A cross-layer vc-tb flow control method and system for a cross-medium heterogeneous communication gateway and a storage medium

Abstract

The application discloses a cross-layer virtual clock-token bucket (VC-TB) flow control method and system for a cross-medium heterogeneous communication gateway and a storage medium, belongs to the technical field of cross-medium and underwater heterogeneous communication, and is suitable for a gateway application scenario of high-speed input, low-speed output, time-varying channel, and interface transmission constraint. The application builds a data queue and a virtual token bucket on the gateway, collects the channel quality and the interface busy state of a downstream link in a cross-layer mode, dynamically adjusts the token generation rate through channel sensing impedance matching, realizes adaptive data injection budget following link carrying capacity, and matches a multi-constraint instant sending gate control mechanism. The application can simultaneously solve heterogeneous link rate mismatch congestion and time domain conflicts caused by half-duplex reception, reduces buffer backlog and invalid packet sending, and can be used in various cross-medium fusion communication networks such as acoustoelectric, acoustooptic, and optoelectric communication networks.

Description

Technical Field

[0001] This invention relates to the fields of cross-media heterogeneous networks, underwater acoustic communication and gateway flow control technology, specifically a cross-layer VC-TB flow control method, system and computer-readable storage medium applicable to heterogeneous link gateways that integrate acoustic-electric, acoustic-optical, optoelectronic and acoustic-optical fusion, which can be applied to buoy gateways, unmanned surface vessel gateways, underwater relays, UAV multi-mode relays and shore-based heterogeneous forwarding gateways. Background Technology

[0002] Applications such as marine environmental monitoring, deep-sea exploration, seabed resource development, marine security, and underwater IoT typically require collaborative communication between underwater nodes, surface nodes, airborne nodes, and shore-based control centers. Because different physical media significantly affect signal propagation, a single communication method is insufficient to cover a complete cross-domain information link. Long-distance underwater communication usually relies on underwater acoustic links, while short-distance high-speed underwater communication can utilize optical communication links. Surface or airborne nodes typically employ radio, electromagnetic, or satellite links. The resulting networks are usually heterogeneous cross-media communication networks.

[0003] Gateway nodes in cross-media communication networks undertake tasks such as caching, protocol conversion, and cross-domain forwarding, and are the key bottleneck connecting links of different media. Taking a marine acoustic-electric cross-media network as an example, control commands, task configurations, and data updates generated by control nodes or UAVs quickly reach the gateway via radio links, and are then forwarded to underwater nodes via underwater acoustic interfaces. Monitoring data uploaded by underwater nodes reaches the gateway via underwater acoustic links, and is then transmitted back via radio links. Radio links typically have speeds in the Mbps range and low propagation delays, while underwater acoustic links typically have only effective bandwidths in the kbps range, accompanied by long propagation delays, noise interference, and time-varying fading. Therefore, a rate mismatch between high-speed injection and low-speed output is prone to occur at the gateway.

[0004] Existing flow control or queue scheduling methods are difficult to adapt to the above scenarios simultaneously. Traditional Static Token Bucket (STB) generates tokens at a fixed rate to limit the average injection strength, but the default link service capacity is relatively stable, lacking awareness of the real-time service capacity of the target forwarding link and the busy / idle status of the interface; Priority Queues (PQ) can reduce the waiting latency of critical downlink services, but they are prone to occupying the target link for a long time and causing uplink or reverse service starvation; Static Weighted Round Robin (WRR) can improve long-term fairness, but its static weights are difficult to dynamically change with link quality and interface busy / idle status; ARED active queue management methods usually adjust the drop probability based on queue length or congestion feedback, and have limited response capabilities under conditions of long propagation delay and feedback lag.

[0005] Therefore, a unified flow control scheme for cross-media heterogeneous communication gateways is urgently needed. This scheme should adjust the logical injection budget in the capacity domain based on the real-time channel quality of the target forwarding link to match the input service with the link's service capacity; simultaneously, in the time domain, it should control the transmission timing based on interface busy / idle or receive protection status to avoid blindly transmitting within the receive window. Summary of the Invention

[0006] To overcome the shortcomings of existing token bucket, queue scheduling, and active queue management methods in simultaneously addressing link time-varying fading and interface busy / idle constraints, this invention provides a lightweight, adaptive cross-layer VC-TB flow control method. Through the collaborative mechanism of channel-aware impedance matching (capacity domain adaptive token generation) and instantaneous transmission gating (time domain interface protection), it achieves heterogeneous link capacity matching and transmit / receive window isolation, reducing queue backlog, packet loss rate, and forwarding latency, and ensuring fairness in uplink and downlink service forwarding.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways is applied to a heterogeneous gateway connecting a first communication medium link and a second communication medium link, wherein the two links differ in at least one parameter among transmission rate, propagation delay, channel quality, effective service capability, and transmit / receive constraints.

[0009] The method includes:

[0010] S1. Establish a data queue and virtual token bucket oriented towards the target forwarding direction in the cross-media heterogeneous communication gateway, and pre-configure: queue capacity limit. , Token bucket capacity limit Benchmark token generation rate Deep decay survival threshold Physical layer demodulation threshold Channel hierarchical step size and time slot length ;

[0011] S2, in each discrete time slot The system retrieves cross-layer status information, which includes the traffic volume arriving via the first communication link interface. Length of the data queue to be forwarded via the second communication link interface Virtual token inventory Channel quality parameters of the second communication medium link and the busy / idle status of the second communication link interface. ;

[0012] S3. Based on the channel quality parameters of the second communication medium link. Perform channel-aware impedance matching to obtain the token generation adjustment factor. and in accordance with Determine the token generation rate for the current time slot to match the logic injection budget with the current service capacity of the second communication medium link;

[0013] S4, according to Update the virtual token bucket to obtain the available transmission budget for the current time slot. The The theoretical total number of tokens added in this time slot to the existing virtual token bucket is the upper limit of the virtual token bucket's capacity. Used to limit the length of a single burst of transmission, prevent the target forwarding link from being occupied for a long time, and reserve reception and access opportunities for reverse services, uplink services or other priority services;

[0014] S5. Determine whether to allow transmission via the second communication link interface based on the instant transmission gating conditions. The instant transmission gating conditions include at least the following: the data queue is not empty, the available transmission budget is sufficient, the second communication medium link has effective service capabilities, and the second communication link interface is not in a receive busy or protection busy state.

[0015] S6. When the instantaneous transmission gating condition is met, transmission is performed via the second communication link interface, and the transmission is based on the data queue length. Available sending budget Second communication medium link effective service capability Determine the actual amount sent If the instantaneous sending gating condition is not met, remain silent and continue accumulating tokens;

[0016] S7. Based on the actual sending volume Update the data queue, virtual token bucket, and busy / idle status of the second communication link interface, and enter the next discrete time slot, thereby realizing capacity domain matching and time domain gating of the receiving window between heterogeneous links across media.

[0017] Furthermore, the cross-medium heterogeneous communication gateway includes one or more of the following: gateway buoy, unmanned surface vessel gateway, underwater relay gateway, shore-based gateway, aerial unmanned aerial vehicle gateway, or multi-mode communication relay node; the first communication medium link and the second communication medium link include any two or more combinations of underwater acoustic link, optical communication link, radio link, electromagnetic link, wired communication link, and satellite link.

[0018] Furthermore, the method is applicable to acoustic-electric cross-medium communication networks, acoustic-optical cross-medium communication networks, optoelectronic cross-medium communication networks, and acoustic-optical-electrical integrated cross-medium communication networks.

[0019] Furthermore, the cross-layer state information in step S2 is obtained from the vertical interaction interface between the data link layer and the physical layer. The physical layer feeds back to the data link layer at least one or more of the following: channel quality parameters, carrier sensing results, receive occupancy status, link synchronization status, packet decoding status, received optical power, bit error rate, or packet power.

[0020] Furthermore, the channel quality parameters This includes one or more of the following: signal-to-noise ratio (SNR), signal-to-interference-plus-noise ratio (SINR), bit error rate (BER), packet error rate (PER), received signal strength indication (RSSI), received optical power, link quality indication (LQI), component power, estimated channel capacity, or estimated effective service rate.

[0021] Furthermore, the channel-aware impedance matching in step S3 specifically includes: adjusting the channel quality parameters of the second communication medium link. Mapping to discrete or continuous token generation adjustment factor When the channel quality is below the deep fading keep-alive threshold, the token generation rate is reduced to the keep-alive level; when the channel quality is in a weak coverage area, the token generation rate is lower than the baseline token generation rate; when the channel quality is in the nominal working area, the baseline token generation rate is used; when the channel quality is in the opportunistic transmission area, the token generation rate is increased.

[0022] Furthermore, step S2 uses a combination of event-driven update and zero-order hold to obtain channel quality parameters. When the second communication link interface is successfully synchronized and receives uplink packets, probe frames, or feedback frames, the corresponding channel quality parameters are extracted and the channel quality record is updated; when no new valid measurement value is detected, the most recent valid measurement value is used as the channel quality estimate for the current time slot.

[0023] Furthermore, the effective service capability of the second communication medium link in step S5 Determined based on the effective link rate, the busy / idle status of the second communication link interface, and random service efficiency, satisfying... ,in, To receive busy / idle indicators, For random service efficiency factor, The time slot length, To standardize the group length, The effective rate of the second communication medium link is determined by the channel quality parameters.

[0024] Furthermore, the busy / idle state of the second communication link interface in step S5 is determined by the remaining state of the receiving / occupied state. Sure, =1 indicates that the second communication link interface is in a busy receiving or busy protection state. =0 indicates that the second communication link interface is in a local idle state; the received residual state is as follows: Update, in which This indicates the average number of time slots used for one uplink packet reception. Indicates time slot A new uplink receive event was detected within the device.

[0025] Furthermore, the decision variables are sent in step S6. Satisfy: When , , and hour, ;otherwise The actual transmission volume meets the requirements. .

[0026] Furthermore, the data queue described in step S7 is arranged according to... Update: The amount of packet loss due to overflow caused by limited queue capacity is expressed as follows: .

[0027] The present invention also provides a cross-layer VC-TB flow control system for cross-media heterogeneous communication gateways, including a first communication link interface, a second communication link interface, a data queue management module, a channel state awareness module, a busy / idle detection module, a virtual token bucket module, a channel awareness impedance matching module, an instant transmission gating module, and a processor;

[0028] The first communication link interface connects to the upstream first communication medium link to collect real-time service data from the entry point. And send it to the queue;

[0029] The second communication link interface connects to the downstream second communication medium link to complete the actual transmission and reception of data packets and to provide feedback on interface occupancy and original channel measurement information.

[0030] The data queue management module: maintains the data queue. ,according to The formula updates the queue length and counts packet loss due to buffer overflow.

[0031] The channel state awareness module collects channel quality parameters across layers from the physical / link layer. Event-driven + zero-order hold-up channel parameter updates;

[0032] The busy / idle detection module: counts the uplink received time slots occupied and calculates the remaining occupied time. Output interface busy / idle indicator ;

[0033] The virtual token bucket module: based on Update token inventory ,by Constraints, output available for sending budget;

[0034] The channel sensing impedance matching module: based on Table lookup generation The system calculates the real-time token generation rate to alleviate heterogeneous link rate mismatch from the capacity domain. When the target forwarding link quality is poor, the token generation rate is reduced to reduce invalid injection and cache bloat. When the target forwarding link is in the nominal working area, the system forwards stably according to the baseline token generation rate. When the target forwarding link quality is good, the token generation rate is increased to quickly release queue backlog by utilizing the high-quality channel window.

[0035] The instant transmission gating module makes a transmission decision based on four conditions: queue, token, link service capability, and interface busy / idle status, and outputs the result. Compared with actual sent volume ;

[0036] The processor: coordinates and schedules all modules, and executes the flow control method of any one of claims 1 to 11 in a time-slot cycle.

[0037] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways as described in any one of claims 1 to 11.

[0038] Compared with the prior art, the beneficial effects of the present invention are:

[0039] (1) Dynamically adjust the token generation rate so that the logic injection budget changes with the time-varying service capability of the target forwarding link, thereby reducing cache backlog, invalid injection and overflow packet loss caused by high-speed injection and low-speed output from the source;

[0040] (2) By using real-time transmission gating, blind transmission is avoided during receive busy or protection busy periods, thereby reducing the risk of half-duplex local collisions, receive blocking and uplink packet blocking.

[0041] (3) The data queue, virtual token bucket, link service capabilities and interface busy / idle status are jointly modeled to achieve lightweight online control without strict synchronization of the entire network, which is suitable for resource-constrained gateway buoys and multi-mode relay nodes;

[0042] (4) The token bucket capacity limit can limit the length of a single burst, prevent the target link from being occupied for a long time, reserve the necessary access opportunities for reverse or uplink services, and thus improve the fairness of bidirectional services. Attached Figure Description

[0043] Figure 1This is a schematic diagram of the overall architecture of the cross-layer VC-TB flow control method of the present invention;

[0044] Figure 2 This is a schematic diagram of gateway-side rate mismatch and interface busy / idle constraints in this invention;

[0045] Figure 3 This is a general framework diagram of the cross-layer VC-TB flow control method of the present invention;

[0046] Figure 4 This is a schematic diagram illustrating an application scenario of the acoustic-electric cross-medium gateway buoy embodiment of the present invention;

[0047] Figure 5 This is a mapping diagram of the channel quality parameters and the token generation adjustment factor in this invention;

[0048] Figure 6 This is a schematic diagram of the state transition of the VC-TB flow control method in this invention;

[0049] Figure 7 This is a flowchart illustrating the execution of the cross-layer VC-TB flow control method of the present invention.

[0050] Figure 8 This is a simulation topology diagram of a typical embodiment of an acoustic-electric cross-medium network;

[0051] Figure 9 This is a verification diagram of the channel state awareness and rate adaptation mechanism in this invention;

[0052] Figure 10 This is a comparison chart of the throughput performance of the present invention and other methods under different node scales and uplink / downlink service directions.

[0053] Figure 11 This is a comparison chart of packet loss rate performance between the present invention and other methods under different node scales and uplink / downlink service directions;

[0054] Figure 12 This is a comparison chart of the end-to-end latency performance of the present invention and other methods under different node scales and uplink / downlink traffic directions;

[0055] Figure 13 This is a comparison chart of the fairness performance of the present invention and other methods under different node scales. Detailed Implementation

[0056] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below through examples.

[0057] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0058] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0059] To verify the applicability of the cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways in real-world cross-media heterogeneous communication gateway scenarios, such as... Figure 1 , 4 As shown, this embodiment constructs an acoustic-electric cross-medium cooperative communication network consisting of airborne control nodes, surface gateway buoys, and underwater acoustic nodes. This network includes an underwater sensing layer, a cross-medium relay layer, and an airborne or surface control layer. The underwater sensing layer, composed of underwater nodes or cluster heads, is used to upload environmental monitoring data; the cross-medium relay layer, composed of surface gateway buoys, is used to perform buffering, protocol conversion, and bidirectional traffic shaping; the airborne or surface control layer, composed of UAVs, unmanned surface vessels, or a control center, is used to receive uplink monitoring data and issue control commands.

[0060] In this scenario, the radio side has high-speed injection capability, while the underwater acoustic side exhibits low speed, long latency, and strong time-varying characteristics. Furthermore, the gateway's underwater acoustic interface is subject to half-duplex transmit / receive mutual exclusion constraints. Therefore, this embodiment needs to simultaneously address capacity domain rate mismatch, interface time domain conflicts, and bidirectional service fairness issues.

[0061] Table 1 below shows the simulation environment of the constructed acoustic-electric cross-medium cooperative communication network (simulation topology diagram as shown). Figure 8 This document outlines key configurations for the link and VC-TB, provides simulation boundary conditions to ensure the reproducibility of the implementation examples, and determines core experimental parameters such as underwater acoustics, radio frequency, node motion, time slots, and thresholds. Figures 9-13 The data sources for the performance curves are listed in Table 2. Table 2 supplements the implicit settings such as simulation platform, event triggering, and statistical methods, unifying the simulation environment of this embodiment with other comparative algorithms to ensure fair comparison; it also clarifies the rules for token measurement, uplink events, and packet loss statistics, eliminating experimental ambiguity and enhancing experimental credibility. Table 3 lists five existing mainstream flow control algorithms—STB, PQ, WRR, ARED, and IoT-MAC—to determine the performance benchmarks. Figures 10-13The performance curves serve as a benchmark for comparison. Table 3 also briefly outlines the core mechanisms and inherent limitations of each algorithm, theoretically demonstrating that traditional solutions cannot simultaneously address rate mismatch and half-duplex blocking. Table 3 specifies that all comparative algorithms and this embodiment use the same simulation environment and the same token unit of measurement to eliminate experimental condition biases, enhance the authority of the experimental data, and support creative innovation.

[0062] Table 1 Simulation Scenarios and Protocol Parameters for Acoustic-Electrical Cross-Medium Networks

[0063] Table 2 Supplementary Simulation Implementation and Reproduction Settings

[0064] Table 3 Comparison Methods and Key Configurations

[0065] The flow control method of this embodiment is described in detail below:

[0066] like Figure 3 , 5 As shown in Figure 7, the cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways in this embodiment specifically includes the following steps:

[0067] S1. Establish a data queue and virtual token bucket oriented towards the target forwarding direction in the cross-media heterogeneous communication gateway. Assume... Indicates time slot The traffic volume reaching the gateway via the first communication link interface. Indicates time slot The initial length of the data queue waiting to be forwarded via the second communication link interface. Indicates time slot The actual service volume completed by the second communication link interface within the inner channel. This indicates the maximum capacity of the gateway cache.

[0068] The data queue is updated according to the following formula:

[0069]

[0070] The packet loss due to limited cache capacity is expressed as:

[0071]

[0072] S2. Configure the virtual token bucket. For time slots Initial virtual token inventory, This is the maximum capacity of the token bucket. For time slots The token generation rate within the timeframe. The current available sending budget after token generation is complete is:

[0073]

[0074] If the actual transmission volume of the current time slot is Then the token status will be updated to:

[0075]

[0076] One token corresponds to one standardized packet, one standardized data frame, or one standardized transmission budget. Unlike the traditional static token bucket, the token generation rate in this embodiment is not a fixed constant, but is dynamically determined by the channel quality parameters of the target forwarding link.

[0077] S3. Perform channel-aware impedance matching. The gateway performs this matching based on the channel quality parameters of the second communication medium link. Determine the token generation adjustment factor The current token generation rate is calculated using the following formula:

[0078]

[0079] In this embodiment of the underwater acoustic link, the token generation adjustment factor Step-by-step mapping is used:

[0080]

[0081]

[0082]

[0083]

[0084] in, To ensure survival in deep fading, a threshold of 5 dB is preferred. The physical layer demodulation threshold is preferably set to 10 dB; The channel segmentation step size is preferably 5 dB. 0.1 corresponds to deep fading keep-alive mode, 0.5 corresponds to weak coverage rate limiting mode, 1.0 corresponds to nominal operating mode, and 1.5 corresponds to opportunistic transmission mode. This segmentation setting is used to suppress invalid injection under low-quality links and release backlogged traffic in high-quality channel windows.

[0085] S4. Establish the busy / idle state of the interface. (Setting...) Indicates time slot Initially, the second communication link interface will continue to be occupied by the remaining time slots of the receive window, guard window, or other occupancy events, and the busy / idle indicator of the current time slot will be obtained by the following formula:

[0086]

[0087] when When =1, the second communication link interface is in a receive busy or protection busy state, and sending is not allowed; when When =0, the second communication link interface is in a local idle state and can proceed to the transmission decision. The remaining state of the receiver can be updated according to the following formula:

[0088]

[0089] in, This indicates the average number of time slots used for one uplink packet reception. Indicates time slot A new uplink receive event was detected within the device.

[0090] S5. Calculate the effective service capacity of the target forwarding link. Effective service capacity is determined by the link's effective rate, interface busy / idle status, and random service efficiency.

[0091]

[0092]

[0093] Where B is the effective bandwidth and η is the overall efficiency coefficient. For random service efficiency factor, The time slot length, To standardize the group length. Because Contains (1- This option sets the effective service capacity to zero when the interface is busy, thus protecting the receiving window.

[0094] S6. Execute immediate send gating. If , , and Then set the sending decision variable. And execute the send; otherwise set The gateway remains silent. The actual transmission volume is represented as follows:

[0095]

[0096] S7. Complete status write-back. If =1, then the gateway sends according to the actual amount. Deduct from data queues and virtual token buckets; if If the value is 0, the current time slot will remain silent and continue accumulating tokens. Update subsequently. , and Then, it enters the next time slot.

[0097] This embodiment can be further implemented by a cross-layer virtual clock-token bucket flow control system. This system includes a first communication link interface, a second link communication interface, a data queue management module, a channel state awareness module, a busy / idle detection module, a virtual token bucket module, a channel-aware impedance matching module, an instantaneous transmission gating module, and a processor.

[0098] The first communication link interface connects to the upstream first communication medium link to collect real-time service data from the ingress point. And send it to the queue;

[0099] The second communication link interface connects to the downstream second communication medium link to complete the actual transmission and reception of data packets and to provide feedback on interface occupancy and original channel measurement information.

[0100] Data queue management module: Maintains data queues ,according to The formula updates the queue length and counts packet loss due to buffer overflow.

[0101] Channel State Aware Module: Collects channel quality parameters from the physical / link layer across layers. Event-driven + zero-order hold-up channel parameter updates;

[0102] Busy / Idle Detection Module: Statistics on uplink received time slots occupied and calculation of remaining occupied time. Output interface busy / idle indicator ;

[0103] Virtual token bucket module: based on Update token inventory ,by Constraints, output available for sending budget;

[0104] Channel-aware impedance matching module: based on Table lookup generation Calculate the real-time token generation rate;

[0105] The real-time sending gating module uses four conditions—joint queue, token, link service capacity, and interface busy / idle—to make a sending decision and output the result. Compared with actual sent volume ;

[0106] Processor: Coordinates and schedules all modules, executing steps S1 to S7 of this embodiment in a time-slot-based loop. The processor's scheduling of the above modules to execute the above program can also be stored in a computer-readable storage medium and run in gateway buoys, underwater relay nodes, unmanned surface vessels, drones, shore-based gateways, or multi-mode communication relay nodes.

[0107] The VC-TB flow control method in this embodiment can also be interpreted from the perspective of stochastic control. Let the system state vector be... =[ , , ],in This indicates a backlog in the data queue in the target forwarding direction. Indicates the virtual token bucket inventory. This indicates the residual state of the receive access on the second communication link interface. To characterize the protection effect on the uplink receive window, the uplink protection indicator is defined as:

[0108]

[0109] In long-term operation, the gateway control objective is to achieve a balance between increasing effective downlink forwarding volume, protecting the uplink receive window, and suppressing drastic fluctuations in the token generation rate. This objective can be expressed as:

[0110]

[0111] The above objectives are subject to ∈{0,1}、 ≤1- , ≤ ≤ Queue capacity limit and the token bucket capacity limit Common constraints. To illustrate the stability heuristic of rule design, we define a quadratic Lyapunov function:

[0112]

[0113] The corresponding one-step conditional drift is:

[0114]

[0115] Combining the relationship between data queue updates and virtual token bucket updates, and utilizing the relationship between any non-negative variable... , , Inequalities that hold true for all The upper bound of the drift term can be obtained as follows:

[0116]

[0117] Where B is a finite constant determined by the maximum arrivals, maximum service volume, and maximum token generation rate per time slot. This upper bound indicates that when When the volume is large, the effective transmission volume should be increased as much as possible. To reduce cache backlog; when When the value is large over a long period, it should be reduced. To avoid the accumulation of invalid tokens; when When =1, downlink transmission should be suppressed to protect the uplink reception window. Therefore, this embodiment uses the channel-aware impedance matching approximation to achieve capacity domain injection regulation and the instantaneous transmission gating approximation to achieve time domain transmission timing selection.

[0118] like Figure 2 As shown in the figure, the left and right sides are a comparison between traditional fixed flow control and the VC-TB dynamic flow control of this embodiment:

[0119] Traditional methods: Impedance mismatch between the source and load links, fixed tokens cannot adapt to time-varying underwater acoustic channels, and queue congestion is likely to occur. The method in this embodiment: Impedance matching is achieved by relying on SINR cross-layer feedback and dynamic injection, which solves the rate mismatch and half-duplex transmit / receive conflict from both the capacity domain and the time domain.

[0120] To comprehensively compare and evaluate the performance of the VC-TB flow control method in this embodiment with existing flow control or queue scheduling methods in cross-media heterogeneous communication gateway scenarios, this embodiment uses quantitative analysis based on indicators such as total network throughput, uplink throughput, downlink throughput, total packet loss rate, uplink packet loss rate, downlink packet loss rate, average end-to-end latency, and Jain fairness index. Figures 10 to 12 This is used to demonstrate the changes in throughput, packet loss rate, and end-to-end latency between VC-TB and the comparison method under different node sizes and uplink / downlink business directions. Figure 13 Used to demonstrate changes in fairness across different node sizes.

[0121] (1) Network throughput: The total network throughput is defined as the sum of the net uplink and downlink load bits that successfully reach the destination within the observation period, divided by the total simulation duration; uplink throughput and downlink throughput are counted separately for their respective directions. This metric is used to measure the effective carrying capacity of a cross-media gateway for bidirectional heterogeneous services.

[0122] like Figure 10 As shown, when the number of nodes is 6, the total network throughput of VC-TB is approximately 0.30 kbps, which is about 2.5 times that of Fixed Token Bucket (STB), 3.4 times that of Weighted Round Robin (WRR), and 5.5 times that of Priority Queue (PQ). This result demonstrates that the synergistic effect of capacity domain adjustment and time domain gating can improve the bidirectional effective carrying capacity of cross-media gateways.

[0123] (2) Packet loss rate: The packet loss rate is defined as the ratio of the number of failed packets caused by buffer overflow, underlying access conflict, half-duplex blocking or timeout failure to the number of generated packets. VC-TB can reduce invalid injection and local half-duplex conflict by reducing the token generation rate when the channel deteriorates and shutting downlink transmission during busy reception periods. Therefore, it can reduce the total packet loss rate and maintain a low packet loss level on the downlink service side.

[0124] (3) End-to-end latency: End-to-end latency is defined as the average time between the generation of a packet at the service layer and its successful reception by the target end. For example... Figure 12 As shown, when the number of nodes is 6, the average end-to-end latency of VC-TB is about 290 ms, which is about 28% lower than that of Fixed Token Bucket (STB), about 41% lower than that of Weighted Round Robin (WRR), and about 59% lower than that of Priority Queue (PQ). This result shows that gateway cache backlog and interface conflicts are suppressed through capacity domain adjustment and time domain gating.

[0125] (4) Jain Fairness Index: The Jain Fairness Index is used to measure the balance of services between two-way traffic or nodes. Its expression is:

[0126] in, This represents the number of successful services obtained by the i-th service flow or node, where n represents the number of service flows or nodes. The closer this metric is to 1, the fairer the resource allocation. Compared to priority queues, VC-TB does not simply prioritize downlink services, but rather protects uplink services and limits downlink burst length during busy reception periods, thus improving the fairness of uplink and downlink services.

[0127] Table 4 Supplementary results of the overall performance of the VC-TB method of this invention

[0128] The above supplementary results demonstrate that, compared to a fixed-rate token bucket, the method in this embodiment can adjust the token generation strength according to changes in link status; compared to a priority queue, it can avoid long-term suppression of uplink reception by downlink services; and compared to static weighted polling, it can perform online adaptive switching based on current channel quality and reception busy / idle status. Therefore, this invention not only improves throughput or latency in a single direction but is also more suitable for application scenarios in cross-media gateways involving bidirectional concurrent services, time-varying link service capabilities, and limited interface transmission and reception.

[0129] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A cross-layer VC-TB flow control method for heterogeneous communication gateways across media, applied to heterogeneous communication gateways connecting a first communication medium link and a second communication medium link, wherein the two links differ in at least one parameter among transmission rate, propagation delay, channel quality, effective service capability, and transmit / receive constraints; characterized in that, The method includes: S1. Establish a data queue and virtual token bucket oriented towards the target forwarding direction in the cross-media heterogeneous communication gateway, and pre-configure: queue capacity limit. , Token bucket capacity limit Benchmark token generation rate Deep decay survival threshold Physical layer demodulation threshold Channel hierarchical step size and time slot length ; S2, in each discrete time slot The system retrieves cross-layer status information, which includes the traffic volume arriving via the first communication link interface. Length of the data queue to be forwarded via the second communication link interface Virtual token inventory Channel quality parameters of the second communication medium link and the busy / idle status of the second communication link interface. ; S3. Based on the channel quality parameters of the second communication medium link. Perform channel-aware impedance matching to obtain the token generation adjustment factor. and in accordance with Determine the token generation rate for the current time slot to match the logic injection budget with the current service capacity of the second communication medium link; S4, according to Update the virtual token bucket to obtain the available transmission budget for the current time slot. The The theoretical total number of tokens added in this time slot to the existing virtual token bucket is the upper limit of the virtual token bucket's capacity. Used to limit the length of a single burst of transmission, prevent the target forwarding link from being occupied for a long time, and reserve reception and access opportunities for reverse services, uplink services or other priority services; S5. Determine whether to allow transmission via the second communication link interface based on the instant transmission gating conditions. The instant transmission gating conditions include at least the following: the data queue is not empty, the available transmission budget is sufficient, the second communication medium link has effective service capabilities, and the second communication link interface is not in a receive busy or protection busy state. S6. When the instantaneous transmission gating condition is met, transmission is performed via the second communication link interface, and the transmission is based on the data queue length. Available sending budget Second communication medium link effective service capability Determine the actual amount sent If the instantaneous sending gating condition is not met, remain silent and continue accumulating tokens; S7. Based on the actual sending volume Update the data queue, virtual token bucket, and busy / idle status of the second communication link interface, and enter the next discrete time slot, thereby realizing capacity domain matching and time domain gating of the receiving window between heterogeneous links across media.

2. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 1, characterized in that, The cross-medium heterogeneous communication gateway includes one or more of the following: gateway buoy, unmanned surface vessel gateway, underwater relay gateway, shore-based gateway, aerial unmanned aerial vehicle gateway, or multi-mode communication relay node; the first communication medium link and the second communication medium link include any two or more combinations of underwater acoustic link, optical communication link, radio link, electromagnetic link, wired communication link, and satellite link.

3. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 2, characterized in that, The method is applicable to acoustic-electric cross-medium communication networks, acoustic-optical cross-medium communication networks, optoelectronic cross-medium communication networks, and acoustic-optical-electrical integrated cross-medium communication networks.

4. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 1, characterized in that: The cross-layer status information in step S2 is obtained from the vertical interaction interface between the data link layer and the physical layer. The physical layer feeds back to the data link layer at least one or more of the following: channel quality parameters, carrier sensing results, receive occupancy status, link synchronization status, packet decoding status, received optical power, bit error rate, or packet power.

5. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 4, characterized in that, The channel quality parameters This includes one or more of the following: signal-to-noise ratio (SNR), signal-to-interference-plus-noise ratio (SINR), bit error rate (BER), packet error rate (PER), received signal strength indication (RSSI), received optical power, link quality indication (LQI), component power, estimated channel capacity, or estimated effective service rate.

6. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 1, characterized in that, The channel-aware impedance matching in step S3 specifically includes: matching the channel quality parameters of the second communication medium link. Mapping to discrete or continuous token generation adjustment factor When the channel quality is below the deep fading keep-alive threshold, the token generation rate is reduced to the keep-alive level; when the channel quality is in a weak coverage area, the token generation rate is lower than the baseline token generation rate; when the channel quality is in the nominal working area, the baseline token generation rate is used; when the channel quality is in the opportunistic transmission area, the token generation rate is increased.

7. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 1, characterized in that, Step S2 uses a combination of event-driven update and zero-order hold to obtain channel quality parameters. When the second communication link interface is successfully synchronized and receives uplink packets, probe frames, or feedback frames, the corresponding channel quality parameters are extracted and the channel quality record is updated; when no new valid measurement value is detected, the most recent valid measurement value is used as the channel quality estimate for the current time slot.

8. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 1, characterized in that, The second communication medium link effective service capability in step S5 Determined based on the effective link rate, the busy / idle status of the second communication link interface, and random service efficiency, satisfying... ,in, To receive busy / idle indicators, For random service efficiency factor, The time slot length, To standardize the group length, The effective rate of the second communication medium link is determined by the channel quality parameters.

9. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 8, characterized in that, The busy / idle state of the second communication link interface in step S5 is determined by the remaining state of the receiving / occupied state. Sure, =1 indicates that the second communication link interface is in a busy receiving or busy protection state. =0 indicates that the second communication link interface is in a local idle state; The remaining state of the received occupancy is according to Update, in which This indicates the average number of time slots used for one uplink packet reception. Indicates time slot A new uplink receive event was detected within the device.

10. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 9, characterized in that, Sending decision variables in step S6 Satisfy: When , , and hour, ;otherwise The actual transmission volume meets the requirements. .

11. The cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways according to claim 1, characterized in that, In step S7, the data queue is arranged according to... Update: The amount of packet loss due to overflow caused by limited queue capacity is expressed as follows: .

12. A cross-layer VC-TB flow control system for cross-media heterogeneous communication gateways, characterized in that, It includes a first communication link interface, a second communication link interface, a data queue management module, a channel state awareness module, a busy / idle detection module, a virtual token bucket module, a channel awareness impedance matching module, an instant transmission gating module, and a processor; The first communication link interface connects to the upstream first communication medium link to collect real-time service data from the entry point. And send it to the queue; The second communication link interface connects to the downstream second communication medium link to complete the actual transmission and reception of data packets and to provide feedback on interface occupancy and original channel measurement information. The data queue management module: maintains the data queue. ,according to The formula updates the queue length and counts packet loss due to buffer overflow. The channel state awareness module collects channel quality parameters across layers from the physical / link layer. Event-driven + zero-order hold-up channel parameter updates; The busy / idle detection module: counts the uplink received time slots occupied and calculates the remaining occupied time. Output interface busy / idle indicator ; The virtual token bucket module: based on Update token inventory ,by Constraints, output available for sending budget; The channel sensing impedance matching module: based on Table lookup generation Calculate the real-time token generation rate; The instant transmission gating module makes a transmission decision based on four conditions: queue, token, link service capability, and interface busy / idle status, and outputs the result. Compared with actual sent volume ; The processor: coordinates and schedules all modules, and executes the flow control method of any one of claims 1 to 11 in a time-slot cycle.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the cross-layer VC-TB flow control method for cross-media heterogeneous communication gateways as described in any one of claims 1 to 11.

Images