Microwave link adaptive transmission method for edge coastal defense radar backhaul

Abstract

The application discloses a microwave link adaptive transmission method for coastal defense radar backhaul, and belongs to the technical field of digital information transmission, and comprises the following steps: collecting a throughput rate, a sending queuing time length and sea clutter energy; segmenting the sending queuing time length to obtain a fluctuation scale table and generate fluctuation phase quantities, and calculating a same direction occupancy rate to obtain a coupling level; generating a sending budget table according to the fluctuation scale table, the fluctuation phase quantities and the coupling level; cutting echo data according to a slice upper limit to obtain a slice sending list and a slice index table, and obtaining an alarm fast packet according to an alarm reserved time slot and associating the slice index table; and rotating a sending window according to the sending budget table, sending the alarm fast packet first, then sending the slice sending list, and limiting retransmission according to a retransmission limit. The application effectively solves the problem that the abnormal propagation of an evaporation waveguide causes the adaptive strategy to fail, and improves the transmission stability and transmission efficiency of coastal defense radar backhaul data on a microwave link.

Description

Technical Field

[0001] This application relates to an adaptive transmission method for microwave links for backhaul of radar for border and coastal defense, and belongs to the field of digital information transmission technology. Background Technology

[0002] In coastal defense detection systems, front-end radars in coastal areas typically transmit echo data, track data, and alarm data to the back-end via microwave links. Microwave links operate in the air-sea boundary layer environment for extended periods, and are affected by evaporation waveguides, tropospheric waveguides, and multipath propagation over the sea surface. As a result, the available capacity, transmission delay, and service queuing status of the link fluctuate continuously over time. At the same time, the energy of sea clutter generated by the front-end radar echoes also changes under abnormal propagation conditions, causing a coupling phenomenon between changes in the detection side and changes in the transmission side.

[0003] Existing link adaptive transmission methods primarily adjust transmission bit rate, redundancy, and retransmission strategies based on link metrics such as throughput, packet loss rate, and signal-to-noise ratio. While these methods can respond to general link fluctuations, they typically treat link status as an independent process, lacking continuous identification of transmission queuing time fluctuations, effective determination of the coupling between sea clutter energy changes and throughput changes, and unified constraints on the resource occupancy relationship of alarm data, echo data, track data, and retransmission data within the same transmission cycle. Therefore, in scenarios where abnormal propagation causes rapid fluctuations in link status, existing methods are prone to mismatches in update rhythm selection, transmission parameter adjustment, and retransmission occupancy control, allowing low-priority retransmission tasks or large data fragments to continuously occupy transmission resources.

[0004] Therefore, even if the link temporarily improves during local periods, high-priority alarm data may still fail to get a timely transmission opportunity, resulting in alarm information transmission delays. This affects the timely response of the backend command and control center to anomalies in the frontend radar and reduces the overall transmission efficiency and reliability of border and coastal defense radar backhaul services. Therefore, in border and coastal defense radar backhaul scenarios, how to accurately characterize and effectively constrain the link fluctuation process and transmission resource occupancy relationship under abnormal propagation conditions, thereby reducing head-end congestion and alarm transmission delays, has become an urgent technical problem to be solved. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this application aims to provide an adaptive transmission method for microwave links in coastal defense radar backhaul. This method involves synchronously collecting throughput, transmission queuing time, and sea clutter energy at each sampling moment to construct an undulation scale table, undulation phase quantity, and coupling level. Based on these parameters, a transmission budget table is generated. Differentiated transmission control is then applied to alarm data, echo data, and track data. This achieves the goals of reducing adaptive strategy mismatch under abnormal propagation conditions such as evaporative waveguides, mitigating head blocking, and improving the timeliness and efficiency of coastal defense radar backhaul data transmission.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] Firstly, this application provides an adaptive transmission method for microwave links for backhaul of border and coastal defense radar, including:

[0008] At each sampling moment, the throughput, transmission queue time, and sea clutter energy are collected.

[0009] The transmission queuing time is segmented to obtain an fluctuation scale table. The fluctuation phase quantity is generated according to the direction of the fluctuation segment in the fluctuation scale table. The coupling level is obtained by calculating the co-directional occupancy rate of sea clutter energy and throughput.

[0010] Based on the fluctuation scale table, fluctuation phase quantity and coupling level, the first update cycle, second update cycle, alarm retention time slot, retransmission limit, fragmentation limit, first bit rate level and first redundancy are determined to obtain the transmission budget table.

[0011] Based on the transmission budget table, the collected echo data is divided into segments according to the upper limit of the segments to obtain a segment transmission list and generate a segment index table. The segment transmission list contains the collected track data. The collected alarm data is processed according to the alarm retention time slot to obtain alarm fast packets, and the alarm fast packets are associated with the segment index table.

[0012] The first bitrate and first redundancy are updated according to the first update cycle. The transmission window is rotated according to the second update cycle. First, alarm fast packets are sent, then echo data and track data are sent according to the fragmented transmission list, and the return data stream is obtained by supplementing the transmission limit according to the supplementary transmission limit.

[0013] Furthermore, the throughput is taken as the sliding window throughput, which is obtained by the transmission count of the front-end microwave link device;

[0014] Set the first window length, and when collecting throughput, record the amount of valid data sent within the first window length to form a throughput record item. The throughput record item contains a timestamp and is arranged in the order of the timestamp.

[0015] By employing a sliding window throughput and throughput record, throughput can participate in subsequent direction marker generation and fluctuation analysis under a unified time caliber, reducing the impact of occasional fluctuations at a single sampling moment on link status judgment.

[0016] Furthermore, methods for obtaining the fluctuation scale table include:

[0017] The queuing time for sending is segmented to obtain fluctuation segments;

[0018] For each fluctuation segment, record the start time and end time, as well as the direction and duration of the fluctuation segment. The duration of the fluctuation segment is determined by the start time and end time.

[0019] Extract the representative value of throughput and the representative value of sending queue time within the current fluctuation segment. The representative value of throughput is taken from the middle position of the throughput record in the current fluctuation segment, and the representative value of sending queue time is taken from the middle position of the sending queue time in the current fluctuation segment.

[0020] The fluctuation scale table is arranged in chronological order of fluctuation segments. Each item in the fluctuation scale table includes a start time marker, an end time marker, the direction of the fluctuation segment, the duration of the fluctuation segment, a representative value of throughput, and a representative value of transmission queuing time.

[0021] By constructing a fluctuation scale table based on the transmission queue duration and generating fluctuation phase quantities, the fluctuations in link capacity and the changes in transmission queue backlog can be unified into the same time structure, providing a direct basis for determining the subsequent update cycle and transmission budget table.

[0022] Furthermore, methods for obtaining fluctuation segments include:

[0023] Traverse the direction markers of the transmission queue duration in the order of the time markers. Record the sampling time when the direction marker of the transmission queue duration is in the upward direction as the rising segment, and record the sampling time when the direction marker of the transmission queue duration is in the downward direction as the falling segment. Summarize the rising segment and the falling segment to obtain the fluctuation segment.

[0024] Furthermore, methods for obtaining fluctuation segments also include:

[0025] Set a lower limit for the segment length. If the length of the rising segment is less than the lower limit, the rising segment will be merged into the adjacent rising segment on the left or the adjacent rising segment on the right with the same direction. If the directions of the adjacent rising segment on the left and the adjacent rising segment on the right are not the same as the rising segment, the lengths of the adjacent rising segment on the left and the adjacent rising segment on the right will be compared, and the rising segment will be merged into the adjacent rising segment with the longer length, resulting in the merged rising segment.

[0026] Furthermore, methods for generating fluctuation phase quantities include:

[0027] Select the most recent fluctuation segment from the fluctuation scale table, and take the direction of the most recent fluctuation segment as the current direction;

[0028] When the current direction is upward, the fluctuation phase is measured as the upward phase; when the current direction is downward, the fluctuation phase is measured as the downward phase.

[0029] Set a boundary duration. When the time interval between the current sampling time and the end time marker of the most recent fluctuation segment is less than or equal to the boundary duration, the fluctuation phase quantity remains the phase corresponding to the fluctuation direction of the most recent fluctuation segment.

[0030] After determining the current direction, the fluctuation phase quantity is obtained.

[0031] Furthermore, the same-direction occupancy rate is determined by the occupancy relationship between the number of same-direction attempts and the number of valid attempts.

[0032] Furthermore, the methods for obtaining the number of times in the same direction and the number of valid times include:

[0033] Set a second window length, and count both the number of times in the same direction and the number of valid times within the time marker covered by the second window length;

[0034] The counting method for valid counts is as follows: when both the probe observation sequence and the link observation sequence are valid at each pair of adjacent sampling times, it is counted as 1 valid count;

[0035] The counting method for the same direction count is as follows: at each pair of adjacent sampling times, if the direction marker of sea clutter energy and the direction marker of throughput are both in the upward direction or both in the downward direction, it is counted as 1 same direction count. If either direction marker is in the same direction, the adjacent sampling times are not counted as same direction count.

[0036] Furthermore, methods for generating directional markers for sea clutter energy include:

[0037] Set the threshold for sea clutter energy variation;

[0038] When the difference between the sea clutter energy at the next sampling time and the sea clutter energy at the previous sampling time is greater than or equal to the sea clutter energy change threshold, the direction of the sea clutter energy is marked in the upward direction.

[0039] When the difference between the sea clutter energy at the previous sampling time and the sea clutter energy at the next sampling time is greater than or equal to the sea clutter energy change threshold, the direction of the sea clutter energy is marked by the falling direction.

[0040] Otherwise, the direction of sea clutter energy is marked in the same direction.

[0041] Furthermore, methods for generating throughput direction markers include:

[0042] Set a threshold for throughput changes;

[0043] The throughput direction marker is generated using the same method as the sea clutter energy direction marker, except that the sea clutter energy change threshold is replaced with the throughput change threshold.

[0044] By adopting a consistent direction marking generation rule for sea clutter energy and throughput, and by using sea clutter energy change thresholds and throughput change thresholds to exclude direction reversals caused by small fluctuations, the consistency of the determination of the coupling relationship between the detection side and the transmission side can be improved.

[0045] Furthermore, the methods for obtaining the link observation sequence and the probe observation sequence include:

[0046] A link observation sequence is formed based on throughput records and transmission queuing time, and a detection observation sequence is obtained based on sea clutter energy.

[0047] Furthermore, methods for obtaining the coupling level include:

[0048] Set a first percentage threshold and a second percentage threshold;

[0049] When the occupancy rate in the same direction is greater than or equal to the second percentage threshold, the coupling level is set to the third level.

[0050] When the occupancy rate in the same direction is less than the second percentage threshold and greater than or equal to the first percentage threshold, the coupling level is the second level.

[0051] When the occupancy rate in the same direction is less than the first percentage threshold, the coupling level is set to the first level.

[0052] By statistically analyzing the occupancy rate within the coverage area of ​​the second window and mapping it to obtain the coupling level, the synchronous direction changes of sea clutter energy and throughput can be transformed into directly callable hierarchical results, which facilitates the subsequent sending of budget tables to execute differentiated configurations.

[0053] Compared with the prior art, the beneficial effects achieved by this application are as follows: This application synchronously collects throughput, transmission queuing time and sea clutter energy at each sampling moment, and forms an fluctuation scale table through the transmission queuing time, and forms a co-directional occupancy rate and coupling level through sea clutter energy and throughput. Then, based on the fluctuation scale table, fluctuation phase quantity and coupling level, it determines the first update cycle, the second update cycle, alarm retention time slot, retransmission limit, fragmentation limit, first code rate level and first redundancy to obtain a transmission budget table. On this basis, according to the transmission budget table, it performs differentiated organization and transmission control on echo data, track data and alarm data, so that alarm data enters the transmission window first, and echo data and track data are transmitted under the constraints of fragmentation limit and retransmission limit.

[0054] Therefore, the link-side fluctuation characteristics and the detection-side propagation characteristics can be jointly introduced into the transmission control process to reduce the probability of adaptive strategy mismatch, follow-up lag and head blocking under abnormal propagation conditions such as evaporation waveguides, thereby improving the timeliness and efficiency of the transmission of data back to the border and coastal defense radar. Attached Figure Description

[0055] Figure 1 A schematic diagram of an adaptive transmission method for microwave links for backhaul of radar data from border and coastal defense.

[0056] Figure 2 This is a flowchart of the method for obtaining the fluctuation scale table in this application;

[0057] Figure 3 This application generates a logic diagram for the generation of directional markers for sea clutter energy and throughput, and for determining the coupling level. Detailed Implementation

[0058] The technical solutions of this application will be described in detail, clearly, and completely below with reference to the accompanying drawings of the embodiments. It should be particularly noted that the specific embodiments described below are only used to better illustrate and explain the technical solutions of this application, and are intended to enable those skilled in the art to better understand and implement this application, and should not be construed as limiting the scope of protection of this application. Without departing from the spirit and substance of this application, those skilled in the art can modify, adjust, or make equivalent substitutions based on the content disclosed in this application, and these modifications, adjustments, or equivalent substitutions should all be considered within the scope of protection of this application.

[0059] Example 1

[0060] Please see Figures 1-3 As shown, this embodiment discloses an adaptive transmission method for microwave links for backhaul of border and coastal defense radar, including:

[0061] At each sampling moment, throughput, transmission queuing time, and sea clutter energy are collected, specifically including:

[0062] A link observation sequence is formed based on throughput records and transmission queuing time, and a detection observation sequence is obtained based on sea clutter energy.

[0063] Set a first sampling period, which is used to constrain the acquisition period of throughput, transmission queuing time and sea clutter energy; generate sampling times sequentially according to the first sampling period, and record a time stamp at each sampling time.

[0064] The throughput is collected at each sampling moment. The throughput is taken as the sliding window throughput, which is obtained by the transmission count of the front-end microwave link equipment. The first window length is set according to the first sampling period so that the sliding window throughput can reflect the fluctuation of the available capacity of the link under the evaporating waveguide condition, and avoid the impact of occasional packet loss at a single sampling moment on the throughput. During the acquisition, the amount of data transmitted within the first window length and the corresponding first window length are recorded at each sampling moment to form a throughput record. The throughput record is arranged in the order of time stamp.

[0065] When forming the link observation sequence, the packet loss rate is synchronously collected at each sampling moment and a packet loss rate record is formed. The packet loss rate record is the occupancy relationship between the amount of data that failed to be sent and did not receive an acknowledgment within the first window length and the amount of data that was sent within the first window length. The packet loss rate record retains a time stamp and corresponds to the throughput record and the transmission queuing time using the same time stamp. When the packet loss rate record is missing, the available status at the current sampling moment is invalidated. Each sequence item of the link observation sequence includes throughput, transmission queuing time, packet loss rate and available status.

[0066] The transmission queuing time is collected at each sampling moment. The transmission queuing time is determined by the queuing time of the data entries leaving the transmission queue in the first sampling period. In the first implementation, for each data entry leaving the transmission queue in the first sampling period, the enqueue time and the dequeue time are recorded. The queuing time of the data entry is obtained by subtracting the enqueue time from the dequeue time. All queuing times in the current first sampling period are arranged according to their numerical values. The value corresponding to the middle position is taken as the transmission queuing time at the current sampling moment, and a transmission queuing time with a time stamp is formed.

[0067] When the enqueue and dequeue times cannot be directly obtained, a second implementation method is used to collect the transmission queuing time. In the second implementation method, at each sampling time, the transmission queue length and dequeue rate are collected from the front-end microwave link equipment. The transmission queue length is the amount of data waiting to be transmitted in the transmission queue, in bytes. The dequeue rate is the amount of data actually transmitted in the first sampling period divided by the duration of the first sampling period, in bytes per second. The transmission queue delay is estimated based on the transmission queue length and dequeue rate, and used as the transmission queuing time for each sampling time. To reduce the impact of short-term fluctuations in the dequeue rate on the estimation results, a second smoothing process is performed on the dequeue rate before estimation. The second smoothing process uses a fixed-length moving average method, and the sample values ​​of the moving average are limited to the sampling times when the available state is valid. A second smoothing length is set, and the fixed length is the second smoothing length. The second smoothing length is used to limit the number of consecutive sampling times participating in the moving average.

[0068] When the transmission queue length and dequeue rate cannot be directly obtained, a third implementation method is used to collect the transmission queue duration. In the third implementation method, the transmission queue delay statistics are collected from the front-end microwave link device at each sampling time. The transmission queue delay statistics are the statistical results of the transmission queue delay output by the front-end microwave link device in the first sampling period. The statistical results include at least one of the mean or quantile values. When both the mean and quantile values ​​exist, the quantile value is preferred as the transmission queue duration at each sampling time. When only the mean is output, the mean value is selected as the transmission queue duration at each sampling time.

[0069] By using any of the above implementation methods, the cumulative effect of link-side congestion and retransmission backlog can be characterized by the transmission queuing time. This can reflect the changes in transmission queue occupancy in the link capacity fluctuations caused by the evaporation waveguide. All three implementation methods rely on common statistical or counting information provided by the front-end microwave link equipment. They do not require support for recording the entry and exit times of each data item, thereby reducing the risk that the acquisition of transmission queuing time may not be possible due to equipment limitations, and reducing the probability of low-priority retransmission occupying the transmission window for a long time and causing head blocking.

[0070] A link observation sequence is formed based on throughput records and transmission queuing time. The link observation sequence is indexed by time stamps. Each item in the link observation sequence contains the throughput and transmission queuing time at the same sampling time. To avoid direction reversal caused by missing throughput records or missing transmission queuing time in the generation of subsequent direction stamps, an available state is added to each item in the link observation sequence. The available state is used to indicate whether the throughput and transmission queuing time at the current sampling time are valid collected values. When both the throughput record and the transmission queuing time exist, the available state is valid. When either record is missing, the available state is invalid.

[0071] If a throughput record or a transmission queue duration is missing at the current sampling time, the most recent valid value is retained for the current sampling time, and the available state is set to invalid. The method for retaining the most recent valid value is to backtrack from the current sampling time, select the most recent available state as the valid sampling time, and assign the throughput and transmission queue duration to the current sampling time respectively. To avoid the most recent valid value retention span being too long due to consecutive missing values, a backtracking upper limit is set, which is set based on the first sampling period. If no valid collected value is found after the backtracking step count exceeds the backtracking upper limit, the current sampling time is kept in the missing state, and the available state is kept as invalid.

[0072] When generating direction markers for transmission queuing time, the system is based on adjacent sampling time pairs with valid available states. Direction markers are generated only for adjacent sampling time pairs. If any sampling time is invalid, no direction marker is generated for the adjacent sampling time pairs. The same method is used when generating direction markers for throughput.

[0073] Sea clutter energy is collected based on time stamps to form a detection and observation sequence. The sea clutter energy is calculated from the echo data of the front-end radar. A sea surface interval is set to limit the sampling range of sea surface echoes in the echo data. At each sampling time, all echo sample values ​​within the sea surface interval are extracted. When the echo sample value is a real number, the square of the amplitude of the echo sample value is taken as the energy sample value. When the echo sample value contains both real and imaginary parts, the sum of the squares of the real and imaginary parts is taken as the energy sample value. The average of all energy sample values ​​is calculated to obtain the sea clutter energy at the current sampling time, and the sea clutter energy and time stamp are written into the sea clutter energy record.

[0074] The transmission queuing time is segmented to obtain an fluctuation scale table. Fluctuation phase quantities are generated based on the direction of the fluctuation segments in the fluctuation scale table. The coupling level is obtained by calculating the co-directional occupancy rate of sea clutter energy and throughput, specifically including:

[0075] To avoid small fluctuations in the transmission queue duration affecting the direction marker, for example, a change threshold is first set based on the changes in transmission queue duration corresponding to the 10 most recent adjacent sampling times. The changes in the 10 most recent adjacent sampling times are taken as the absolute value of the difference between the transmission queue duration of the previous sampling time and the transmission queue duration of the next sampling time. The changes in the 10 most recent adjacent sampling times are arranged in ascending order of value, and the change corresponding to the 7th position is taken as the initial value of the change threshold. When the change corresponding to the 7th position is less than 0.5 milliseconds, the change threshold is 0.5 milliseconds; when the change corresponding to the 7th position is greater than 10 milliseconds, the change threshold is 10 milliseconds; when the change corresponding to the 7th position is between 0.5 milliseconds and 10 milliseconds, the change threshold is the change corresponding to the 7th position. When the link observation sequence has less than 10 changes in adjacent sampling times before the current time, the changes in the existing adjacent sampling times are arranged in ascending order of value, and the change corresponding to the last position is taken as the change threshold, and boundary constraints are applied from 0.5 milliseconds to 10 milliseconds.

[0076] The transmission queue durations corresponding to the previous and next sampling times are compared sequentially. When the difference between the transmission queue duration of the next sampling time and the transmission queue duration of the previous sampling time is greater than or equal to the change threshold, the direction of the transmission queue duration is set to the upward direction; when the difference between the transmission queue duration of the previous sampling time and the transmission queue duration of the next sampling time is greater than or equal to the change threshold, the direction of the transmission queue duration is set to the downward direction; when both of the above differences are less than the change threshold, the direction of the transmission queue duration is set to the horizontal direction.

[0077] The transmission queuing time is segmented to obtain fluctuation segments. The direction markers of the transmission queuing time are traversed in the order of the time markers. The sampling time when the direction marker of the transmission queuing time is in the upward direction is recorded as the rising segment, and the sampling time when the direction marker of the transmission queuing time is in the downward direction is recorded as the falling segment. Both the rising segment and the falling segment are fluctuation segments. A lower limit for the segment length is set to avoid the fluctuation segment being too short, which would cause the fluctuation scale table to lose its representativeness. The lower limit for the segment length is set based on the first sampling period. If the length of the rising segment is less than the lower limit for the segment length, the rising segment is merged into the fluctuation segment with the same direction in the adjacent fluctuation segment on the left or right. If the directions of the adjacent fluctuation segments on the left and right are not consistent with the rising segment, the lengths of the adjacent fluctuation segments on the left and right are compared, and the rising segment is merged into the adjacent fluctuation segment with the longer length. The falling segment is processed in the same way.

[0078] Record the duration of fluctuation segments to form a fluctuation scale table; record the start and end timestamps for each fluctuation segment, as well as the direction and duration of the fluctuation segment; the direction of the fluctuation segment is determined based on the formation process of the fluctuation segment. When the fluctuation segment is formed by the sampling time when the direction mark of the sending queue duration maintains the upward direction, the direction of the fluctuation segment is the upward direction; when the fluctuation segment is formed by the sampling time when the direction mark of the sending queue duration maintains the downward direction, the direction of the fluctuation segment is the downward direction; when an upward or downward segment with a segment length less than the lower limit of the segment length is merged into an adjacent fluctuation segment, the direction of the fluctuation segment is the direction of the adjacent fluctuation segment after the merger; the duration of the fluctuation segment is determined by the number of the first sampling periods between the start and end timestamps.

[0079] To ensure that the fluctuation scale table simultaneously reflects both link capacity fluctuations and queue backlog fluctuations, while recording the duration of fluctuation segments, representative values ​​of throughput and transmission queue duration within the current fluctuation segment are extracted. The representative value of throughput is taken from the middle position of the throughput record within the current fluctuation segment, and the representative value of transmission queue duration is taken from the middle position of the transmission queue duration within the current fluctuation segment. The fluctuation scale table is arranged according to the order of fluctuation segments in the link observation sequence. Each entry includes a start time marker, an end time marker, the direction of the fluctuation segment, the duration of the fluctuation segment, the representative value of throughput, and the representative value of transmission queue duration. In this way, the fluctuation scale table uses transmission queue duration as the main line to characterize the temporal structure of link fluctuations, while retaining the quantization level of throughput. This reduces the short-term packet loss pull caused by relying solely on throughput and suppresses the repeated fluctuations of transmission rate and redundancy within adjacent control cycles.

[0080] The fluctuation phase quantity is generated based on the direction of the fluctuation segment in the fluctuation scale table. Specifically, the most recent fluctuation segment is selected from the fluctuation scale table, and the direction of the fluctuation segment corresponding to the most recent fluctuation segment is taken as the current direction. The fluctuation phase quantity is used to characterize the current stage state of rising or falling. When the current direction is rising, the fluctuation phase quantity takes the rising phase; when the current direction is falling, the fluctuation phase quantity takes the falling phase. To reduce the impact of direction switching at the boundary of the fluctuation segment on the fluctuation phase quantity, a boundary duration is set, which is set to multiple sampling times. When the time interval between the current sampling time and the end time mark of the most recent fluctuation segment is less than or equal to the boundary duration, the fluctuation phase quantity remains the phase corresponding to the direction of the most recent fluctuation segment and does not change immediately with the single direction mark of the next sampling time.

[0081] The coupling level is obtained by calculating the co-directional occupancy rate of sea clutter energy and throughput from the detection and observation sequence. The detection and observation sequence is indexed by time stamps, and each sequence item contains the sea clutter energy at the same sampling time. The sea clutter energy is obtained from echo data within the sea surface interval. At each sampling time, the echo sampling content corresponding to the sea surface interval is selected. The echo sampling content contains multiple echo sampling values. For each echo sampling value, an energy sampling value is formed by squared amplitude. When the echo sampling value contains real and imaginary parts, the energy sampling value is formed by the sum of the squares of the real and imaginary parts. The average value of all energy sampling values ​​is calculated to obtain the sea clutter energy at the current sampling time. The sea clutter energy is recorded as a sea clutter energy recording item, and the sea clutter energy recording items are arranged in time stamp order to form the detection and observation sequence. To avoid direction reversal caused by missing echo data, an available state is set in the detection and observation sequence. The available state is used to indicate whether the sea clutter energy recording item is a valid acquisition value. When the echo data is missing at the current sampling time, the sea clutter energy is processed by keeping the most recent valid value, and the available state is invalidated.

[0082] Under evaporative waveguide or tropospheric waveguide conditions, near-sea surface refractive index gradient anomalies simultaneously alter the propagation and scattering of radar near-range sea surface echoes, causing directional fluctuations in sea clutter energy over short timescales. The same anomalous propagation conditions also change the effective path loss and multipath fading state of microwave links, causing throughput and queuing backlog to exhibit synchronous or quasi-synchronous directional changes. Based on this, this embodiment uses the directional occupancy relationship between sea clutter energy and throughput at adjacent sampling times as an observable proxy quantity for coupling strength. By using available state, smoothing processing, and small fluctuation thresholds to suppress the influence of occasional noise on direction judgment, the coupling level can stably characterize the degree of synchronous disturbance between the detection side and the transmission side under anomalous propagation conditions.

[0083] The link observation sequence also includes an availability status. The availability status of the link observation sequence is valid when both the throughput record and the transmission queuing time exist, and invalid when either record is missing. When a record is missing, the throughput and transmission queuing time are kept at the most recent valid value and are subject to the upper limit constraint of the backtracking. The second window length is set based on the fluctuation scale table. The second window length is used to limit the statistical time range of the same-direction occupancy rate.

[0084] To ensure that the direction markers for sea clutter energy and throughput have a consistent calculation process, the same direction marker generation rules are applied to both sea clutter energy and throughput within the time marker covered by the second window length. A first smoothing length is set based on the first sampling period. The sea clutter energy undergoes a first smoothing process, and the throughput undergoes a second smoothing process. Both the first and second smoothing processes use a fixed-length moving average method, with the fixed length being the first smoothing length and the second smoothing length, respectively. A sea clutter energy change threshold and a throughput change threshold are set. The sea clutter energy change threshold is used to generate the direction marker for sea clutter energy, and the throughput change threshold is used to generate the direction marker for throughput. The sea clutter energy change threshold and the throughput change threshold are used to exclude direction reversals caused by small fluctuations.

[0085] The method for setting the sea clutter energy change threshold is as follows: Within the time-marked range covered by the second window length, the sea clutter energy after the first smoothing process is read. All adjacent sampling time pairs with valid available states are extracted according to the time-marked order. The absolute value of the sea clutter energy difference corresponding to each adjacent sampling time pair is calculated, and the absolute values ​​of the sea clutter energy difference are arranged in ascending order. To ensure that the sea clutter energy change threshold can exclude direction reversal caused by small fluctuations and to avoid direct direction judgment by individual abnormal changes, the absolute value of the sea clutter energy difference in the middle position after sorting is determined as the sea clutter energy change threshold. When there is no sea clutter energy difference absolute value available for calculation within the time-marked range covered by the second window length, the sea clutter energy change threshold corresponding to the previous sampling time remains unchanged. When there is no sea clutter energy change threshold corresponding to the previous sampling time, the threshold is set based on the currently obtained sea clutter energy record. The absolute value of the first non-zero sea clutter energy difference formed between the two samples determines the sea clutter energy change threshold. The method for setting the throughput change threshold is the same as that for the sea clutter energy change threshold. Within the time-marked range covered by the second window length, the throughput after the second smoothing process is read. All adjacent sampling time pairs with valid available states are extracted according to the time-marked order. The absolute value of the throughput difference corresponding to each adjacent sampling time pair is calculated, and the absolute values ​​of the throughput difference are arranged in ascending order. The absolute value of the throughput difference in the middle position after sorting is determined as the throughput change threshold. When there is no absolute value of throughput difference that can be used for calculation within the time-marked range covered by the second window length, the throughput change threshold corresponding to the previous sampling time remains unchanged. When there is no throughput change threshold corresponding to the previous sampling time, the throughput change threshold is determined by the absolute value of the first non-zero throughput difference formed between the currently obtained throughput records.

[0086] The previous and subsequent sampling times are selected sequentially for comparison. Sea clutter energy direction markers and throughput direction markers are generated only if both the previous and subsequent sampling times are valid in the detection and link observation sequences. When the difference between the sea clutter energy at the subsequent sampling time and the sea clutter energy at the previous sampling time is greater than or equal to the sea clutter energy change threshold, the sea clutter energy direction marker is set to an upward direction; when the difference is greater than or equal to the sea clutter energy change threshold, the sea clutter energy direction marker is set to a downward direction; otherwise, the sea clutter energy direction marker is set to a level direction. The throughput direction marker is generated in the same way, except that the sea clutter energy change threshold is replaced with the throughput change threshold.

[0087] The occupancy rate in the same direction is determined by the occupancy relationship between the number of times in the same direction and the number of valid times. Both the number of times in the same direction and the number of valid times are counted within the time stamp range covered by the second window length. The counting method for valid times is as follows: within the time stamp range covered by the second window length, for each group of previous and next sampling times, a judgment is made. If both sampling times are valid in the available state in the detection observation sequence and the link observation sequence, one valid count is recorded. If there is an invalid available state in either the previous or next sampling time, the adjacent sampling times before and after the current group are not counted as valid counts. The counting method includes: in adjacent sampling times before and after the valid count, when the direction marker of sea clutter energy and the direction marker of throughput are both rising, it is counted as one count; when the direction marker of sea clutter energy and the direction marker of throughput are both falling, it is counted as one count; when either direction marker is level, it is not counted as a count. The count only reflects the synchronous occupancy relationship when there is a clear directional change in sea clutter energy and throughput; the output range of the occupancy rate is limited to 0 to 1. When the valid count is 0, the occupancy rate is set to 0 and the second window length is kept unchanged.

[0088] Set a first proportion threshold and a second proportion threshold; to determine the value positions of the first proportion threshold and the second proportion threshold, first count the number of fluctuation segments within the time mark covered by the second window length, and count the total number of sampling times corresponding to the second window length, and then divide the total number of sampling times corresponding to the second window length by the number of fluctuation segments to obtain the average fluctuation segment duration.

[0089] After determining the alternation state, a first percentage threshold and a second percentage threshold are determined according to the alternation state. For example, when the rising and falling segments of the sending queue time alternate frequently, the first percentage threshold is 0.65 to 0.7, and the second percentage threshold is 0.8 to 0.9; when the rising and falling segments of the sending queue time alternate infrequently, the first percentage threshold is 0.55 to 0.6, and the second percentage threshold is 0.7 to 0.78; when the rising and falling segments of the sending queue time alternate in an intermediate state, the first percentage threshold is 0.6 to 0.65, and the second percentage threshold is 0.78 to 0.8.

[0090] When the occupancy rate in the same direction is greater than or equal to the second occupancy threshold, the coupling level is set to the third level; when the occupancy rate in the same direction is less than the second occupancy threshold but greater than or equal to the first occupancy threshold, the coupling level is set to the second level; when the occupancy rate in the same direction is less than the first occupancy threshold, the coupling level is set to the first level. The coupling level is determined by the occupancy relationship between sea clutter energy and throughput at the same sampling time. The acquisition aperture, direction marker generation rules, counting rules, rejection rules and output range of sea clutter energy are all clearly defined. This can reduce the probability of the coupling level shifting due to unclear input scale or improper handling of missing items, thereby reducing the probability of delay caused by alarm information still being occupied by the transmission window during the short-term link improvement phase.

[0091] Based on the fluctuation scale table, fluctuation phase quantity, and coupling level, the first update cycle, second update cycle, alarm retention time slot, retransmission limit, fragmentation upper limit, first bit rate level, and first redundancy are determined, resulting in a transmission budget table, which specifically includes:

[0092] Short and long scales are extracted based on the fluctuation scale table; the most recent fluctuation segments are selected; the selected fluctuation segment durations are arranged in ascending order of value, and the middle value of the fluctuation segment duration in the first half of the range is taken as the short scale, and the middle value of the fluctuation segment duration in the second half of the range is taken as the long scale; both the short and long scales are represented by the number of sampling times, and the short scale is less than or equal to the long scale; by defining the short and long scales as two fixed values, the subsequent process of determining the update cycle has a direct execution basis.

[0093] The second update cycle and the first update cycle are determined based on short scale, long scale, fluctuation phase quantity, and coupling level; when the coupling level is third level, the second update cycle takes the short scale; when the coupling level is first level, the second update cycle takes the long scale; when the coupling level is second level, the second update cycle takes the intermediate value between the short scale and the long scale; fluctuation phase quantity is used to determine the correspondence between the first update cycle and the second update cycle.

[0094] The alarm retention time slot is determined based on the second update cycle, the fluctuation phase quantity, and the coupling level; the continuous interval of sampling time corresponding to the second update cycle is taken as a transmission window, and only one pre-continuous alarm retention time slot is set in each transmission window. The pre-continuous alarm retention time slot extends continuously from the start time mark of the transmission window.

[0095] The retransmission limit and fragmentation limit are determined based on the fluctuation scale table, alarm retention time slot and coupling level; the representative value of throughput and the representative value of transmission queuing time corresponding to the most recent fluctuation segment in the fluctuation scale table are selected as the link level quantification basis for the current transmission window; the retransmission limit is used to limit the upper limit of the occupancy for retransmission within the transmission window, and the retransmission limit is expressed in the number of sampling time points.

[0096] To determine the range of the upper limit for fragmentation, firstly, representative throughput values ​​corresponding to the most recent fluctuation segments are selected from the fluctuation scale table, and these values ​​are arranged in ascending order. The value at the middle position is taken as the baseline throughput value. Next, representative transmission queuing times corresponding to the most recent fluctuation segments are selected from the fluctuation scale table, and these values ​​are arranged in ascending order. The value at the middle position is taken as the baseline transmission queuing time value. If the representative throughput value of the current fluctuation segment is less than the baseline throughput value, it is determined to be in a low position. If the representative throughput value of the current fluctuation segment is greater than or equal to the baseline throughput value, it is determined to be in a high position. If the representative transmission queuing time value of the current fluctuation segment is greater than or equal to the baseline transmission queuing time value, it is determined to be in a high position. If the representative transmission queuing time value of the current fluctuation segment is less than the baseline transmission queuing time value, it is determined to be in a low position.

[0097] When the coupling level is level 3, the value closest to the lower limit of the predetermined value range is taken. When the coupling level is level 1, the value closest to the upper limit of the predetermined value range is taken. When the coupling level is level 2, the value in the middle of the predetermined value range is taken.

[0098] Based on the fluctuation scale table, the first update cycle, and the coupling level, the first code rate level and the first redundancy are determined, and a transmission budget table is generated. The larger the level number, the larger the corresponding first quota value.

[0099] To determine the range of the first redundancy, the representative values ​​of the packet loss rate corresponding to the most recent fluctuation segments are selected from the fluctuation scale table. These representative values ​​are then arranged in ascending order, and the value at the middle position is taken as the baseline value of the packet loss rate. When the representative value of the packet loss rate corresponding to the current fluctuation segment is greater than or equal to the baseline value, it is determined that the representative value of the packet loss rate corresponding to the current fluctuation segment is in a high position. When the representative value of the packet loss rate corresponding to the current fluctuation segment is less than the baseline value, it is determined that the representative value of the packet loss rate corresponding to the current fluctuation segment is in a low position.

[0100] In this embodiment, the first code rate level is used to characterize the throttling level at the entry of the transmission queue. The first code rate level corresponds one-to-one with the first quota value. The first code rate level does not directly represent the physical layer modulation and coding level. The first quota value is used to limit the total amount of data allowed to enter the transmission queue within a first update cycle.

[0101] Alarm fast packets are sent within the current update cycle execution interval, but alarm fast packets do not occupy the first quota value; fragment index tables, echo data fragments, track data entries, redundant fragments, and retransmission entries are sent within the current update cycle execution interval, and all occupy the first quota value; by distinguishing the applicable relationship of transmission parameters and the occupation relationship of the first quota value within the update cycle execution interval, the execution boundary between alarm fast packets and other objects can be kept clear.

[0102] In terms of execution, at the start of the first update cycle, the first quota value is loaded as the first remaining quota. The first remaining quota decreases as data enters the transmission queue during the first update cycle. When generating cycle record entries, the first quota value corresponding to any first update cycle is made greater than or equal to twice the fragmentation limit. The echo data fragments, track data entries, redundant fragments, and retransmission entries in the fragmentation transmission list have their corresponding lengths calculated before entering the transmission queue. When the corresponding length is less than or equal to the first remaining quota, the current object enters the transmission queue, and the corresponding number of bytes is deducted from the first remaining quota. When the corresponding length is greater than the first remaining quota, the current object does not enter the transmission queue during the current first update cycle. The current object is entered into the transmission queue and retained until the next first update cycle for further processing. If the current echo data fragment, current track data entry, current redundant fragment, or current retransmission entry has not entered the transmission queue for two consecutive first update cycles, the current object is re-segmented and then added to the fragment transmission list corresponding to the next first update cycle. The current echo data fragment, current redundant fragment, and current retransmission entry are re-segmented according to the fragment upper limit, and the current track data entry is re-segmented according to the single-entry length upper limit, which is the fragment upper limit. By synchronously applying the re-segmentation rules to the current track data entry, the execution boundary of the track data entry can be kept consistent with that of the echo data fragment.

[0103] Alarm fast packets are not constrained by the first remaining quota. Alarm fast packets are given priority to enter the transmission queue within the alarm reservation time slot to prevent alarm fast packets from entering the transmission queue if the first quota value is occupied by echo data fragments, track data entries, redundant fragments, or retransmission entries in the current first update cycle. The fragment index table enters the transmission queue along with the alarm fast packets, and the length of the fragment index table is included in the first remaining quota so that the amount of data entering the transmission queue other than alarm fast packets is constrained by the first quota value.

[0104] By implementing the first code rate level as the throttling at the entry of the transmission queue, the position of the first code rate level is consistent with the position of the transmission queue duration acquisition. The first code rate level can directly constrain the amount of data entering the transmission queue without relying on the switching of physical transmission parameters of the front-end microwave link equipment. The first code rate level, together with the fluctuation scale table, coupling level, and first update cycle, limits the intensity of data entering the transmission queue, reduces the probability that the transmission queue duration will be cumulatively lengthened in the fluctuation segment, and reduces the probability that the retransmission entries and redundant fragments will squeeze the alarm retention time slot after the expansion of the occupation in the transmission queue.

[0105] To ensure the selection of the first redundancy has a quantifiable basis, packet loss rate records are synchronously collected at the front-end microwave link equipment according to the first sampling period when forming the link observation sequence. The packet loss rate record is the occupancy relationship between the amount of data not transmitted and the amount of data sent within the first window length. The packet loss rate record retains a time stamp and corresponds to the throughput and transmission queuing time using the same time stamp. When a packet loss rate record is missing, the most recent valid value is retained and the available state is invalidated, so that the packet loss rate record has a time structure consistent with the throughput and transmission queuing time in the link observation sequence. When determining the first redundancy, the start and end time stamps corresponding to the most recent fluctuation segment are selected based on the fluctuation scale table. Packet loss rate records are extracted from the link observation sequence, and the value corresponding to the middle position of the valid packet loss rate record is taken as the representative value of the packet loss rate, so that the representative value of the packet loss rate, the representative value of the throughput, and the representative value of the transmission queuing time are constrained by the same time range.

[0106] The first redundancy is implemented by taking redundant fragments. Redundant fragments are fragment entries added to the fragment sending list. The length of the redundant fragment is less than or equal to the upper limit of the fragment. The redundant fragment has a corresponding entry in the fragment index table.

[0107] To ensure a reproducible recovery path for redundant fragments, the receiver first locates the target fragment group based on the fragment group identifier according to the fragment index table, and then locates the continuous coverage segment corresponding to the redundant fragment based on the coverage start fragment number and coverage end fragment number. When only one echo data fragment is missing in the continuous coverage segment, the receiver selects the corresponding redundant fragment and the remaining echo data fragments already received in the same continuous coverage segment, and performs an XOR operation sequentially according to byte position to recover the missing echo data fragment. When the number of missing fragments in the continuous coverage segment is greater than or equal to two, no recovery is performed on the continuous coverage segment. After recovery, the receiver removes the trailing zeros according to the original fragment length recorded in the fragment index table. By recording the coverage start fragment number and coverage end fragment number in the fragment index table, the receiver can determine the correspondence between the redundant fragment and the original fragment.

[0108] The time stamp corresponding to the second update cycle is used as the effective starting point of the transmission budget table, and the effective ending point is obtained by extending the second update cycle backward from the effective starting point. The transmission budget table forms window record items according to the second update cycle, and each window record item corresponds to a transmission window. Each window record item records the time stamp corresponding to the effective starting point, the time stamp corresponding to the effective ending point, the second update cycle, the alarm retention time slot, the retransmission limit, and the fragmentation limit. Within the same window record item, multiple cycle record items are recorded in the order of the first update cycle. Each cycle record item corresponds to a first update cycle starting point and includes the time stamp corresponding to the first update cycle starting point, the first bitrate, the first redundancy, and the first quota value. When the second update cycle contains only one first update cycle, only one cycle record item is recorded within the window record item. When the second update cycle contains multiple first update cycles, multiple cycle record items are recorded within the window record item in chronological order.

[0109] Based on the transmission budget table, the collected echo data is segmented according to the segmentation limit to obtain a segmented transmission list and generate a segmentation index table. The segmented transmission list contains the collected track data. The collected alarm data is processed according to the alarm retention time slot to obtain alarm fast packets, and the alarm fast packets are associated with the segmentation index table, specifically including:

[0110] The data organization boundary within the sending window is determined based on the sending budget table. The sending budget table includes the second update cycle and alarm retention slots. The time stamp corresponding to the effective start point of the sending budget table is used as the starting point of the sending window, and the time stamp range covered by the second update cycle is used as the sending window boundary. Within the sending window, the time stamp range corresponding to a preceding continuous alarm retention slot is determined based on the alarm retention slots to obtain the alarm retention slot range. The alarm retention slot range adopts the continuous time stamp range after the starting point of the sending window, and the value rule follows the occupancy ratio setting result of the alarm retention slots in the sending budget table to avoid the alarm retention slots being discretely distributed within the sending window and introducing additional queuing time changes. The sending budget table is transformed into boundary conditions that can directly constrain data segmentation and sending order, reducing the disordered resource occupation caused by unclear boundaries within the sending window and reducing the probability of head blocking.

[0111] Within the transmission window, echo data, track data, and alarm data are collected and aligned. Within the time stamp range corresponding to the transmission window, the front-end radar processing chain continuously outputs echo data, which includes echo sampling content arranged according to the time stamp. The front-end radar processing chain continuously outputs track data, which is track information output by the front-end tracking processing, and includes track time information and location information. The front-end radar processing chain continuously outputs alarm data, which is high-priority alarm information output by the front-end radar, and includes time information, location information, and confidence information. The corresponding time stamps are retained for the above echo data, track data, and alarm data. Based on the time stamps, the sampling time correspondence of the three types of data is established to obtain the alignment relationship within the transmission window. The correspondence between the detection side output and the link side scheduling boundary is established using time stamps, reducing the probability of difficulty in locating the source of resource occupation when alarm data is squeezed by echo data or track data, and reducing the probability of alarm transmission delay being amplified.

[0112] Based on the transmission budget table, echo data is segmented according to the segmentation limit to obtain a segmented transmission list. The transmission budget table contains the segmentation limit. Taking the echo data within the transmission window as the object, the echo data is segmented according to the segmentation limit. The segmentation principle is that the length of a single data segment is less than or equal to the segmentation limit. The data segments cover the continuous position range of the echo data within the transmission window. The data segments retain the start time marker and end time marker. The data segments are arranged in order of their start time markers to obtain the segmented transmission list. Each entry in the segmented transmission list corresponds to a data segment.

[0113] A fragmentation index table is generated based on the fragmentation transmission list, and track data is included in the fragmentation transmission list. Echo data is divided according to the upper limit of the fragmentation to form echo data fragmentation entries. Track data is not divided according to the upper limit of the fragmentation, but is formed as track data entries by single track records and added to the fragmentation transmission list. The fragmentation index table records the entry type, start time marker, end time marker, and entry length for each entry in the fragmentation transmission list. The entry type must at least distinguish between echo data fragmentation entries, track data entries, alarm fast packet entries, redundant fragmentation entries, and retransmission entries. Track data is arranged by time marker and added to the fragmentation transmission list as independent track data entries, and the corresponding entry type is recorded in the fragmentation index table.

[0114] Alarm data is processed according to the alarm retention time slot to obtain alarm fast packets, and the alarm fast packets are associated with the fragment index table; the alarm data processing period within the current sending window is limited by the alarm retention time slot range, and alarm fast packet generation processing is only performed on alarm data corresponding to the timestamps within the alarm retention time slot range within the current sending window; alarm data that does not fall within the alarm retention time slot range of the current sending window is retained and processed first in the alarm retention time slot of the next sending window; time information, location information, and confidence information are retained for each alarm data to form an alarm fast packet; when the length of the alarm fast packet formed by a single alarm data is less than or equal to When the current alarm fast packet length limit is reached, a single alarm data is formed into an alarm fast packet. When the length of the alarm fast packet formed by a single alarm data exceeds the current alarm fast packet length limit, the single alarm data is further divided into multiple alarm fast packets according to the current alarm fast packet length limit. Multiple alarm fast packets are added to the fragmentation transmission list consecutively according to the same timestamp order. The alarm fast packet is added to the fragmentation transmission list as an independent entry. A corresponding entry is formed for each alarm fast packet in the fragmentation index table. The entry contains the timestamp of the alarm fast packet, the corresponding time information, and the range of the associated echo data entry.

[0115] The first bitrate and first redundancy are updated according to the first update cycle. The transmission window is rotated according to the second update cycle. Alarm fast packets are sent first, followed by echo data and track data according to the fragmented transmission list. The echo data stream is obtained by supplementing transmission according to the supplementary transmission limit. Specifically, it includes:

[0116] Based on the transmission budget table, establish the boundaries of the first and second update cycles; the window record items in the transmission budget table provide the timestamps corresponding to the effective start point, the timestamps corresponding to the effective end point, and the second update cycle; the cycle record items in the transmission budget table provide the timestamps corresponding to the start points of each first update cycle, the first bitrate, the first redundancy, and the first quota value; arrange the timestamps corresponding to all sampling moments between the timestamps corresponding to the effective start point and the timestamps corresponding to the effective end point in chronological order to form a timestamp sequence; determine the second update cycle boundaries sequentially in the timestamp sequence based on the window record items to form a second update cycle boundary sequence; based on the cycle record items... The first update cycle boundary is determined sequentially in the time stamp sequence to form the first update cycle boundary sequence; a sending window is formed between two adjacent second update cycle boundaries, and an update cycle execution interval is formed between two adjacent first update cycle boundaries; only one pre-alarm retention time slot is set in each sending window, and the alarm retention time slot extends continuously from the start time stamp of the sending window, and the corresponding number of sampling times is determined by the alarm retention time slot in the window record item; the retransmission limit corresponds to the upper limit of the number of sampling times allowed for retransmission in the current sending window; by jointly establishing the time boundary through the window record item and the cycle record item, the sending budget table can be transformed into a directly executable time structure.

[0117] The first code rate and first redundancy are updated according to the first update cycle, and the transmission queuing time is constrained to increase. When processing each first update cycle starting point along the first update cycle boundary sequence, the first code rate, first redundancy, and first quota value are read from the cycle record item corresponding to the current first update cycle starting point, and the reading result is used as the execution parameter of the current update cycle execution interval. Within the current update cycle execution interval, the alarm fast packets, fragment index table, echo data fragments, track data entries, redundant fragments, and retransmission entries to be transmitted all adopt the first code rate and first redundancy corresponding to the current first update cycle, until entering the next update cycle execution interval, and then switching to the first code rate and first redundancy corresponding to the next first update cycle.

[0118] The transmission window is rotated according to the second update cycle, and a transmission order constraint is formed. The transmission windows are sequentially entered along the boundary sequence of the second update cycle. At the beginning of each transmission window, the alarm retention time slot range corresponding to this transmission window is first determined, taking the time marker range of the previous period of this transmission window. Then, the retransmission limit corresponding to this transmission window is determined, represented by the number of sampling times allowed for retransmission within this transmission window. Subsequently, the range of fragment transmission list entries allowed to be transmitted within this transmission window is determined, taking the entries in the fragment transmission list whose starting time marker falls within the time marker range of this transmission window. At the same time, entries in the fragment index table corresponding to the fragment transmission list entry range are retained. A fixed transmission window is formed by rotating according to the second update cycle. The alarm retention time slot and retransmission limit maintain consistent constraints within the window, reducing the probability of delays caused by long fragments left over from the previous transmission window during the short-term link improvement phase, and reducing the probability of head blocking conditions being continued at window switching points.

[0119] First, an alarm fast packet is sent, and the alarm reservation time slot is maintained in the return data stream. After entering the sending window, alarm fast packet entries are selected from the fragmented sending list, with the selection rule being that the timestamp of the alarm fast packet entry is within the alarm reservation time slot range. The alarm fast packet entries are sent in the order of their timestamps, using the first bitrate and first redundancy determined in this first update cycle. After the alarm fast packet is sent, the fragmented index table is sent as part of the return data stream. The fragmented index table takes the range of entries corresponding to this sending window, and the sending of the fragmented index table still uses the first bitrate and first redundancy of this first update cycle. The timing record records the timestamp, sending sequence number, corresponding first bitrate, and corresponding first redundancy during the alarm fast packet and fragmented index table sending processes. At the beginning of each sending window, the alarm reservation time slot is occupied first to complete the alarm fast packet sending, reducing the probability of alarm data not entering the return data stream in a timely manner due to the retransmission task and echo data fragments occupying the sending queuing time first, and reducing the risk of increased time delay in receiving alarm information by the command and control center.

[0120] Then, echo data and track data are sent according to the fragmented transmission list, and retransmission is limited according to the retransmission limit to form the echo data stream and timing record. After completing the transmission of alarm fast packets and fragmented index table, echo data fragments and track data entries are sent sequentially within the entry range of the fragmented transmission list in this transmission window. The transmission order is the order of the entries in the fragmented transmission list. Both echo data fragments and track data entries use the first code rate and the first redundancy of this first update cycle. If there are retransmission entries in the fragmented transmission list, the transmission occupation of the retransmission entries is included in the retransmission limit. The number of sampling times occupied by the retransmission entries in this transmission window is less than or equal to the retransmission limit. Retransmission entries exceeding the retransmission limit are not sent in this transmission window. The timing record is recorded in the echo data. During the transmission of fragmented, track data entries, and retransmission entries, the time stamp, transmission sequence number, corresponding first code rate, and corresponding first redundancy are recorded, along with the second update cycle boundary information corresponding to the current transmission window. The return data stream contains alarm fast packets, fragmented index tables, echo data fragments, and track data entries in the actual transmission order, and also includes fragmented index tables. During non-alarm phases, the fragmented transmission list maintains an interpretable order, and retransmission quotas constrain retransmission usage, reducing the probability of low-priority retransmission occupying the transmission queue for a long time, reducing the probability of alarm-related information being squeezed out by retransmission during short-term link improvement phases, and improving the ability of border and coastal defense radar data transmission to accommodate both alarm information and normal data under evaporative waveguide conditions.

[0121] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An adaptive transmission method for microwave links for radar backhaul in border and coastal defense, characterized in that, include: At each sampling moment, the throughput, transmission queue time, and sea clutter energy are collected. The queuing time for sending is segmented to obtain fluctuation segments; For each fluctuation segment, record the start time and end time, as well as the direction and duration of the fluctuation segment. The duration of the fluctuation segment is determined by the start time and end time. Extract the representative value of throughput and the representative value of sending queue time within the current fluctuation segment. The representative value of throughput is taken from the middle position of the throughput record in the current fluctuation segment, and the representative value of sending queue time is taken from the middle position of the sending queue time in the current fluctuation segment. The fluctuation scale table is arranged in chronological order of fluctuation segments. Each item in the fluctuation scale table includes a start time marker, an end time marker, the direction of the fluctuation segment, the duration of the fluctuation segment, a representative value of throughput, and a representative value of transmission queuing time. Select the most recent fluctuation segment from the fluctuation scale table, and take the direction of the most recent fluctuation segment as the current direction; When the current direction is upward, the fluctuation phase is measured as the upward phase; when the current direction is downward, the fluctuation phase is measured as the downward phase. Set a boundary duration. When the time interval between the current sampling time and the end time marker of the most recent fluctuation segment is less than or equal to the boundary duration, the fluctuation phase quantity remains the phase corresponding to the fluctuation direction of the most recent fluctuation segment. After determining the current direction, the fluctuation phase quantity is generated; The coupling level is obtained by calculating the co-directional occupancy rate of sea clutter energy and throughput; the co-directional occupancy rate is determined by the occupancy relationship between the number of co-directional waves and the effective number of waves. Based on the fluctuation scale table, fluctuation phase quantity and coupling level, the first update cycle, second update cycle, alarm retention time slot, retransmission limit, fragmentation limit, first bit rate level and first redundancy are determined to obtain the transmission budget table. Based on the transmission budget table, the collected echo data is divided into segments according to the upper limit of the segments to obtain a segment transmission list and generate a segment index table. The segment transmission list includes the collected track data. The collected alarm data is processed according to the alarm retention time slot to obtain alarm fast packets, and the alarm fast packets are associated with the segment index table. The first bitrate and first redundancy are updated according to the first update cycle. The transmission window is rotated according to the second update cycle. First, alarm fast packets are sent, then echo data and track data are sent according to the fragmented transmission list, and the return data stream is obtained by supplementing the transmission limit according to the supplementary transmission limit.

2. The microwave link adaptive transmission method for radar backhaul for border and coastal defense according to claim 1, characterized in that, Methods for obtaining throughput record entries include: The throughput is taken as the sliding window throughput, which is obtained by the transmit count of the front-end microwave link equipment; Set the first window length, and when collecting throughput, record the amount of valid data sent within the first window length to form a throughput record item.

3. The microwave link adaptive transmission method for radar backhaul for border and coastal defense according to claim 1, characterized in that, Methods for obtaining fluctuation segments include: Traverse the direction markers of the transmission queue duration in the order of the time markers. Record the sampling time when the direction marker of the transmission queue duration is in the upward direction as the rising segment, and record the sampling time when the direction marker of the transmission queue duration is in the downward direction as the falling segment. Summarize the rising segment and the falling segment to obtain the fluctuation segment.

4. The microwave link adaptive transmission method for radar backhaul for border and coastal defense according to claim 3, characterized in that, Other methods for obtaining fluctuation segments include: Set a lower limit for the segment length. If the length of the rising segment is less than the lower limit, the rising segment will be merged into the adjacent rising segment on the left or the adjacent rising segment on the right with the same direction. If the directions of the adjacent rising segment on the left and the adjacent rising segment on the right are not the same as the rising segment, the lengths of the adjacent rising segment on the left and the adjacent rising segment on the right will be compared, and the rising segment will be merged into the adjacent rising segment with the longer length, resulting in the merged rising segment.

5. The microwave link adaptive transmission method for radar backhaul for border and coastal defense according to claim 1, characterized in that, The methods for obtaining the number of times in the same direction and the number of valid times include: Set a second window length, and count both the number of times in the same direction and the number of valid times within the time marker covered by the second window length; The counting method for valid counts is as follows: when both the probe observation sequence and the link observation sequence are valid at each pair of adjacent sampling times, it is counted as 1 valid count; The counting method for the same direction count is as follows: at each pair of adjacent sampling times, if the direction marker of sea clutter energy and the direction marker of throughput are both in the upward direction or both in the downward direction, it is counted as 1 same direction count. If either direction marker is in the same direction, the adjacent sampling times are not counted as same direction count.

6. The microwave link adaptive transmission method for radar backhaul for border and coastal defense according to claim 5, characterized in that, Methods for obtaining the directional markers of sea clutter energy include: Set the threshold for sea clutter energy variation; When the difference between the sea clutter energy at the next sampling time and the sea clutter energy at the previous sampling time is greater than or equal to the sea clutter energy change threshold, the direction of the sea clutter energy is marked in the upward direction. When the difference between the sea clutter energy at the previous sampling time and the sea clutter energy at the next sampling time is greater than or equal to the sea clutter energy change threshold, the direction of the sea clutter energy is marked by the falling direction. Otherwise, the direction of sea clutter energy is marked in the same direction.

7. The microwave link adaptive transmission method for radar backhaul for border and coastal defense according to claim 6, characterized in that, Methods for obtaining the direction marker of throughput include: Set a threshold for throughput changes; The throughput direction marker is generated using the same method as the sea clutter energy direction marker, except that the sea clutter energy change threshold is replaced with the throughput change threshold.

8. The microwave link adaptive transmission method for radar backhaul for border and coastal defense according to claim 6, characterized in that, Methods for obtaining link observation sequences and probe observation sequences include: A link observation sequence is formed based on throughput records and transmission queuing time, and a detection observation sequence is obtained based on sea clutter energy.

9. The microwave link adaptive transmission method for radar backhaul for border and coastal defense according to claim 1, characterized in that, Methods for obtaining coupling levels include: Set a first percentage threshold and a second percentage threshold; When the occupancy rate in the same direction is greater than or equal to the second percentage threshold, the coupling level is set to the third level. When the occupancy rate in the same direction is less than the second percentage threshold and greater than or equal to the first percentage threshold, the coupling level is the second level. When the occupancy rate in the same direction is less than the first percentage threshold, the coupling level is set to the first level.

Images