A bidirectional transmission verification method for HDMI interface data

Abstract

The present invention provides a method for verifying bidirectional transmission of interface data for HDMI. The method first obtains a bidirectional transmission data set of a target device, including data packet sequences captured from two mutually opposite transmission paths. Then, the two data packet sequences are subjected to time feature alignment processing to generate a bidirectional transmission time feature set. Then, the bidirectional transmission time feature set is subjected to path consistency analysis based on preset transmission path verification rules to generate a path verification parameter set. The path verification parameter set includes key parameters such as path timing matching degree, data integrity index, and transmission interruption frequency. Based on the comparison result of the path verification parameter set with a preset transmission compliance threshold, a dynamic adjustment signal is generated and mapped to the transmission configuration module of the target device to activate the transmission parameter update operation. Finally, after the transmission parameters are updated, a secondary verification operation is triggered to generate a final verification result set to ensure improved transmission performance and stability.

Description

Technical Field

[0001] The present invention relates to the technical field of high-definition multimedia interfaces, and in particular to a method for verifying bidirectional transmission of interface data for HDMI. Background Art

[0002] In the field of high-definition multimedia interface technology, the stability and accuracy of data transmission are crucial to ensuring high-quality transmission of audio and video signals. With the widespread application of multimedia data such as high-definition video and audio, the data transmission volume of the HDMI interface continues to increase, and the requirements for bidirectional verification of data transmission are also becoming increasingly higher. However, in the existing HDMI interface data transmission verification technology, it is often only focused on data verification in a single transmission direction, while ignoring the consistency and coordination of the bidirectional transmission path. The one-way verification method cannot fully evaluate the performance of the HDMI interface during bidirectional data transmission, which can easily lead to data transmission errors, signal loss or transmission delays in actual applications. In addition, the existing verification method lacks a dynamic adjustment mechanism for transmission parameters and cannot optimize the transmission configuration according to the real-time transmission status, thereby affecting the overall transmission efficiency and stability of the HDMI interface. Summary of the Invention

[0003] In view of the above-mentioned problems, in combination with the first aspect of the present invention, an embodiment of the present invention provides a method for verifying bidirectional transmission of interface data for HDMI, the method comprising:

[0004] Obtaining a bidirectional transmission data set of a target device, the bidirectional transmission data set comprising a first data packet sequence captured from a first transmission direction and a second data packet sequence captured from a second transmission direction, wherein the first transmission direction and the second transmission direction are mutually opposite transmission paths in an HDMI interface;

[0005] Performing time feature alignment processing on the first data packet sequence and the second data packet sequence to generate a bidirectional transmission time feature set;

[0006] Performing path consistency analysis on the bidirectional transmission time feature set based on a preset transmission path verification rule to generate a path verification parameter set for the first transmission direction and the second transmission direction, the path verification parameter set including a path timing matching degree, a data integrity index, and a transmission interruption frequency;

[0007] generating a dynamic adjustment signal according to a comparison result between the path verification parameter set and a preset transmission compliance threshold, and mapping the dynamic adjustment signal to a transmission configuration parameter change queue corresponding to a transmission configuration module of the target device to activate a transmission parameter update operation;

[0008] After the transmission parameter update operation is performed, a secondary verification operation for the updated transmission configuration is triggered, and a final verification result set is generated.

[0009] On the other hand, an embodiment of the present invention further provides an interface data bidirectional transmission verification system for HDMI, comprising a processor and a machine-readable storage medium, wherein the machine-readable storage medium is connected to the processor, the machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine-readable storage medium to implement the above method.

[0010] Based on the above aspects, the embodiment of the present invention realizes comprehensive monitoring and analysis of the bidirectional transmission path of the HDMI interface by obtaining the bidirectional transmission data set of the target device and performing time feature alignment processing on the first data packet sequence and the second data packet sequence, performs path consistency analysis processing on the bidirectional transmission time feature set based on the preset transmission path verification rules, generates a path verification parameter set, generates a dynamic adjustment signal according to the comparison result between the path verification parameter set and the preset transmission compliance threshold, and maps the dynamic adjustment signal to the transmission configuration parameter change queue corresponding to the transmission configuration module of the target device, thereby realizing dynamic adjustment and optimization of the transmission parameters and improving the transmission efficiency and stability of the HDMI interface. After the transmission parameter update operation is executed, a secondary verification operation for the updated transmission configuration is triggered, and a final verification result set is generated, thereby ensuring the effectiveness and reliability of the transmission parameter adjustment. Therefore, by comprehensively evaluating the performance of the bidirectional transmission path of the HDMI interface and realizing dynamic adjustment of the transmission parameters, the stability and accuracy of the HDMI interface data transmission are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 The present invention is a flowchart of an implementation method for verifying bidirectional transmission of HDMI interface data provided by an embodiment of the present invention.

[0012] Figure 2 FIG. 4 is a schematic diagram of exemplary hardware and software components of a system for verifying bidirectional transmission of interface data for HDMI provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The present invention will be described in detail below with reference to the accompanying drawings. Figure 1 FIG1 is a flow chart of a method for verifying bidirectional transmission of interface data for HDMI provided by an embodiment of the present invention. The method for verifying bidirectional transmission of interface data for HDMI is introduced in detail below.

[0014] Step S110: Obtain a bidirectional transmission data set of the target device, wherein the bidirectional transmission data set includes a first data packet sequence captured from a first transmission direction and a second data packet sequence captured from a second transmission direction, wherein the first transmission direction and the second transmission direction are mutually opposite transmission paths in the HDMI interface.

[0015] In actual HDMI data transmission scenarios, the target device can be any combination of devices that exchange data via the HDMI interface, such as a computer and a high-definition monitor. To capture bidirectionally transmitted data, a data capture module can be implemented at the physical level of the HDMI interface. This data capture module can be a hardware-based circuit that connects to the HDMI interface's data transmission lines and monitors and records data transmission in real time.

[0016] For the first transmission direction, let's assume data is being transmitted from a computer to a high-definition display. During the data transmission process, the data capture module captures each data packet in the order in which they are transmitted and stores them, forming a first data packet sequence. Let the first data packet sequence be P1 = {p11, p12, p13, ...}, where p1i represents the i-th data packet in the first transmission direction.

[0017] For the second transmission direction, that is, transmitting data from the HD display to the computer host, the data capture module also captures each data packet to form a second data packet sequence. Let the second data packet sequence be P2 = {p21, p22, p23, ...}, where p2j represents the jth data packet in the second transmission direction.

[0018] These data packets contain various types of data, such as video frame data, audio data, control signals, etc. When capturing data packets, the data capture module will record some basic information of each data packet, such as the size and data type of the data packet, for subsequent processing.

[0019] Step S120: performing time feature alignment processing on the first data packet sequence and the second data packet sequence to generate a bidirectional transmission time feature set.

[0020] Because the transmission of packets in the first and second transmission directions may be affected by different factors, resulting in time differences between them, it is necessary to perform time alignment on the two packet sequences in order to accurately analyze the time characteristics of data transmission in the first and second transmission directions.

[0021] Step S121: extracting a first initial timestamp set from the first data packet sequence, where the first initial timestamp set includes a sending end start timestamp and a receiving end end timestamp of each first data packet.

[0022] In the first data packet sequence P1, for each data packet p1i, its sending end timestamp and receiving end timestamp need to be recorded. The sending end timestamp refers to the time when the computer host starts sending the data packet, and the receiving end timestamp refers to the time when the high-definition display receives the data packet.

[0023] A time recorder can be installed on the host computer's sending port and the HD display's receiving port. When the host computer starts sending data packet p1i, the time recorder records the time at that time, which is recorded as t1si. When the HD display receives data packet p1i, the time recorder records the time at that time, which is recorded as t1ri.

[0024] In this way, the first initial timestamp set T1={(t1s1, t1r1), (t1s2, t1r2), (t1 step S26, t1r3), ...}, where (t1si, t1ri) represents the start timestamp of the sending end and the end timestamp of the receiving end of the i-th first data packet.

[0025] Step S122: extracting a second initial timestamp set from the second data packet sequence, where the second initial timestamp set includes a sending end start timestamp and a receiving end end timestamp of each second data packet.

[0026] Similarly, for each data packet p2j in the second data packet sequence P2, a time recording device is set up at the HD display's sending port and the host computer's receiving port. When the HD display begins to send data packet p2j, the time t2sj is recorded; when the host computer receives data packet p2j, the time t2rj is recorded.

[0027] The second initial timestamp set T2={(t2s1, t2r1), (t2s2, t2r2), (t2 step S26, t2r3), ...}, where (t2sj, t2rj) represents the start timestamp of the sending end and the end timestamp of the receiving end of the j-th second data packet.

[0028] Step S123: Perform cross-path synchronization processing on the first initial timestamp set and the second initial timestamp set, determine the time base offset between the first data packet sequence and the second data packet sequence, and perform compensation calibration on the first initial timestamp set and the second initial timestamp set based on the time base offset to generate a calibrated first calibration timestamp set and a calibrated second calibration timestamp set.

[0029] Since the clocks in the two transmission directions may differ, resulting in inconsistent time bases, cross-path synchronization is required. The specific steps are as follows:

[0030] Step S1231: Filter out a first reference timestamp pair that meets a preset time interval condition from the first initial timestamp set, where the first reference timestamp pair includes a sending end start timestamp and a corresponding receiving end end timestamp of two consecutive first data packets.

[0031] The preset time interval condition can be set according to actual conditions, for example, a fixed time interval Δt is set. In the first initial timestamp set T1, two adjacent timestamp pairs (t1si, t1ri) and (t1s(i+1), t1r(i+1)) are checked in sequence. If the difference between their sending end start timestamps |t1s(i+1)-t1si| meets the preset time interval condition Δt, these two timestamp pairs are used as the first reference timestamp pairs. Let the screened first reference timestamp pair set be R1={((t1s1, t1r1), (t1s2, t1r2)), ((t1 step S26, t1r3), (t1s4, t1r4)), ...}.

[0032] Step S1232: Filter out a second reference timestamp pair that meets the preset time interval condition from the second initial timestamp set, where the second reference timestamp pair includes a sending end start timestamp and a corresponding receiving end end timestamp of two consecutive second data packets.

[0033] Similarly, in the second initial timestamp set T2, according to the preset time interval condition Δt, two adjacent timestamp pairs (t2sj, t2rj) and (t2s(j+1), t2r(j+1)) are screened. If |t2s(j+1)-t2sj| satisfies Δt, they are used as the second reference timestamp pairs. Let the screened second reference timestamp pair set be R2 = {((t2s1, t2r1), (t2s2, t2r2)), ((t2 in step S26, t2r3), (t2s4, t2r4)), ...}.

[0034] Step S1233: Calculate a first difference between the sender start timestamps of two consecutive first data packets in the first reference timestamp pair, and calculate a second difference between the sender start timestamps of two consecutive second data packets in the second reference timestamp pair.

[0035] For each pair ((t1si, t1ri), (t1s(i+1), t1r(i+1))) in the first reference timestamp pair set R1, a first difference d1i=t1s(i+1)-t1si is calculated, resulting in a first difference set D1={d11, d12, d13, ...}.

[0036] For each pair ((t2sj, t2rj), (t2s(j+1), t2r(j+1))) in the second reference timestamp pair set R2, a second difference d2j=t2s(j+1)-t2sj is calculated. A second difference set D2={d21, d22, d23, ...} is obtained.

[0037] Step S1234: Determine a clock frequency difference coefficient between the first transmission direction and the second transmission direction according to a ratio between the first difference and the second difference.

[0038] For each set of first and second differences d1i and d2i (assuming i corresponds to the same time reference point), calculate their ratio k = d1i / d2i. Since there may be multiple sets of ratios, to obtain a more accurate clock frequency difference coefficient, these ratios can be statistically processed, such as by taking an average. Let the clock frequency difference coefficient be C, C = average(k), where average represents the averaging operation.

[0039] Step S1235: performing frequency normalization processing on the first initial timestamp set and the second initial timestamp set based on the clock frequency difference coefficient to generate a third timestamp set and a fourth timestamp set after frequency synchronization.

[0040] For each timestamp pair (t1si, t1ri) in the first initial timestamp set T1, multiply the sender's start timestamp t1si by the clock frequency difference coefficient C to obtain the normalized sender's start timestamp t1s'i = C*t1si. The same process is performed on the receiver's end timestamp t1ri, resulting in t1r'i = C*t1ri. This yields the third frequency-synchronized timestamp set T3 = {(t1s'1, t1r'1), (t1s'2, t1r'2), (t1s'3, t1r'3), ...}.

[0041] For each timestamp pair (t2sj, t2rj) in the second initial timestamp set T2, the same processing is performed to obtain the normalized sending end start timestamp t2s'j=C*t2sj and the receiving end end timestamp t2r'j=C*t2rj, thereby obtaining the fourth timestamp set T4={(t2s'1, t2r'1), (t2s'2, t2r'2), (t2s'3, t2r'3), ...}.

[0042] Step S1236: Determine the time base offset according to the offset between the sender start timestamps of the same data packet identifier in the third timestamp set and the fourth timestamp set.

[0043] In the third timestamp set T3 and the fourth timestamp set T4, find timestamp pairs with the same packet identifier. For the sender-start timestamps t1s'i and t2s'j with the same packet identifier, calculate their offset Δt = t1s'i - t2s'j. Similarly, to obtain a more accurate time base offset, statistical processing can be performed on multiple offsets with the same packet identifier, such as taking an average. Let the time base offset be Offset, and Offset = average(Δt).

[0044] Step S1237: performing compensation calibration on the first initial timestamp set and the second initial timestamp set based on the time base offset to generate a calibrated first calibrated timestamp set and a calibrated second calibrated timestamp set.

[0045] For each timestamp pair (t1si, t1ri) in the first initial timestamp set T1, subtract the time base offset Offset from the sender's start timestamp t1si to obtain the calibrated sender start timestamp t1s"i = t1si - Offset. The same process is performed on the receiver's end timestamp t1ri to obtain t1r"i = t1ri - Offset. This yields the calibrated first timestamp set T1' = {(t1s"1, t1r"1), (t1s"2, t1r"2), (t1s"3, t1r"3), ...}.

[0046] For each timestamp pair (t2sj, t2rj) in the second initial timestamp set T2, the same processing is performed to obtain the calibrated sending end start timestamp t2s"j=t2sj-Offset and the receiving end end timestamp t2r"j=t2rj-Offset, thereby obtaining the calibrated second calibration timestamp set T2'={(t2s"1, t2r"1), (t2s"2, t2r"2), (t2s"3, t2r"3), ...}.

[0047] Step S124: Generate a first transmission delay sequence based on the difference between the start timestamp of the sending end and the end timestamp of the receiving end in the first calibration timestamp set, and generate a second transmission delay sequence based on the difference between the start timestamp of the sending end and the end timestamp of the receiving end in the second calibration timestamp set.

[0048] For each timestamp pair (t1s″i, t1r″i) in the first calibration timestamp set T1′, calculate the difference between the start timestamp at the transmitter and the end timestamp at the receiver, i.e., the first transmission delay τ1i = t1r″i - t1s″i. This yields the first transmission delay sequence τ1 = {τ11, τ12, τ13, …}.

[0049] For each timestamp pair (t2s"j, t2r"j) in the second calibration timestamp set T2', calculate the difference between the start timestamp at the transmitter and the end timestamp at the receiver, i.e., the second transmission delay τ2j = t2r"j - t2s"j. This yields the second transmission delay sequence τ2 = {τ21, τ22, τ23, ...}.

[0050] Step S125: Perform windowed statistical analysis on the first transmission delay sequence and the second transmission delay sequence to determine the delay fluctuation range and average delay value within each preset time window, and associate the delay fluctuation range and average delay value with the corresponding data packet identifier to generate the bidirectional transmission time feature set.

[0051] The preset time window can be set according to actual needs, for example, a fixed time length W is set. The first transmission delay sequence τ1 and the second transmission delay sequence τ2 are divided into respective preset time windows in chronological order.

[0052] For each preset time window, in the first transmission delay sequence τ1, find the maximum delay value τ1max and the minimum delay value τ1min in the preset time window, and calculate the delay fluctuation range Δτ1=τ1max-τ1min; at the same time, calculate the average delay value μ1=average(τ1) in the preset time window, where average represents the operation of finding the average value.

[0053] Similarly, in the second transmission delay sequence τ2, find the maximum delay value τ2max and the minimum delay value τ2min in the preset time window, calculate the delay fluctuation range Δτ2=τ2max-τ2min; calculate the average delay value μ2=average(τ2) in the preset time window.

[0054] The latency fluctuation range and average latency value within each preset time window are associated with the corresponding packet identifier. For example, for the nth preset time window, the associated packet identifier is the set In of identifiers of all packets contained in the preset time window. The bidirectional transmission time feature set F = {((Δτ1n, μ1n, In), (Δτ2n, μ2n, In))}, where n represents the number of the preset time window.

[0055] Step S130: Based on the preset transmission path verification rules, the path consistency analysis is performed on the bidirectional transmission time feature set to generate a path verification parameter set for the first transmission direction and the second transmission direction. The path verification parameter set includes path timing matching, data integrity index and transmission interruption frequency.

[0056] The preset transmission path verification rules are a series of rules pre-set according to the characteristics and requirements of HDMI interface data transmission, and are used to evaluate the path consistency of the two transmission directions.

[0057] Step S131: Based on the delay fluctuation range and average delay value corresponding to each data packet identifier in the bidirectional transmission time feature set in each preset time window, a bidirectional delay difference vector is generated for each data packet, wherein the bidirectional delay difference vector includes the ratio of the delay fluctuation range to the average delay corresponding to the same data packet in the bidirectional transmission path, wherein the delay fluctuation range is the difference between the maximum and minimum delays of the same data packet transmitted multiple times in the preset time window.

[0058] For each data packet identifier In in the bidirectional transmission time feature set F, in the first transmission direction, the corresponding delay fluctuation range is Δτ1n, and the average delay value is μ1n; in the second transmission direction, the corresponding delay fluctuation range is Δτ2n, and the average delay value is μ2n.

[0059] Calculate the delay fluctuation range Rn=|Δτ1n-Δτ2n| and the average delay ratio Sn=μ1n / μ2n for the same data packet in the bidirectional transmission path.

[0060] Then the two-way delay difference vector of each data packet is Vn=(Rn, Sn).

[0061] Step S132: performing sliding window segmentation processing on the two-way delay difference vector to obtain multiple delay analysis windows, and calculating the cosine similarity between the two-way delay difference vectors of adjacent data packets in each delay analysis window to generate a window timing similarity sequence.

[0062] The size and step length of the sliding window can be set according to actual conditions, for example, the sliding window size is set to M and the step length is set to S.

[0063] The two-way delay difference vector sequence {V1, V2, V3, ...} is divided into sliding windows to obtain multiple delay analysis windows. For example, the two-way delay difference vectors contained in the k-th delay analysis window are {Vk, Vk+1, ..., Vk+M-1}.

[0064] Within each delay analysis window, the cosine similarity between the two-way delay difference vectors of adjacent data packets is calculated. For two adjacent two-way delay difference vectors Vi and Vi+1, the cosine similarity is calculated as cosθi=(Vi·Vi+1) / (||Vi||*||Vi+1||), where · represents the vector dot product and |||| represents the vector modulus.

[0065] The cosine similarities calculated in each delay analysis window are combined into a window timing similarity sequence S = {cosθ1, cosθ2, cosθ3, ...}.

[0066] Step S133: determining the path timing matching degree according to the degree of deviation between the window timing similarity sequence and a preset similarity threshold.

[0067] The preset similarity threshold is a value pre-set according to actual needs, and is used to measure whether the similarity in the window time series similarity sequence meets the requirements.

[0068] For each cosine similarity cosθi in the window time series similarity sequence S, the degree of deviation δi=|cosθi-Th| from the preset similarity threshold Th is calculated.

[0069] To obtain the path timing matching degree P, all deviations are statistically processed, for example, by taking the inverse of the average value. Let the path timing matching degree P = 1 / average(δi).

[0070] Step S134: extracting the absolute value of the delay difference between the first transmission delay sequence and the second transmission delay sequence in each delay analysis window, and calculating the frequency of occurrence of consecutive exceeding threshold intervals based on the time distribution density function of the absolute value of the delay difference to generate the transmission interruption frequency.

[0071] In each delay analysis window, for the delay values ​​τ1i and τ2i corresponding to the same data packet identifier in the first transmission delay sequence τ1 and the second transmission delay sequence τ2, the absolute value of the delay difference |τ1i-τ2i| is calculated.

[0072] Assume that the preset delay difference threshold is Thd. According to the time distribution density function of the absolute value of the delay difference, find the continuous threshold exceeding intervals, that is, the intervals in which the absolute value of the delay difference is greater than Thd in the continuous time period.

[0073] The occurrence frequency N of these consecutive exceeding-threshold intervals is counted and is used as the transmission interruption frequency.

[0074] Step S135: Perform packet identifier mapping verification on the first data packet sequence and the second data packet sequence, count the number of data packets that are not successfully mapped and the number of data packets that are repeatedly mapped in the first data packet sequence and the second data packet sequence, and proportionally convert the number of data packets with a preset integrity reference value to generate the data integrity index.

[0075] When performing packet identifier mapping verification on the first and second data packet sequences, it is important to first ensure that each data packet has a unique packet identifier. Let the first data packet sequence be P1, with data packets p1i (where i is the index), and the second data packet sequence be P2, with data packets p2j (where j is the index).

[0076] When performing packet identification mapping verification, the packet identification of each data packet in the first data packet sequence is compared one by one with the packet identification of each data packet in the second data packet sequence. In this process, the number of data packets that are not successfully mapped and the number of data packets that are repeatedly mapped need to be recorded.

[0077] For data packets that were not successfully mapped, this means that a packet identifier in the first data packet sequence does not match a corresponding packet in the second data packet sequence; or a packet identifier in the second data packet sequence does not match a corresponding packet in the first data packet sequence. Let Nu be the number of data packets that were not successfully mapped.

[0078] Repeatedly mapped packets refer to packets where a packet identifier is found to appear multiple times in another sequence during the comparison process. Let the number of repeatedly mapped packets be Nr.

[0079] The preset integrity reference value is a value preset based on an ideal situation of HDMI interface data transmission or historical experience, and is denoted as Nb.

[0080] The data integrity index, D, is calculated by proportionally converting the sum of the number of unsuccessfully mapped and duplicated packets to a preset integrity baseline. First, the total number of unsuccessfully mapped and duplicated packets is calculated as Nt = Nu + Nr. Then, using this proportional relationship, the data integrity index, D, is calculated as 1 - (Nt / Nb). This ratio is subtracted from 1 because a larger value indicates a higher number of problematic packets during data transmission and poorer data integrity. Therefore, subtracting this ratio from 1 yields an indicator of data integrity. A value closer to 1 indicates better data integrity.

[0081] Step S140: Generate a dynamic adjustment signal based on the comparison result between the path verification parameter set and the preset transmission compliance threshold, and map the dynamic adjustment signal to the transmission configuration parameter change queue corresponding to the transmission configuration module of the target device to activate the transmission parameter update operation.

[0082] The path verification parameter set includes the path timing matching degree P, the data integrity index D and the transmission interruption frequency N. The preset transmission compliance thresholds correspond to the first compliance threshold Ph, the second compliance threshold Dh and the third compliance threshold Nh respectively.

[0083] Step S141: When the path timing matching degree is lower than the first compliance threshold, generate a first adjustment instruction, which is used to increase the clock calibration frequency of the target device.

[0084] Compare the path timing matching degree P with the first compliance threshold Ph. If P < Ph, it means that the timing matching degree of the two transmission directions is poor, which may be caused by clock asynchronization. At this time, generate a first adjustment instruction. In the target device, the clock calibration frequency controls the frequency of the device's clock calibration. Increasing the clock calibration frequency can make the device adjust the clock more frequently, thereby improving the timing matching degree of the two transmission directions.

[0085] Step S142: When the data integrity index is lower than the second compliance threshold, generate a second adjustment instruction, which is used to adjust the upper limit of the retransmission times of the data packet retransmission mechanism of the target device.

[0086] Compare the data integrity index D with the second compliance threshold Dh. If D < Dh, it indicates that there are较多 losses or errors in the data during transmission, and the data integrity does not meet the standard. At this time, generate a second adjustment instruction. The data packet retransmission mechanism is set to ensure data integrity. When the receiving party finds that the data packet is incorrect or missing, it will request the sending party to retransmit. Adjusting the upper limit of the retransmission times can optimize the retransmission mechanism according to the data integrity situation. If the data integrity is poor, the upper limit of the retransmission times can be appropriately increased to increase the probability of correct data transmission.

[0087] Step S143: When the transmission interruption frequency is higher than the third compliance threshold, generate a third adjustment instruction, which is used to reduce the transmission rate of the target device and enable the redundant data packet insertion strategy.

[0088] Compare the transmission interruption frequency N with the third compliance threshold Nh. If N > Nh, it means that the interruption situation in the transmission process is relatively frequent, which may be caused by too high transmission rate or insufficient data reliability. At this time, generate a third adjustment instruction. Reducing the transmission rate can reduce the amount of data transmitted per unit time, thereby reducing the pressure during transmission and reducing the occurrence of interruptions. Enabling the redundant data packet insertion strategy is to insert some redundant data packets when transmitting data. These redundant data packets can be used to recover data when the original data packets are lost or damaged, improving the reliability of data transmission.

[0089] Step S144: Generate the dynamic adjustment signal according to the preset dynamic adjustment priority rule, wherein, when the transmission interruption frequency is higher than the third compliance threshold, the execution priority of the third adjustment instruction is higher than the first adjustment instruction and the second adjustment instruction, and when the dynamic adjustment signal is mapped to the transmission configuration parameter change queue, the parameter update operation is executed in priority order.

[0090] The preset dynamic adjustment priority rule specifies the execution order of different adjustment instructions. Because excessive transmission interruption frequency can seriously affect the stability of data transmission, when the transmission interruption frequency exceeds the third compliance threshold, the third adjustment instruction has the highest execution priority.

[0091] When generating a dynamic adjustment signal, the system determines which adjustment instructions need to be generated based on the comparison of the path verification parameter set with the preset transmission compliance threshold. If multiple conditions are met simultaneously, the order of the instructions is determined according to the priority rules. These adjustment instructions are then combined into a dynamic adjustment signal.

[0092] When mapping the dynamic adjustment signal to the transmission configuration parameter change queue corresponding to the target device's transmission configuration module, the adjustment commands are inserted into the queue in order of priority. For example, if the third adjustment command has the highest priority, it is inserted at the head of the queue, followed by the remaining adjustment commands. The transmission configuration module then executes the parameter update operations in the order listed in the queue, adjusting the transmission parameters of the target device.

[0093] Step S150: After the transmission parameter update operation is performed, a secondary verification operation for the updated transmission configuration is triggered, and a final verification result set is generated.

[0094] Step S151: recapture the third updated data packet sequence in the first transmission direction and the fourth updated data packet sequence in the second transmission direction within a preset time interval after the transmission parameter update operation is completed.

[0095] The preset time interval is a pre-set period of time based on the time it takes for the device to stabilize after the transmission parameter update. After the transmission parameter update is complete, wait for the preset time interval to elapse, allowing the device to stabilize under the new transmission configuration.

[0096] Then, use the same data capture module as in step S110 to recapture the data packets in the first transmission direction to form a third data packet sequence P3={p31, p32, p33, ...}, where p3i represents the i-th data packet in the first transmission direction after update; at the same time, capture the data packets in the second transmission direction to form a fourth data packet sequence P4={p41, p42, p43, ...}, where p4j represents the j-th data packet in the second transmission direction after update.

[0097] Step S152: performing time feature alignment processing on the third data packet sequence and the fourth data packet sequence to generate an updated bidirectional transmission time feature set.

[0098] Step S152 is similar to step S120, and its purpose is to align the time characteristics of the updated third data packet sequence and the fourth data packet sequence so as to accurately analyze the time characteristics under the new transmission configuration.

[0099] First, the sender start timestamp and receiver end timestamp of each data packet are extracted from the third data packet sequence P3 to form an updated first initial timestamp set T3. Similarly, the sender start timestamp and receiver end timestamp of each data packet are extracted from the fourth data packet sequence P4 to form an updated second initial timestamp set T4.

[0100] Then, cross-path synchronization processing is performed on T3 and T4 to determine the time reference offset, and compensation calibration is performed to generate updated first calibration timestamp set T3' and second calibration timestamp set T4'.

[0101] Then, an updated first transmission delay sequence τ3 is generated according to the difference between the start timestamp of the transmitter and the end timestamp of the receiver in T3', and an updated second transmission delay sequence τ4 is generated according to the difference between the start timestamp of the transmitter and the end timestamp of the receiver in T4'.

[0102] Finally, windowed statistical analysis is performed on τ3 and τ4 to determine the delay fluctuation range and average delay value within each preset time window. These values ​​are then associated with the corresponding data packet identifiers to generate an updated two-way transmission time feature set F'.

[0103] Step S153: performing path consistency analysis on the updated bidirectional transmission time feature set based on the transmission path verification rule to generate an updated path verification parameter set.

[0104] The same transmission path verification rule as in step S130 is used to perform path consistency analysis on the updated bidirectional transmission time feature set F'.

[0105] First, based on the delay fluctuation range and average delay value corresponding to each data packet identifier in F' within each preset time window, a two-way delay difference vector of each data packet is generated.

[0106] Then, the two-way delay difference vector is segmented into sliding windows, and the cosine similarity between the two-way delay difference vectors of adjacent data packets is calculated to generate a window timing similarity sequence.

[0107] Next, the updated path timing matching degree P' is determined according to the degree of deviation between the window timing similarity sequence and a preset similarity threshold.

[0108] Then, the absolute value of the delay difference between the updated first transmission delay sequence τ3 and the updated second transmission delay sequence τ4 in each delay analysis window is extracted, and the frequency of occurrence of consecutive exceeding threshold intervals is calculated based on the time distribution density function of the absolute value of the delay difference to generate the updated transmission interruption frequency N'.

[0109] Finally, packet identification mapping verification is performed on the third data packet sequence P3 and the fourth data packet sequence P4, and the number of unsuccessfully mapped data packets and the number of repeatedly mapped data packets are counted and proportionally converted to the preset integrity reference value to generate an updated data integrity index D'.

[0110] The updated path verification parameter set includes P', D', and N'.

[0111] Step S154: Calculate the difference between the updated path verification parameter set and the transmission compliance threshold, and determine the optimization efficiency index of the transmission parameter update operation according to the difference.

[0112] Calculate the difference ΔP=P'-Ph between the updated path timing matching degree P' and the first compliance threshold Ph, the difference ΔD=D'-Dh between the updated data integrity index D' and the second compliance threshold Dh, and the difference ΔN=N'-Nh between the updated transmission interruption frequency N' and the third compliance threshold Nh.

[0113] The optimization efficiency index E can be calculated by comprehensively calculating these differences. For example, these differences can be weighted and summed, with weights wp, wd, and wn, respectively. The optimization efficiency index E = wp*ΔP + wd*ΔD + wn*ΔN. The weights can be determined based on the importance of different parameters to transmission performance.

[0114] Step S155: Compare the optimization efficiency index with the preset efficiency evaluation threshold. If the optimization efficiency index reaches the efficiency evaluation threshold, a verification pass signal is generated; if it does not reach the efficiency evaluation threshold, a verification failure signal and a corresponding failure reason code are generated.

[0115] The preset efficiency evaluation threshold is a value preset according to the expected effect of the transmission parameter update operation, and is denoted as Eh.

[0116] The optimization efficiency index E is compared with the efficiency evaluation threshold Eh. If E≥Eh, it means that the transmission parameter update operation has achieved the expected optimization effect, and a verification pass signal is generated.

[0117] If E < Eh, it indicates that the transmission parameter update operation has not achieved the expected effect, and the reasons need to be analyzed. Determine the failure reason based on the comparison result between the updated path verification parameter set and the transmission compliance threshold. For example, if P’ is still lower than Ph, it means that the path timing matching degree is still not up to standard, and the corresponding failure reason code can be set to the code representing the path timing matching problem; if D’ is lower than Dh, it means that there are still problems with data integrity, and the corresponding failure reason code can be set to the code representing the data integrity problem; if N’ is higher than Nh, it means that the transmission interruption frequency is still too high, and the corresponding failure reason code can be set to the code representing the transmission interruption problem.

[0118] Step S156: Associate the verification passed signal or verification failed signal with the transmission configuration parameter change queue to generate the final verification result set.

[0119] Step S1561: If the verification passed signal is generated, extract the currently effective transmission configuration parameter set from the transmission configuration parameter change queue, and associate and store the transmission configuration parameter set with the updated path verification parameter set.

[0120] When the verification passed signal is generated, it indicates that the transmission parameter update operation is effective. At this time, extract the currently effective transmission configuration parameter set from the transmission configuration parameter change queue. The transmission configuration parameter set contains various parameters of the target device under the new transmission configuration, such as clock calibration frequency, upper limit of packet retransmission times, transmission rate, etc.

[0121] Associate and store these transmission configuration parameter sets with the updated path verification parameter set (including P’, D’ and N’). They can be stored in a database or a file for subsequent query and analysis. This can record under which transmission configuration parameters the device can achieve better transmission performance.

[0122] Step S1562: If the verification failed signal is generated, extract the historical transmission configuration parameter set from the transmission configuration parameter change queue, and map the historical transmission configuration parameter set with the failure reason code.

[0123] When the verification failed signal is generated, it is necessary to trace and analyze the failure reason. Extract the historical transmission configuration parameter set from the transmission configuration parameter change queue, and these parameters are the configurations before the transmission parameter update operation.

[0124] Map historical transmission configuration parameter sets to failure cause codes. For example, if the failure cause code indicates a path timing mismatch, the parameters related to clock calibration frequency in the historical transmission configuration are associated with that code. If the failure cause code indicates a data integrity issue, the parameters related to the packet retransmission mechanism are associated with that code. This allows analysis of the relationship between historical configurations and failure causes to identify factors that may have contributed to poor transmission performance.

[0125] Step S1563: Generate a corresponding configuration validation report or configuration rollback report according to the verification pass signal or verification fail signal, and integrate the configuration validation report or configuration rollback report with the final verification result set.

[0126] If the verification is successful, a configuration validation report is generated. This report contains the currently effective transmission configuration parameter set, the updated path validation parameter set, and the verification conclusion. The report can be presented in text or chart format for easy viewing and understanding.

[0127] If the verification signal fails, a configuration rollback report is generated. This report contains the historical transmission configuration parameter set, the failure reason code, and recommended adjustment directions. For example, if the path timing matching problem is caused by an improper clock calibration frequency setting, the configuration rollback report may recommend adjusting the clock calibration frequency appropriately.

[0128] Integrate the configuration validation report or configuration rollback report with the final verification result set. The final verification result set includes the verification pass signal or verification failure signal, the relevant configuration parameter set, and the corresponding report, forming a complete verification result record.

[0129] Step S1564: Send the final verification result set to the log storage module and remote monitoring terminal of the target device to trigger an alarm notification or status synchronization operation.

[0130] Step S1564-1: Encode the verification pass signal or the verification fail signal in the final verification result set to generate a standardized verification result coding sequence.

[0131] To facilitate transmission and storage, the verification pass or failure signals in the final verification result set need to be encoded. A predefined encoding rule can be used to encode the verification pass and failure signals into specific binary sequences. For example, the verification pass signal is encoded as "1" and the verification failure signal is encoded as "0." At the same time, other signal-related information, such as transmission configuration parameters and path verification parameters, is also encoded according to the specified format to form a standardized verification result code sequence.

[0132] Step S1564-2: Bind the verification result coding sequence with the device identifier and verification timestamp of the target device to generate a verification record set with a temporal tag.

[0133] To accurately identify the device and time to which the verification result belongs, the verification result code sequence is bound to the target device's device identifier and verification timestamp. The device identifier uniquely identifies the target device, and the verification timestamp records the specific time when the verification operation occurred. This information is combined to generate a temporally tagged verification record set. This facilitates subsequent querying and analysis of the verification records, allowing for understanding the transmission performance of different devices at different times.

[0134] Step S1564-3: Divide the verification record set into multiple data blocks, and encrypt each data block to generate an encrypted verification data stream.

[0135] Considering the security and privacy of data, the set of verification records with temporal tags is divided into multiple data blocks. The size of the division can be set according to the actual situation, for example, according to a fixed number of bytes.

[0136] Each data block is encrypted. A symmetric encryption algorithm, such as the AES algorithm, can be used to encrypt the data block using a pre-set encryption key. Each encrypted data block forms an encrypted data unit, which is arranged in sequence to generate an encrypted verification data stream. This ensures the security of the verification record during transmission and storage, preventing unauthorized access and tampering.

[0137] Step S1564-4: The encrypted verification data stream is transmitted to the log storage module for distributed storage, and synchronously transmitted to the remote monitoring terminal for real-time analysis.

[0138] The encrypted verification data stream is transmitted to the log storage module of the target device. The log storage module can use a distributed storage system, such as Hadoop Distributed File System (HDFS), to store the encrypted verification data stream across multiple storage nodes. This improves data reliability and availability and prevents data loss due to a single storage node failure.

[0139] Simultaneously, the encrypted verification data stream is transmitted to a remote monitoring terminal. This terminal, which can be a dedicated monitoring server or monitoring software, receives the encrypted verification data stream and decrypts it using the same encryption key, then performs real-time parsing. During parsing, the verification result encoding sequence is converted back to the original verification pass or failure signal, along with the associated transmission configuration parameters and path verification parameters.

[0140] Step S1564-5: In the remote monitoring terminal, a visual verification report is generated based on the parsed encrypted verification data stream, and a corresponding alarm notification is triggered based on the verification pass signal or verification fail signal.

[0141] On the remote monitoring terminal, a visual verification report is generated based on the parsed encrypted verification data stream. This report, in the form of charts or tables, intuitively displays the target device's transmission performance, including path timing matching, data integrity indicators, and transmission interruption frequency. This visual report allows monitoring personnel to quickly understand the device's operating status.

[0142] Alarm notifications are triggered based on verification pass or failure signals. If verification passes, a message indicating normal device operation is displayed on the monitoring interface. If verification fails, an alarm is triggered based on the failure reason code. For example, if there's a path timing mismatch issue, the alarm might prompt adjustments to the clock calibration frequency; if there's a data integrity issue, the alarm might prompt a review of the packet retransmission mechanism. This allows timely notification of equipment maintenance and adjustments, ensuring stable and reliable data transmission.

[0143] In a possible implementation, after step S150, the method further includes:

[0144] Step S210: Obtain the two-way transmission time feature set stored in the historical verification process and the corresponding final verification result set, wherein the two-way transmission time feature set includes the delay fluctuation range, average delay value and delay distribution statistics of the data packet identifier within each preset time window, and the final verification result set includes a binary classification label of a verification pass signal or a verification fail signal.

[0145] The two-way transmission time feature set and the final verification result set stored in the historical verification process are obtained from the log storage module or related database of the target device. The two-way transmission time feature set is obtained by performing time feature alignment and windowed statistical analysis on the data packet sequence during the previous verification process. It includes the delay fluctuation range, average delay value and delay distribution statistics of the data packet identifier within each preset time window. The final verification result set records the results of each verification in the form of a binary classification label of a verification pass signal or a verification fail signal. These historical data are the basis for the subsequent construction and training of the classification model, and can reflect the feature information under different transmission states and the corresponding verification results.

[0146] Step S220: perform feature extraction and normalization on the bidirectional transmission time feature set, calculate the difference between the maximum and minimum values ​​of the delay fluctuation range in each time window as the first feature dimension, count the standard deviation of the delay values ​​of all data packets in the time window as the second feature dimension, calculate the sliding difference of the average delay values ​​of adjacent time windows as the third feature dimension, and generate a normalized multi-dimensional time series feature vector.

[0147] After obtaining the bidirectional transmission time feature set, it is necessary to extract valuable feature information from it and perform normalization processing to ensure comparability between different feature dimensions.

[0148] First, for each time window, calculate the difference between the maximum and minimum values ​​of the latency fluctuation range and use this as the first feature dimension. Let the maximum value of the latency fluctuation range in the i-th time window be Max_i and the minimum value be Min_i. Then the value of the first feature dimension F1_i = Max_i - Min_i. This feature reflects the fluctuation amplitude of packet transmission latency within that time window.

[0149] Next, the standard deviation of all packet delay values ​​in each time window is calculated as the second characteristic dimension. The standard deviation can measure the degree of dispersion of the packet delay value relative to the average value. Assume that the delay values ​​of all packets in the i-th time window are {T1_i, T2_i, ..., Tn_i}, and their average value is Avg_i. Then the value F2_i of the second characteristic dimension is obtained by calculating the standard deviation of these delay values. The calculation process of the standard deviation is to first calculate the square of the difference between each delay value and the average value, then calculate the average of these square values, and finally take the square root. The value F2_i of the second characteristic dimension reflects the stability of the packet delay.

[0150] Next, we calculate the sliding difference of the average delay values ​​of adjacent time windows as the third characteristic dimension. Let the average delay value of the i-th time window be Avg_i, and the average delay value of the i+1-th time window be Avg_(i+1), then the value of the third characteristic dimension F3_i = Avg_(i+1) - Avg_i. This third characteristic dimension value F3_i reflects the change in the average delay between adjacent time windows.

[0151] Combine these three feature dimensions of each time window to form a three-dimensional feature vector F_i=(F1_i, F2_i, F3_i).

[0152] In order to make different feature dimensions comparable, these feature vectors need to be normalized. Normalization can be done using common methods, such as minimum-maximum normalization. For each feature dimension, find the maximum value Max_all and the minimum value Min_all of the feature dimension in all time windows. For the j-th feature dimension value Fij in the i-th time window, the normalized feature dimension value Fij_normalized = (Fij-Min_all) / (Max_all-Min_all). After normalization, the value of each feature dimension is mapped to the interval [0, 1], thus generating a normalized multidimensional time series feature vector.

[0153] Step S230: Encode the verification pass signal in the final verification result set as a first numerical label, and encode the verification fail signal as a second numerical label, and align them with the multidimensional time series feature vector in a time window order to generate a training sample set.

[0154] In order to combine the final verification result set with the multi-dimensional time series feature vector for model training, the verification results need to be encoded. The verification pass signal is encoded as a first numerical label, such as 1; the verification fail signal is encoded as a second numerical label, such as 0.

[0155] Following the order of the time windows, the encoded numerical labels are aligned with the corresponding multidimensional time series feature vectors. That is, the multidimensional time series feature vector F_i of the i-th time window corresponds to the encoded verification result label L_i of the i-th time window. Combining all such feature vector and label pairs generates a training sample set. The training sample set can be represented as {(F_1, L_1), (F_2, L_2), …, (F_n, L_n)}, where n is the number of time windows. This training sample set contains feature information and corresponding classification labels and serves as input data for subsequent classification model training.

[0156] Step S240: Construct a classification model including a temporal convolutional layer and a fully connected layer, wherein the convolution kernel size of the temporal convolutional layer matches the division interval of the time window, and the input dimension of the fully connected layer is consistent with the dimension of the multidimensional time series feature vector.

[0157] The constructed classification model mainly consists of temporal convolutional layers and fully connected layers. This structure can effectively process time series features and perform classification predictions.

[0158] The primary function of the temporal convolution layer is to extract local temporal features. The size of the convolution kernel must match the time window interval. This allows the convolution operation to effectively extract features along the temporal dimension. In the temporal convolution layer, multiple convolution kernels of varying sizes are set, each sliding along the temporal dimension to capture local delay fluctuations of varying lengths. The convolution kernel step size is set to the same as the sliding step size of the time window to ensure that the convolution operation covers the entire time series.

[0159] As the convolution kernel slides along the time dimension, it performs a convolution operation on the input multidimensional time series feature vector, generating multiple feature maps. Each feature map represents local feature information at a specific scale. These feature maps are concatenated along the channel dimension to generate a multi-scale fused feature matrix. This multi-scale fused feature matrix contains local delay fluctuation feature information at different scales.

[0160] Next, the multi-scale fusion feature matrix is ​​divided into multiple sub-feature blocks according to the time window sequence. A maximum pooling operation is performed on each sub-feature block. This operation extracts significant eigenvalues ​​within each sub-feature block, thereby reducing the feature dimension and generating a reduced-dimensional time series feature vector. The maximum pooling operation can highlight important features in each sub-feature block and remove some redundant information.

[0161] The fully connected layer's input dimensions match those of the multidimensional time series feature vector. Its function is to further transform and classify the reduced time series feature vector. In the fully connected layer, the time series feature vector is linearly transformed with a preset weight matrix and then mapped into a hidden feature vector using an activation function. The activation function introduces nonlinearity, enhancing the model's expressiveness.

[0162] At the output layer of the classification model, the hidden feature vector is converted into a binary probability value. By mapping the hidden feature vector, a probability value is obtained, which represents the probability that the sample belongs to the verified category. A threshold is preset. If the probability value exceeds the threshold, it is judged as a verification pass signal; otherwise, it is judged as a verification failure signal.

[0163] Step S250: Input the training sample set into the classification model, calculate the cross entropy loss between the predicted label and the true label through forward propagation, and use the gradient descent algorithm to update the model parameters until the loss function converges.

[0164] The previously generated training sample set is input into the constructed classification model for training. During the training process, forward propagation is performed first. Forward propagation involves passing the feature vectors in the training sample set from the input layer through the temporal convolution layer and the fully connected layer, ultimately obtaining the model's predicted label.

[0165] For each training example (F_i, L_i), the classification model outputs a predicted binary probability value. This predicted probability value is compared with the true label L_i to calculate the cross-entropy loss. Cross-entropy loss is a commonly used metric to measure the difference between the classification model's prediction and the true label, which can reflect the accuracy of the model's prediction.

[0166] After calculating the cross-entropy loss, the gradient descent algorithm is used to update the model parameters. The core idea of ​​the gradient descent algorithm is to gradually decrease the loss function by calculating the gradient of the loss function with respect to the model parameters and then updating the parameters in the opposite direction of the gradient. Specifically, the partial derivatives of the loss function with respect to each parameter in the model are first calculated. These partial derivatives form the gradient vector. The model parameters are then updated in the opposite direction of the gradient according to the preset learning rate. The learning rate controls the step size of each parameter update. Excessively large learning rates may prevent the model from convergence, while too small learning rates will slow the training process.

[0167] The process of forward propagation, loss calculation, and parameter update is repeated until the loss function converges. Convergence of the loss function means that the difference between the model's predictions and the true labels has reached a small and stable value. At this point, the model parameters are optimized and can better classify the input features.

[0168] Step S260: During the real-time verification process, the currently captured bidirectional transmission time feature set is processed into a real-time multi-dimensional time series feature vector, which is input into the trained classification model to output a predicted label. If the predicted label is a verification failure signal, the pre-adjustment process is triggered.

[0169] During the real-time verification process, after obtaining the currently captured bidirectional transmission time feature set, it needs to be processed into a real-time multidimensional time series feature vector in the same format as the training sample set. The processing process is similar to step S220, that is, the difference between the maximum and minimum values ​​of the delay fluctuation range in each time window is calculated as the first feature dimension, the standard deviation of the delay values ​​of all packets in the time window is calculated as the second feature dimension, and the sliding difference of the average delay values ​​of adjacent time windows is calculated as the third feature dimension. Then, normalization is performed to generate a real-time multidimensional time series feature vector.

[0170] The generated real-time multi-dimensional time series feature vector is input into the trained classification model. The classification model performs forward propagation based on the input feature vector and outputs a predicted label. If the predicted label indicates a verification failure, there may be a problem with the current transmission status, and the pre-adjustment process needs to be triggered.

[0171] For example, in a further embodiment, step S260 may include:

[0172] Step S261: According to the verification failure prediction result output by the classification model, extract the multi-dimensional time series feature vector in the corresponding time window, and analyze the degree of numerical deviation of each feature dimension.

[0173] When the classification model outputs a verification failure prediction result, further analysis of the cause of the failure is required. First, a multidimensional time series feature vector is extracted for the corresponding time window. This multidimensional time series feature vector contains three characteristic dimensions within the time window: the difference between the maximum and minimum values ​​of the delay fluctuation range, the standard deviation of the packet delay values, and the sliding difference of the average delay values ​​of adjacent time windows.

[0174] Analyze the degree of numerical deviation for each feature dimension. The degree of deviation can be determined by comparing the numerical values ​​of these feature dimensions with the average or median of similar feature dimensions in the training sample set. For example, calculate the difference between the value of the first feature dimension within the time window and the average value of the first feature dimension in the training sample set. The size of this difference reflects the degree of numerical deviation for that feature dimension. Similarly, perform a similar analysis for the second and third feature dimensions.

[0175] Step S262: Calculate the deviation of the delay fluctuation range difference, the deviation of the delay standard deviation, and the deviation of the delay sliding difference to generate a deviation coefficient vector of each feature dimension.

[0176] For the delay fluctuation range difference, calculate its deviation. The deviation of the delay fluctuation range difference can be calculated by dividing the absolute value of the difference between the delay fluctuation range difference within the time window and the average delay fluctuation range difference in the training sample set by the standard deviation of the delay fluctuation range difference in the training sample set. This deviation measures the degree of deviation of the delay fluctuation range difference within the time window from the training sample set.

[0177] Similarly, the deviations of the delay standard deviation and the delay sliding differential are calculated separately. The deviations of these three characteristic dimensions are combined to form a deviation coefficient vector for each characteristic dimension. For example, the deviation coefficient vector D = (D1, D2, D3), where D1 is the deviation of the delay fluctuation range difference, D2 is the deviation of the delay standard deviation, and D3 is the deviation of the delay sliding differential.

[0178] Step S263: Match the deviation coefficient vector with the preset adjustment strategy mapping table to determine the transmission parameter type and adjustment direction that need to be adjusted first. When the delay fluctuation range difference deviation is the highest, it is mapped to the clock calibration parameter adjustment; when the delay standard deviation deviation is the highest, it is mapped to the data retransmission parameter adjustment.

[0179] The preset adjustment strategy mapping table is pre-set based on a large number of experiments and experiences, and records the correspondence between the deviation of different feature dimensions and the transmission parameter adjustment strategy.

[0180] Match the deviation coefficient vectors for each feature dimension with the adjustment strategy mapping table. If the deviation D1 of the delay fluctuation range difference is the highest in the deviation coefficient vector, adjust the clock calibration parameters first according to the adjustment strategy mapping table. Because a large delay fluctuation range difference may be caused by clock asynchrony, adjusting the clock calibration parameters can improve clock synchronization and reduce delay fluctuation.

[0181] If the delay standard deviation (D2) is the highest, prioritize adjusting the data retransmission parameters. A large delay standard deviation indicates a high degree of dispersion in packet delays, potentially indicating packet loss or delayed retransmissions. Adjusting the data retransmission parameters can optimize the retransmission mechanism and improve data transmission stability.

[0182] Step S264: Generate a parameter pre-adjustment instruction including the adjustment type, direction and amplitude, and insert the parameter pre-adjustment instruction into the head of the queue to be executed of the transmission configuration module.

[0183] Based on the transmission parameter types and adjustment directions that require priority adjustment, a parameter pre-adjustment instruction is generated. This instruction includes the adjustment type (e.g., clock calibration parameter adjustment, data retransmission parameter adjustment), the adjustment direction (whether to increase or decrease the parameter value), and the adjustment magnitude (the specific magnitude of the adjustment parameter value).

[0184] For example, if it is determined that the clock calibration parameters need to be adjusted, the adjustment direction is to increase, and the adjustment amplitude is a preset step value, then the parameter pre-adjustment instruction will clearly record this information.

[0185] Insert the generated parameter pre-adjustment instruction into the head of the transmission configuration module's pending queue. The transmission configuration module will execute the parameter adjustment operations in the order of the pending queue. Placing the parameter pre-adjustment instruction at the head of the queue ensures that it is executed first, thereby adjusting the transmission status as quickly as possible.

[0186] Step S265: After the parameter pre-adjustment instruction is executed, the updated bidirectional transmission time feature set is re-collected to generate a real-time multi-dimensional time series feature vector, and after sliding window mean normalization processing, the normalized feature vector is input into the classification model for secondary verification result prediction.

[0187] After executing the parameter pre-adjustment command, wait for a while to allow the device to stabilize under the new transmission parameter configuration. Then, re-collect the updated bidirectional transmission time feature set. The collection process is the same as the previous method for obtaining the bidirectional transmission time feature set. A data capture device is set up on the data transmission line of the HDMI interface to record the transmission information of the data packet.

[0188] The collected, updated bidirectional transmission time feature set is processed into a real-time multidimensional time series feature vector using the same process as before. To make the feature vectors more comparable, sliding window mean normalization is applied. Sliding window mean normalization calculates the mean and standard deviation of the feature dimension within a sliding window and then normalizes the feature dimension values ​​within that sliding window. This reduces data fluctuations and makes the feature vector more stable.

[0189] The normalized feature vector is input into the trained classification model for a second validation result prediction. The model will output a predicted label based on the input feature vector to determine whether the adjusted transmission status has improved.

[0190] Step S266: If the secondary prediction result is still a verification failure signal, the adjustment amplitude of the parameter pre-adjustment instruction is iteratively updated according to the latest deviation coefficient vector until the prediction result turns into a verification pass signal or reaches the maximum iteration number threshold.

[0191] If the secondary prediction result still indicates a verification failure, it means that the current parameter adjustment has not achieved the expected effect and further adjustment is required. Based on the latest calculated deviation coefficient vector, the adjustment range of the parameter pre-adjustment instruction is iteratively updated.

[0192] For example, if it is previously determined that the clock calibration parameters need to be adjusted, and the adjustment amplitude is a preset step value, after the secondary prediction fails, the adjustment amplitude can be appropriately increased or decreased according to the latest deviation coefficient vector.

[0193] Repeat this process—adjusting parameters, recollecting data, generating feature vectors, and making predictions—until the predicted label output by the classification model becomes a verification pass signal, indicating that the transmission status has improved. Or, if the maximum iteration threshold is reached, meaning that multiple adjustments still fail to convert the prediction result into a verification pass signal, adjustments should be stopped. Further analysis of the cause or other methods may be needed to resolve the transmission issue.

[0194] Figure 2 A schematic diagram illustrating exemplary hardware and software components of a system 100 for verifying bidirectional transmission of interface data for HDMI, which can implement the concepts of the present application, is provided in some embodiments of the present application. For example, the processor 120 can be used in the system 100 for verifying bidirectional transmission of interface data for HDMI and perform the functions of the present application.

[0195] The interface data bidirectional transmission verification system 100 for HDMI can be a general-purpose server or a special-purpose server, both of which can be used to implement the interface data bidirectional transmission verification method for HDMI of the present application. Although only one server is shown in this application, for convenience, the functions described in this application can be implemented in a distributed manner on multiple similar platforms to balance the processing load.

[0196] For example, the interface data bidirectional transmission verification system 100 for HDMI may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and storage media 140 of different forms, such as a disk, ROM, or RAM, or any combination thereof. Exemplarily, the interface data bidirectional transmission verification system 100 for HDMI may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present application may be implemented according to these program instructions. The interface data bidirectional transmission verification system 100 for HDMI also includes an I / O interface 150 between a computer and other input and output devices.

[0197] For ease of explanation, only one processor is described in the interface data bidirectional transmission verification system 100 for HDMI. However, it should be noted that the interface data bidirectional transmission verification system 100 for HDMI in the present application may also include multiple processors, so the steps performed by one processor described in the present application may also be performed jointly or individually by multiple processors. For example, if the processor of the interface data bidirectional transmission verification system 100 for HDMI performs step A and step B, it should be understood that step A and step B may also be performed jointly by two different processors or individually in one processor. For example, the first processor performs step A, the second processor performs step B, or the first processor and the second processor perform steps A and B together.

[0198] In addition, an embodiment of the present invention further provides a readable storage medium having computer executable instructions preset therein. When a processor executes the computer executable instructions, the above interface data bidirectional transmission verification method for HDMI is implemented.

[0199] It should be noted that in order to simplify the description of the present invention and thus help understand one or more embodiments of the invention, in the foregoing description of the embodiments of the present invention, multiple features are sometimes combined into one embodiment, figure or description thereof.

Claims

1. A method for verifying bidirectional transmission of HDMI interface data, characterized in that: The method comprises: Obtaining a bidirectional transmission data set of a target device, the bidirectional transmission data set comprising a first data packet sequence captured from a first transmission direction and a second data packet sequence captured from a second transmission direction, wherein the first transmission direction and the second transmission direction are mutually opposite transmission paths in an HDMI interface; Performing time feature alignment processing on the first data packet sequence and the second data packet sequence to generate a bidirectional transmission time feature set; Performing path consistency analysis on the bidirectional transmission time feature set based on a preset transmission path verification rule to generate a path verification parameter set for the first transmission direction and the second transmission direction, the path verification parameter set including a path timing matching degree, a data integrity index, and a transmission interruption frequency; generating a dynamic adjustment signal according to a comparison result between the path verification parameter set and a preset transmission compliance threshold, and mapping the dynamic adjustment signal to a transmission configuration parameter change queue corresponding to a transmission configuration module of the target device to activate a transmission parameter update operation; After the transmission parameter update operation is performed, a secondary verification operation for the updated transmission configuration is triggered, and a final verification result set is generated.

2. The method for verifying bidirectional transmission of interface data for HDMI according to claim 1, wherein: The performing time feature alignment processing on the first data packet sequence and the second data packet sequence to generate a bidirectional transmission time feature set includes: Extracting a first initial timestamp set from the first data packet sequence, the first initial timestamp set including a sending end start timestamp and a receiving end end timestamp of each first data packet; Extracting a second initial timestamp set from the second data packet sequence, the second initial timestamp set including a sending end start timestamp and a receiving end end timestamp of each second data packet; performing cross-path synchronization processing on the first initial timestamp set and the second initial timestamp set, determining a time base offset between the first data packet sequence and the second data packet sequence, and performing compensation calibration on the first initial timestamp set and the second initial timestamp set based on the time base offset to generate a calibrated first calibrated timestamp set and a calibrated second calibrated timestamp set; generating a first transmission delay sequence according to a difference between a start timestamp of the transmitting end and an end timestamp of the receiving end in the first calibration timestamp set, and generating a second transmission delay sequence according to a difference between a start timestamp of the transmitting end and an end timestamp of the receiving end in the second calibration timestamp set; Windowed statistical analysis is performed on the first transmission delay sequence and the second transmission delay sequence to determine the delay fluctuation range and average delay value within each preset time window, and the delay fluctuation range and average delay value are associated with the corresponding data packet identifier to generate the bidirectional transmission time feature set.

3. The method for verifying bidirectional transmission of interface data for HDMI according to claim 2, wherein: The performing cross-path synchronization processing on the first initial timestamp set and the second initial timestamp set to determine a time base offset between the first data packet sequence and the second data packet sequence includes: Filtering a first reference timestamp pair that meets a preset time interval condition from the first initial timestamp set, the first reference timestamp pair comprising a sending end start timestamp and a corresponding receiving end end timestamp of two consecutive first data packets; Filtering a second reference timestamp pair that meets the preset time interval condition from the second initial timestamp set, the second reference timestamp pair including a sending end start timestamp and a corresponding receiving end end timestamp of two consecutive second data packets; Calculating a first difference between the sender start timestamps of two consecutive first data packets in the first reference timestamp pair, and calculating a second difference between the sender start timestamps of two consecutive second data packets in the second reference timestamp pair; determining a clock frequency difference coefficient between the first transmission direction and the second transmission direction according to a ratio between the first difference and the second difference; performing frequency normalization processing on the first initial timestamp set and the second initial timestamp set based on the clock frequency difference coefficient to generate a third timestamp set and a fourth timestamp set after frequency synchronization; The time base offset is determined according to an offset between a sending end start timestamp of the same data packet identifier in the third timestamp set and the fourth timestamp set.

4. The method for verifying bidirectional transmission of interface data for HDMI according to claim 2, wherein: The performing path consistency analysis on the bidirectional transmission time feature set based on a preset transmission path verification rule to generate a path verification parameter set for the first transmission direction and the second transmission direction includes: Generate a two-way delay difference vector for each data packet based on the delay fluctuation range and average delay value corresponding to each data packet identifier in the two-way transmission time feature set within each preset time window, wherein the two-way delay difference vector includes the ratio of the delay fluctuation range to the average delay corresponding to the same data packet in the two-way transmission path, wherein the delay fluctuation range is the difference between the maximum and minimum delays of multiple transmissions of the same data packet within the preset time window; Performing sliding window segmentation processing on the two-way delay difference vector to obtain multiple delay analysis windows, and calculating the cosine similarity between the two-way delay difference vectors of adjacent data packets in each delay analysis window to generate a window timing similarity sequence; Determining the path timing matching degree according to the degree of deviation between the window timing similarity sequence and a preset similarity threshold; Extracting the absolute value of the delay difference between the first transmission delay sequence and the second transmission delay sequence in each delay analysis window, and calculating the frequency of occurrence of consecutive exceeding-threshold intervals based on a time distribution density function of the absolute value of the delay difference to generate the transmission interruption frequency; Perform packet identifier mapping verification on the first data packet sequence and the second data packet sequence, count the number of data packets that are not successfully mapped and the number of data packets that are repeatedly mapped in the first data packet sequence and the second data packet sequence, and proportionally convert the number of data packets with a preset integrity reference value to generate the data integrity index.

5. The method for verifying bidirectional transmission of interface data for HDMI according to claim 1, wherein: The step of generating a dynamic adjustment signal based on a comparison result between the path verification parameter set and a preset transmission compliance threshold, and mapping the dynamic adjustment signal to a transmission configuration parameter change queue corresponding to a transmission configuration module of the target device to activate a transmission parameter update operation includes: When the path timing matching degree is lower than a first compliance threshold, generating a first adjustment instruction, wherein the first adjustment instruction is used to increase a clock calibration frequency of the target device; When the data integrity indicator is lower than a second compliance threshold, generating a second adjustment instruction, wherein the second adjustment instruction is used to adjust an upper limit on the number of retransmissions of a data packet retransmission mechanism of the target device; When the transmission interruption frequency is higher than a third compliance threshold, generating a third adjustment instruction, wherein the third adjustment instruction is used to reduce the transmission rate of the target device and enable a redundant data packet insertion strategy; The dynamic adjustment signal is generated according to a preset dynamic adjustment priority rule, wherein, when the transmission interruption frequency is higher than a third compliance threshold, the execution priority of the third adjustment instruction is higher than the first adjustment instruction and the second adjustment instruction, and when the dynamic adjustment signal is mapped to the transmission configuration parameter change queue, the parameter update operation is executed in order of priority.

6. The method for verifying bidirectional transmission of interface data for HDMI according to claim 1, wherein: After the transmission parameter update operation is executed, triggering a secondary verification operation for the updated transmission configuration and generating a final verification result set, including: Recapturing the third data packet sequence after the first transmission direction is updated and the fourth data packet sequence after the second transmission direction is updated within a preset time interval after the transmission parameter update operation is completed; Performing time feature alignment processing on the third data packet sequence and the fourth data packet sequence to generate an updated bidirectional transmission time feature set; Performing path consistency analysis on the updated bidirectional transmission time feature set based on the transmission path verification rule to generate an updated path verification parameter set; Calculating a difference between the updated path verification parameter set and the transmission compliance threshold, and determining an optimization efficiency index of the transmission parameter update operation according to the difference; Comparing the optimization efficiency index with a preset efficiency evaluation threshold, generating a verification pass signal if the optimization efficiency index reaches the efficiency evaluation threshold; generating a verification fail signal and a corresponding failure reason code if the optimization efficiency index does not reach the efficiency evaluation threshold; The verification pass signal or the verification fail signal is associated with the transmission configuration parameter change queue to generate the final verification result set.

7. The method for verifying bidirectional transmission of interface data for HDMI according to claim 6, wherein: The associating the verification pass signal or the verification fail signal with the transmission configuration parameter change queue to generate the final verification result set includes: If the verification pass signal is generated, extracting the currently effective transmission configuration parameter set from the transmission configuration parameter change queue, and associating the transmission configuration parameter set with the updated path verification parameter set for storage; If the verification failure signal is generated, extracting a historical transmission configuration parameter set from the transmission configuration parameter change queue, and mapping the historical transmission configuration parameter set to the failure reason code; Generate a corresponding configuration validation report or configuration rollback report according to the verification pass signal or verification fail signal, and integrate the configuration validation report or configuration rollback report with the final verification result set; The final verification result set is sent to the log storage module and the remote monitoring terminal of the target device to trigger an alarm notification or a status synchronization operation.

8. The method for verifying bidirectional transmission of interface data for HDMI according to claim 7, wherein: The sending of the final verification result set to the log storage module and the remote monitoring terminal of the target device to trigger an alarm notification or a status synchronization operation includes: Encoding the verification pass signal or the verification fail signal in the final verification result set to generate a standardized verification result coding sequence; Binding the verification result code sequence with the device identifier and verification timestamp of the target device to generate a verification record set with a temporal tag; Splitting the verification record set into multiple data blocks, and encrypting each data block to generate an encrypted verification data stream; Transmitting the encrypted verification data stream to the log storage module for distributed storage, and synchronously transmitting it to the remote monitoring terminal for real-time analysis; In the remote monitoring terminal, a visual verification report is generated according to the parsed encrypted verification data stream, and a corresponding alarm notification is triggered according to the verification pass signal or the verification fail signal.

9. The method for verifying bidirectional transmission of interface data for HDMI according to any one of claims 1 to 8, wherein: The method further comprises: Obtaining a set of bidirectional transmission time features stored in the historical verification process and a corresponding final verification result set, wherein the bidirectional transmission time feature set includes the delay fluctuation range, average delay value, and delay distribution statistics of the data packet identifier within each preset time window, and the final verification result set includes a binary classification label of a verification pass signal or a verification fail signal; Perform feature extraction and normalization on the bidirectional transmission time feature set, calculate the difference between the maximum and minimum values ​​of the delay fluctuation range in each time window as the first feature dimension, calculate the standard deviation of the delay values ​​of all data packets in the time window as the second feature dimension, calculate the sliding difference of the average delay values ​​of adjacent time windows as the third feature dimension, and generate a normalized multidimensional time series feature vector; Encoding the verification pass signal in the final verification result set as a first numerical label and the verification fail signal as a second numerical label, and aligning them with the multidimensional time series feature vector in a time window order to generate a training sample set; Constructing a classification model including a temporal convolutional layer and a fully connected layer, wherein the convolution kernel size of the temporal convolutional layer matches the division interval of the time window, and the input dimension of the fully connected layer is consistent with the dimension of the multidimensional time series feature vector; Input the training sample set into the classification model, calculate the cross entropy loss between the predicted label and the true label through forward propagation, and use the gradient descent algorithm to update the model parameters until the loss function converges; During the real-time verification process, the currently captured bidirectional transmission time feature set is processed into a real-time multi-dimensional time series feature vector, which is input into the trained classification model to output a predicted label. If the predicted label is a verification failure signal, the pre-adjustment process is triggered.

10. The method for verifying bidirectional transmission of interface data for HDMI according to claim 9, wherein: The construction of a classification model including a temporal convolutional layer and a fully connected layer includes: Setting a plurality of convolution kernels of different scales in the temporal convolution layer, each convolution kernel slides along the time dimension to extract local delay fluctuation features, wherein the step size of each convolution kernel is the same as the sliding step size of the time window; splicing the output feature maps of the temporal convolution layer along the channel dimension to generate a multi-scale fusion feature matrix, and dividing the multi-scale fusion feature matrix into multiple sub-feature blocks in order of the time window; Perform a maximum pooling operation on each sub-feature block to extract the significant feature values ​​within each time window and generate a time series feature vector after dimensionality reduction; In the fully connected layer, the temporal feature vector is linearly transformed with a preset weight matrix and mapped into a hidden feature vector through an activation function; At the output layer of the classification model, the hidden feature vector is converted into a binary classification probability value, wherein the category with a probability value higher than a preset threshold is determined as a verification pass signal, otherwise it is a verification fail signal; During the training process, a sliding window incremental update strategy is adopted to dynamically expand the training sample set according to real-time verification results.

Images