions change rapidly. Summary of the Invention

[0006] To overcome the aforementioned deficiencies in the prior art, embodiments of the present invention provide a routing method for free-space optical fiber communication networks under a delay-tolerant network, thereby addressing the problems mentioned in the background art. Specifically, in a free-space optical communication delay-tolerant network environment, the routing decision may not fully consider the real-time changing physical link state, potentially leading to low transmission success rates and poor network resource utilization efficiency.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a routing method for a free-space optical fiber communication network under a delay-tolerant network, comprising steps S1 to S3: S1: The node acquires the signal quality observation values ​​of the links with each neighboring node in real time, and assigns a discrete current turbulence state identifier to each link according to the preset range to which the observation value belongs, so as to transform the continuous physical channel state into a discrete routing decision input. S2: For each neighboring node, obtain its corresponding historical connection strength reference value; according to the current turbulence state identifier corresponding to the link of the neighboring node, apply a state-related correction operation to the historical connection strength reference value to generate a channel state-corrected connection evaluation value. This operation is used to dynamically adjust the credibility of historical information based on real-time channel quality. S3: Calculate a comprehensive forwarding priority for each neighbor node based on the connection evaluation value corrected by the channel state; if there is a neighbor node whose comprehensive forwarding priority is higher than the first priority threshold, forward the message to the neighbor node with the highest comprehensive forwarding priority; otherwise, store the message in the local cache queue.

[0008] Furthermore, in step S1, the allocation of the current turbulence state identifier based on the signal quality observation value specifically involves: comparing the signal quality observation value with a first quality threshold and a second quality threshold that is greater than the first quality threshold; when the observation value is better than the first quality threshold, allocating a first state identifier; when the observation value is between the first and second quality thresholds, allocating a second state identifier; and when the observation value is worse than the second quality threshold, allocating a third state identifier, thereby dividing the channel quality into multiple discrete levels to facilitate standardization processing by the routing algorithm.

[0009] Furthermore, in step S2, the state-related correction operation is as follows: a set of correction coefficients associated with the turbulence state identifier is pre-defined; a corresponding correction coefficient is selected from the set according to the current turbulence state identifier; the historical connection strength reference value is multiplied by the selected correction coefficient to obtain the connection evaluation value corrected by the channel state; wherein, the set of correction coefficients satisfies the following: the correction coefficient corresponding to the first state identifier is greater than the correction coefficient corresponding to the second state identifier, and the correction coefficient corresponding to the second state identifier is greater than the correction coefficient corresponding to the third state identifier, so as to achieve a lower dependence on historical connection information when the real-time channel conditions are worse.

[0010] Furthermore, in step S3, the method for calculating the comprehensive forwarding priority is as follows: the connection evaluation value corrected by the channel state is multiplied by a factor reflecting the proportion of available cache resources of neighboring nodes, and the product is used as the comprehensive forwarding priority, so that the forwarding decision considers both the link connection probability and the immediate load capacity of the next hop node.

[0011] Furthermore, following step S3, step S4 is also included: S4: For messages stored in the local cache queue, continuously monitor the current turbulence status identifier of their potential next-hop neighbor nodes; When the current turbulence state identifier of any potential next-hop neighbor node is detected to change from the third state identifier to the first state identifier or the second state identifier, steps S2 and S3 are re-executed for that neighbor node. If the recalculated overall forwarding priority is higher than the first priority threshold, the message is retrieved from the cache queue and forwarded to the neighboring node, thereby providing a secondary forwarding opportunity for the cached message based on channel state improvement.

[0012] Furthermore, in step S4, when the current turbulence status indicators of multiple potential next-hop neighbor nodes are simultaneously improved and trigger recalculation, the node with the highest overall forwarding priority after recalculation is selected for message forwarding, so as to achieve the optimal selection among multiple potential opportunities.

[0013] Furthermore, in step S3, when multiple neighboring nodes have the same overall forwarding priority and are all the highest, the neighboring node with the better channel quality represented by the current turbulence state identifier is selected for message forwarding, so as to give priority to the link with better real-time status when the connection evaluation is similar.

[0014] Furthermore, it also includes step S5: dynamically updating the correction coefficients used in the state-related correction operation; S5: Record the neighbor nodes used for each message forwarding and their corresponding current turbulence status identifier, and record whether the forwarding was successfully completed; For each type of turbulence state identifier, the percentage of messages successfully forwarded under that identifier is calculated. If the success rate under a certain type of identifier is consistently lower than the preset expected rate, the correction coefficient corresponding to that type of identifier in the set of correction coefficients is increased, so that the system can adaptively optimize routing parameters based on historical performance feedback.

[0015] Furthermore, when there are multiple pending messages in the node's local cache queue, step S3 includes: For each message, the overall forwarding priority of all reachable neighbor nodes is calculated independently; For each message, among its reachable neighbor nodes, nodes with a comprehensive forwarding priority higher than the first priority threshold are selected to form a candidate node set for the message; According to the preset scheduling rules, the message to be finally forwarded is selected from the candidate node set of all messages and paired with the neighbor node to perform global scheduling when there are multiple messages.

[0016] Furthermore, the preset scheduling rule is as follows: prioritize scheduling messages whose candidate node set contains neighbor nodes whose current turbulence state identifier is the first state identifier; for messages whose candidate node set does not contain any first state identifier nodes, compare the optimal comprehensive forwarding priority of each message candidate node set node, and schedule the message with the highest priority, thereby prioritizing the use of high-quality channels globally and maximizing the benefits of a single forwarding.

[0017] The technical effects and advantages of this invention are as follows: Nodes periodically convert link signal quality observations into turbulence state identifiers, reflecting the current transmission conditions of the link. Subsequently, nodes adjust historical connection probability values ​​based on these identifiers, resulting in a connection evaluation value that integrates historical patterns and real-time status. This evaluation value is combined with a resource factor reflecting the current load of neighboring nodes to calculate forwarding priority. This calculation process allows routing decisions to consider both real-time channel availability and the current carrying capacity of the next-hop node when assessing connection opportunities, thus balancing transmission success probability and network load distribution at the decision-making level. Compared to methods relying solely on historical statistics, this approach more accurately assesses instantaneous forwarding opportunities, helping to reduce invalid transmission attempts under adverse channel conditions.

[0018] For messages cached due to the lack of a suitable path, this invention establishes a re-evaluation mechanism triggered by improved channel conditions. When an improvement in the link condition of a potential next-hop node is detected, the system automatically re-executes route calculation for that node and the corresponding message. If the calculation result meets the conditions, forwarding is triggered immediately. This mechanism reduces the reliance on fixed retransmission intervals or passive waiting, enabling messages to be forwarded more promptly when link conditions improve. This helps reduce the average end-to-end latency of messages and improves the utilization efficiency of short-lived connection opportunities.

[0019] This invention includes a parameter adaptive update process. The system records forwarding results under different turbulence state identifiers and calculates the actual success rate. By comparing the actual success rate with a preset target, the correction parameters in the routing calculation can be slowly adjusted. This process enables the routing method to fine-tune the decision threshold or correction coefficients based on long-term operational statistics, thereby giving the entire system the ability to adapt to specific network environments and channel variation patterns, improving the applicability and stability of the method in different deployment scenarios. Attached Figure Description

[0020] Figure 1 This is a flowchart of the routing decision process of the present invention; Figure 2 This is a schematic diagram of the dynamic correction process for connection evaluation values ​​according to the present invention; Figure 3 This is a branch structure diagram of the forwarding decision and caching processing of the present invention; Figure 4 This is a schematic diagram of the adaptive update process for the correction parameters of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1 As attached Figures 1 to 4 The above describes a routing method for free-space optical fiber communication networks under a delay-tolerant network. This method is implemented through the following process: First, nodes perceive and quantify the atmospheric turbulence state of communication links with their neighboring nodes in real time; second, the quantified real-time turbulence state is used as input to dynamically correct traditional routing metrics based on historical statistics. Then, based on the corrected metric and node context information, a decision is made on whether to forward the message or cache it locally; at the same time, a triggered forwarding mechanism based on channel state improvement is established for cached messages; in addition, the system can adaptively update the internal correction parameters according to the historical decision effects. Finally, when a node faces multiple messages to be forwarded, it performs queue scheduling based on global link state assessment.

[0023] I. Implementation Scenarios and System Framework This method is deployed in a delay-tolerant network consisting of multiple mobile or fixed nodes, with data transmission between nodes via free-space optical communication links. Each network node needs to have the following basic functional modules: a laser signal transceiver, a processor for running a routing protocol stack, a non-volatile data storage device, and an internal bus for communication between modules.

[0024] The routing decision logic described in this method is integrated as a software module into the node's protocol stack. Its operation depends on inputs provided by other basic modules and generates corresponding control outputs. The specific interaction relationships are as follows: Dependencies on inputs and external modules: 1. Neighbor Discovery Module: Through physical layer and link layer protocols, it maintains a dynamic "neighbor node list," providing the identifiers of currently detectable neighboring nodes. This list is the source of the next-hop candidate set in routing decisions.

[0025] 2. Historical Connection Information Database: Stores and maintains "historical connection strength reference values" related to each known node, denoted as... This value is a real number in the interval [0, 1], and its value is updated based on historical encounter events between nodes and a time decay function. This module provides connection probability predictions based on long-term statistics.

[0026] 3. Node State Exchange Module: Responsible for periodically exchanging control messages containing resource status with neighboring nodes. By parsing the received messages, it obtains and updates the "current remaining cache capacity" of each neighboring node. Compared to "total cache capacity" .

[0027] 4. Local Message Management Module: Manages the queue of data messages to be forwarded. Each message is accompanied by metadata such as destination address, generation time, and time to life.

[0028] The resulting output and actions: 1. Forwarding instruction: Contains a clear target message identifier and the identifier of the selected next-hop neighbor node, and is sent to the underlying communication module to perform the sending operation.

[0029] 2. Caching directive: Instructs a specific message to be stored in a local cache queue, and may include its initial priority position in the queue.

[0030] 3. Queue scheduling instruction: When it is necessary to select messages from the cache queue for sending, output the sorting or selection result of the messages to be processed according to the strategy.

[0031] 4. Parameter update command: Based on the system's operating history, output the adjustment amount for the internal routing calculation parameters.

[0032] II. Specific Implementation Steps Step S1: Real-time link status awareness and turbulence label allocation The goal of this step is to transform the continuously varying physical layer channel quality on the free-space optical communication link into a discrete, digital state identifier, providing a standardized input for subsequent routing metric calculations.

[0033] Acquisition of signal quality observations: A node uses its optical receiver to measure the received power of optical signals from specific neighboring nodes. To obtain stable statistical characteristics, a fixed observation time window is set. . The value of this parameter needs to balance the real-time nature of state updates with the stability of measurement results, and its range is typically between 50 milliseconds and 300 milliseconds. For example, it can be set to... millisecond.

[0034] In the time window Within, the nodes operate at a fixed sampling frequency. (For example The received power is sampled at kilohertz to obtain a signal containing... Discrete sequence of sampling points .

[0035] Calculate the variance of this sequence, which serves as a signal quality observation characterizing the intensity of signal fluctuations, denoted as Var: ; in, This is the arithmetic mean of the sequence. The magnitude of the Var value directly reflects the degree of fluctuation in signal power within the observation window, which is mainly caused by atmospheric turbulence.

[0036] Assignment of turbulence state identifiers: Each node is pre-configured with two non-negative decision thresholds: a first quality threshold and a second quality threshold. Second mass threshold And satisfy These two thresholds are used to classify channel quality levels and can be determined through theoretical analysis, simulation calibration, or field experiments.

[0037] The allocation process involves deterministic logical judgments: 1. If Then, a first state identifier is assigned to the link, denoted as . (It can be represented by the value 1 in internal processing).

[0038] 2. If Then, a second state identifier is assigned to the link, denoted as . (This can be represented by the value 2 in internal processing).

[0039] 3. If Then, a third state identifier is assigned to the link, denoted as . (This can be represented by the value 3 in internal processing).

[0040] Each node maintains a state variable for each neighbor node to store the "current turbulence state identifier" of that link. This variable is updated after each observation calculation is completed.

[0041] Step S2: Correction of connection evaluation value based on real-time status This step is executed when the routing decision process is triggered (e.g., when there are messages to be forwarded and processed, or when triggering conditions are met). Its purpose is to incorporate the real-time channel state information obtained in step S1 into the connection probability assessment based on historical statistics.

[0042] Input retrieval: Read the neighbor nodes to be evaluated from the historical connection information database. corresponding .

[0043] Obtain the neighbor node from the output of step S1. corresponding .

[0044] Correction coefficient mapping: Internally, each node stores a static table or function called the "correction coefficient set". This set defines the parameters derived from the turbulence state identifier. To the correction factor The mapping relationship. A feasible mapping example is as follows: This mapping relationship must satisfy the following condition: the worse the quality of the channel state indicator (i.e., the lower the quality of the channel state indicator), the more... When the value of ( ) is larger, the corresponding correction factor is The smaller the value, the better. This design reflects the moderating effect of real-time channel conditions on the reliability of historical information.

[0045] Corrected calculation: The correction coefficient is obtained based on the above mapping. Then, calculate the channel-state-corrected connectivity assessment value using the following formula. : ; At this point, It becomes a comprehensive evaluation value that simultaneously contains historical connection patterns and current instantaneous link quality information.

[0046] Step S3: Comprehensive forwarding priority calculation and decision This step utilizes the output of step S2 and combines it with the real-time resource status of neighboring nodes to select the next-hop node or make a caching decision for the message to be forwarded.

[0047] Resource factor calculation: First, calculate the neighboring nodes. Available cached resource ratio factor This factor reflects the node's current load capacity: ; in, and The latest information provided by the node state exchange module.

[0048] Overall forwarding priority calculation: Next, calculate the neighboring nodes. The overall forwarding priority for the next hop of this message : ; This calculation means that the priority of a node becoming a forwarding target depends on both its connectivity evaluation value and its current available storage space.

[0049] Forwarding decision logic: The node sets a first priority threshold. This threshold is a configurable system parameter, typically ranging from 0.2 to 0.5. For example, it can be set to... .

[0050] The decision-making process is as follows: 1. For the message to be forwarded, iterate through all its reachable neighbor nodes and calculate the distance to each node. .

[0051] 2. Find the largest one. Value, denoted as And record the corresponding nodes.

[0052] 3. Judgment condition: If Then a "forwarding instruction" is generated, instructing the recipient to... The corresponding neighbor node sends a message.

[0053] 4. Otherwise, generate a "caching instruction" and store the message in the local cache queue.

[0054] Priority is handled in parallel: If the calculation reveals two or more neighboring nodes Values ​​are equal and both are the maximum value. If so, a second decision is required. The decision is based on the current turbulence state identifiers of these nodes. .choose Nodes with smaller values ​​(i.e., nodes with better real-time channel quality) are selected as forwarding targets.

[0055] Step S4: Triggered forwarding of cached messages For messages already stored in the local cache queue, this step provides a passive, channel state-triggered reassessment and forwarding opportunity, independent of the arrival of new messages.

[0056] Status monitoring: Each node maintains an associated "potential next-hop node list" for each cached message. A separate background monitoring task runs at fixed intervals (e.g., every second) to check the latest update for each node in this list. value.

[0057] Triggering conditions and re-evaluation: When the monitoring task detects a node in the list of The value changed, and the trend of change was from a worse level ( ) transformed into a better level ( or When this happens, a trigger is activated for that message and node. The reassessment process.

[0058] Upon triggering, the system immediately activates the node. Re-execute steps S2 and S3: that is, use the node The latest And new Recalculate And then combined with the latest The new integrated forwarding priority is calculated. .

[0059] Trigger action: Judgment condition: If If the message is not found, a "forwarding instruction" is generated, which removes the message from the cache queue and sends it to the node. .

[0060] Multi-trigger arbitration: If multiple potential next-hop nodes for the same message meet the triggering conditions within the same monitoring period, the system will calculate the triggering conditions of all triggering nodes in parallel. Values, and select one of them. The node with the largest value is selected as the forwarding target.

[0061] Step S5: Adaptive update of correction coefficients This step aims to enable the system to adjust the correction coefficients used in step S2 based on actual operating results. Make slow, adaptive adjustments to optimize long-term performance.

[0062] Decision Records: The node logs each forwarding attempt initiated by step S3 or S4, and each log entry includes at least: the neighbor node identifier, the link's... The decision-making process The value, and a boolean flag indicating whether the forwarding was successful.

[0063] Periodic statistics: This step is executed automatically once at a low frequency (e.g., hourly). During execution, each turbulence state is identified. The following statistics were performed: Count the total number of forwards initiated under this identifier. .

[0064] Count the number of successful reposts. .

[0065] Calculate the actual success rate under this identifier. .

[0066] Parameter adjustment logic: The system presets a desired target success rate for each status indicator. ,For example: , , .

[0067] Define two adjustment parameters: number of evaluation periods (For example ) and adjusting step size (For example ).

[0068] Adjustment rules: For status indicators If its In recent consecutive All statistical periods were lower than Then the corresponding correction coefficient is considered to be... The value may be too high, the system will... value decrease ,Right now To prevent the coefficient from being too small, a lower limit is set. ,make sure After the adjustment is completed, the new The value will replace the original value and be used in all subsequent route calculations.

[0069] III. Multi-Message Queue Scheduling When a node's local cache queue contains multiple (e.g., more than 5) messages awaiting forwarding, the following global scheduling strategy can be used to efficiently manage forwarding opportunities. This strategy operates whenever one or more messages need to be selected from the queue for transmission.

[0070] 1. Independent and parallel evaluation: For each candidate message in the cache queue The node performs the steps from step S1 (getting) in parallel on all its reachable neighbor nodes. ) to step S3 (calculation) The complete evaluation process.

[0071] 2. Construct a candidate set: for each message Based on the evaluation results, a set of candidate nodes is generated. This set contains all that satisfy... Neighbor nodes of the condition .

[0072] 3. Two-level scheduling rules: Level 1 Scheduling – Prioritizing High-Quality Links: First, select all candidate links. It contains at least one Messages from neighboring nodes (i.e., those in the best state). These messages constitute a high-priority message set, HighPriSet. The scheduler prioritizes selecting messages from HighPriSet for processing. During selection, it can choose from candidate nodes... The message with the highest value.

[0073] Second-level scheduling – Optimal opportunity utilization: For the remaining messages not included in the HighPriSet, the scheduler compares each message with its candidate set. The highest of all nodes Value, denoted as Then, choose Process the message with the highest value.

[0074] 4. Parallel transmission support: If the node hardware supports simultaneous transmission... road( For parallel data transmission (where the integer is a positive integer), the two-level scheduling rule described above can be applied iteratively. That is, first, one message is selected, and then the rule is run again among the remaining messages to select the next message, until all messages are selected. Until a message is received or no valid message is received.

[0075] IV. Experimental Verification Examples To verify the actual effectiveness of this routing method, a simulation comparison experiment was conducted in a simulated free-space optical communication delay-tolerant network scenario.

[0076] The experimental design is as follows: 1. Simulation environment and parameters: Scene setting: In a Within a square area, 50 mobile communication nodes are randomly deployed.

[0077] Node mobility: A mobility model based on a real map is adopted, with nodes moving along roads. to Move at a speed of [speed], and stay at a randomly selected location for 10 to 60 seconds.

[0078] Communication parameters: Effective range of laser communication: 800 meters.

[0079] Data transfer rate: .

[0080] Node cache capacity: .

[0081] Atmospheric channel model: adopts a log-normal turbulence model, with refractive index structure constant. exist arrive The range changes dynamically to simulate different atmospheric conditions.

[0082] State observation window ms.

[0083] Business load: in total During a simulation lasting one hour, 500 messages are randomly generated, with the message generation interval following a uniform distribution of 25 to 35 seconds. The message size is within... arrive Randomly selected from among several options. Each message has a lifespan of 300 minutes.

[0084] Comparison Protocols: Two typical routing protocols for delay-tolerant networks were selected as performance benchmarks: Epidemic routing: uses flooding to transmit message copies.

[0085] PROPHET routing: Makes forwarding decisions based on the historical probability of node encounters.

[0086] Performance metrics: Message delivery success rate: The percentage of successfully delivered messages out of the total number of generated messages.

[0087] Average end-to-end latency: The average time taken for all successfully delivered messages to be received from generation to receipt.

[0088] Network overhead: The ratio of the total number of message copies transmitted within the network to the number of successfully delivered messages during the simulation.

[0089] 2. Parameter settings for the method of this invention: Observation window ms.

[0090] Initial values ​​of correction coefficients: , , .

[0091] Forwarding threshold .

[0092] Neighbor state exchange cycle: 10 seconds.

[0093] Coefficient update cycle: Hour.

[0094] The experiment was conducted in multiple rounds to eliminate randomness, and the average results are recorded in the table below: Routing methods Message delivery success rate (%) Average end-to-end delay (min) Network overhead (number of replicas / messages) Epidemic Routing 98.2 85.3 135.7 PROPHET Router 74.6 142.1 18.4 Method of the present invention 82.7 121.5 22.1 The above embodiments detail the specific implementation of the technical solution of the present invention. This method achieves dynamic routing optimization for free-space optical communication delay-tolerant networks by sequentially executing steps such as link state awareness and quantization, real-time state-based routing metric correction, resource-aware comprehensive decision-making, state-triggered reassessment of cached messages, and adaptive parameter updates. Those skilled in the art can adjust and set the configuration parameters mentioned in the embodiments, such as the time window, various thresholds, and initial values ​​of correction coefficients, according to specific application scenarios and network parameters. Such adaptive modifications based on the basic principles of the present invention should all be included within the protection scope of the present invention.

[0095] Finally, the following points should be noted: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installation", "connection", and "linkage" should be interpreted broadly, and can be mechanical or electrical connections, or internal connections between two components, or direct connections. "Up", "down", "left", "right", etc. are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may change. Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A routing method for free-space optical fiber communication networks under delay-tolerant networks, characterized in that, Including steps S1 to S3: S1: The node acquires the signal quality observation values ​​of the links with each neighboring node in real time, and assigns a discrete current turbulence state identifier to each link according to the preset range to which the observation value belongs; S2: For each neighboring node, obtain its corresponding historical connection strength reference value; based on the current turbulence state identifier corresponding to the link of the neighboring node, apply a state-related correction operation to the historical connection strength reference value to generate a connection evaluation value corrected by the channel state. S3: Calculate a comprehensive forwarding priority for each neighbor node based on the connection evaluation value corrected by the channel state; if there is a neighbor node whose comprehensive forwarding priority is higher than the first priority threshold, then forward the message to the neighbor node with the highest comprehensive forwarding priority; otherwise, store the message in the local cache queue.

2. The routing method for free-space optical fiber communication networks under delay-tolerant networks according to claim 1, characterized in that, In step S1, the step of assigning the current turbulence state identifier based on the signal quality observation value specifically involves: comparing the signal quality observation value with a first quality threshold and a second quality threshold that is greater than the first quality threshold; assigning a first state identifier when the observation value is better than the first quality threshold; assigning a second state identifier when the observation value is between the first and second quality thresholds; and assigning a third state identifier when the observation value is worse than the second quality threshold.

3. The routing method for free-space optical fiber communication networks under delay-tolerant networks according to claim 2, characterized in that, In step S2, the state-related correction operation is as follows: a set of correction coefficients associated with the turbulence state identifier is pre-defined; a corresponding correction coefficient is selected from the set according to the current turbulence state identifier; the historical connection strength reference value is multiplied by the selected correction coefficient to obtain the connection evaluation value corrected by the channel state; wherein, the set of correction coefficients satisfies the following: the correction coefficient corresponding to the first state identifier is greater than the correction coefficient corresponding to the second state identifier, and the correction coefficient corresponding to the second state identifier is greater than the correction coefficient corresponding to the third state identifier.

4. The routing method for free-space optical fiber communication networks under delay-tolerant networks according to claim 3, characterized in that, In step S3, the method for calculating the overall forwarding priority is as follows: multiply the channel state-corrected connection evaluation value by a factor that reflects the proportion of available cache resources of neighboring nodes, and use the product as the overall forwarding priority.

5. The routing method for free-space optical fiber communication networks under delay-tolerant networks according to claim 1, characterized in that, After step S3, step S4 is also included: S4: For messages stored in the local cache queue, continuously monitor the current turbulence status identifier of their potential next-hop neighbor nodes; When the current turbulence state identifier of any potential next-hop neighbor node is detected to change from the third state identifier to the first state identifier or the second state identifier, steps S2 and S3 are re-executed for that neighbor node. If the recalculated overall forwarding priority is higher than the first priority threshold, the message is retrieved from the cache queue and forwarded to the neighboring node.

6. The routing method for free-space optical fiber communication networks under delay-tolerant networks according to claim 5, characterized in that, In step S4, when the current turbulence status indicators of multiple potential next-hop neighbor nodes are simultaneously improved and trigger recalculation, the node with the highest overall forwarding priority after recalculation is selected for message forwarding.

7. The routing method for free-space optical fiber communication networks under delay-tolerant networks according to claim 1, characterized in that, In step S3, when multiple neighboring nodes have the same overall forwarding priority and all of them are the highest, the neighboring node with the better channel quality represented by the current turbulence state identifier is selected for message forwarding.

8. The routing method for free-space optical fiber communication networks under delay-tolerant networks according to claim 1, characterized in that, It also includes step S5: dynamically updating the correction coefficients used in the state-related correction operation; S5: Record the neighbor nodes used for each message forwarding and their corresponding current turbulence status identifier, and record whether the forwarding was successfully completed; For each type of turbulence state identifier, the percentage of messages successfully forwarded under that identifier is calculated. If the success rate of a certain type of identifier is consistently lower than the preset expected rate, then the correction coefficient corresponding to that type of identifier in the set of correction coefficients will be increased.

9. The routing method for free-space optical fiber communication networks under delay-tolerant networks according to claim 1, characterized in that, When there are multiple messages to be processed in the node's local cache queue, step S3 includes: For each message, the overall forwarding priority of all reachable neighbor nodes is calculated independently; For each message, among its reachable neighbor nodes, nodes with a comprehensive forwarding priority higher than the first priority threshold are selected to form a candidate node set for the message; According to the preset scheduling rules, the message to be finally executed for forwarding is selected from the candidate node set of all messages and paired with the neighbor node.

10. The routing method for free-space optical fiber communication networks under delay-tolerant networks according to claim 9, characterized in that, The preset scheduling rule is as follows: prioritize scheduling messages whose candidate node set contains neighbor nodes whose current turbulence state identifier is the first state identifier; for messages whose candidate node set does not contain any first state identifier nodes, compare the optimal comprehensive forwarding priority of each message candidate node set, and schedule the message with the highest priority.

Images