Video doorbell network optimization and fault self-recovery system

Abstract

The present application relates to a video doorbell network optimization and fault self-recovery system, comprising: a video frame signature generation module, which computes the color distribution, edge information and texture features of each video frame on the basis of frame content feature encoding, establishes a composite signature in combination with a hash value, generates frame signature data, and creates, on the basis of the frame signature data, integrity signature information before a video frame is sent.

Description

A video doorbell network optimization and fault self-recovery system

[0001] Related applications

[0002] This application claims priority to Chinese patent application No. 202410926360.8, filed on July 11, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application relates to the field of fault self-healing technology, and in particular to a video doorbell network optimization and fault self-recovery system. Background Technology

[0004] The field of fault self-healing technology involves developing and implementing systems, networks, or applications to automatically diagnose problems and restore normal operation in the event of a failure. This technology typically includes fault detection, problem diagnosis, decision support systems, and the automated execution of remedial measures. Its aim is to reduce system downtime and improve system reliability and efficiency. Fault self-healing technology is particularly important in various scenarios, such as critical infrastructure, large-scale data centers, and commercial applications requiring high availability. By using artificial intelligence, machine learning algorithms, and sophisticated event processing systems, fault self-healing systems can predict potential failures, respond to anomalies in real time, and restore service without human intervention.

[0005] A video doorbell network optimization and fault self-recovery system primarily refers to enhancing the network performance and stability of video doorbells using fault self-healing technology. This system is designed to automatically detect and resolve problems in the network connection of video doorbell devices, ensuring smooth and reliable communication within the doorbell system. Its main uses include optimizing data transmission, reducing video stream latency, improving image quality, and quickly restoring service in the event of network failures. This is crucial for ensuring users can promptly receive and view video footage of visitors at their door, especially in security monitoring and home automation systems.

[0006] Existing technologies rely on manual intervention or limited automated repair strategies, which may fail to respond promptly to data corruption or loss. This is particularly problematic in critical infrastructure and high-availability commercial applications, where system downtime or data inconsistencies can have severe consequences. For example, in large-scale data centers experiencing network failures, delayed detection and repair can lead to the inaccessibility or corruption of large amounts of data, impacting service continuity and customer trust. Furthermore, traditional systems are often unable to effectively adapt to rapid changes in network conditions, resulting in degraded video streaming and data transmission quality, especially under network congestion or instability. Summary of the Invention

[0007] The purpose of this application is to address the shortcomings of existing technologies by proposing a video doorbell network optimization and fault self-recovery system.

[0008] To achieve the above objectives, this application adopts the following technical solution: A video doorbell network optimization and fault self-recovery system includes:

[0009] S1: The video frame signature generation module calculates the color distribution, edge information, and texture features of each video frame based on the frame content feature encoding, and establishes a composite signature by combining the hash value to generate frame signature data. Based on the frame signature data, it creates integrity signature information before the video frame is sent.

[0010] S2: Based on the integrity signature information, the video frame signature verification module recalculates the signature of the received video frame at the receiving end, compares it with the signature at the sending end, records the information of mismatched frames, and establishes the signature comparison result.

[0011] S3: Based on the signature comparison result, the network fault repair request module determines whether the error type is data corruption or data loss, selects to resend the request or apply forward error correction technology for local repair, calls additional redundant data, establishes retransmission or repair instructions, executes data packet retransmission or local error repair, and generates a repaired video frame.

[0012] S4: The network status dynamic adjustment module, based on the data integrity of the repaired video frames, combined with bandwidth changes, packet loss rate, and latency indicators, creates adjustment parameters, adjusts the parameters of forward error correction and the size of data packets, optimizes network resource usage, and obtains network optimization results.

[0013] As a further aspect of this application, the steps for obtaining the integrity signature information are specifically as follows:

[0014] S111: Collect the raw color data of each video frame, extract color frequencies using the color histogram method, and apply the normalized color frequency formula:

[0015] Calculate the normalized frequency of each color to obtain a normalized color frequency array;

[0016] Among them, C i The normalized frequency of a color, n i α is the number of times color i appears, N is the total number of pixels, α is the weighted color frequency coefficient, β is the normalized offset, and S is the number of times color i appears. i It is the standard deviation of color frequency;

[0017] S112: Analyze the edge information of the video frames, apply the Sobel operator for edge detection, and use the edge strength formula:

[0018] Calculate the edge intensity of pixel (x,y) to obtain the edge intensity matrix;

[0019] Among them, E x,y G represents the edge intensity of pixel (x, y). x G y These represent the gradients in the x and y directions, respectively, and λ and μ are the gradients corresponding to G. x and G y The weighting coefficients of the gradient;

[0020] S113: Extract texture features from video frames, calculate texture contrast using the gray-level co-occurrence matrix, and apply the texture contrast formula:

[0021] Calculate the texture complexity of the entire image and generate texture contrast values;

[0022] Where T represents the contrast of texture features, p i,j γ is the co-occurrence probability of gray levels i and j, γ enhances the sensitivity of contrast to differences, and Ω is used for normalization calculations.

[0023] S114: Combining the normalized color frequency array, the edge intensity matrix, and the texture contrast value, the feature formula is applied:

[0024] Generate frame signature data and establish integrity signature information before sending video frames;

[0025] Where H represents integrity signature information, · represents element-wise multiplication of the array, hash is a hash function used to generate a unique identifier, and ρ is a normalization parameter for overall feature weighting.

[0026] As a further aspect of this application, the steps for obtaining the signature comparison result are as follows:

[0027] S211: Based on the integrity signature information, after receiving video frame data at the receiving end, extract the normalized color frequency array, edge intensity matrix, and texture contrast value, and use the signature calculation formula:

[0028] The recalculated signature data is obtained;

[0029] Where H′ represents the recalculated signature data, C′, E′, and T′ are the normalized color frequency array, edge intensity matrix, and texture contrast value of the received video frame, respectively, κ, ε, and τ are adjustment constants added for each feature, ρ′ is the normalization constant, and ξ is the adjustment parameter.

[0030] S212: Compare the recalculated signature data with the sender's signature, using the difference measurement formula:

[0031] Calculate the difference between the signatures of the sender and receiver, and generate signature difference data;

[0032] Where Δ represents the signature difference data, and λ is the normalization factor for the difference measure;

[0033] S213: Based on the signature difference data, identify mismatched frames and use the logical filtering formula: R={f∣Δ(f)>θ}

[0034] Generate signature comparison results;

[0035] Where R represents the signature comparison result, f is the frame index, Δ(f) is the signature difference degree of each frame, and θ is the decision threshold.

[0036] As a further aspect of this application, the step of obtaining the repaired video frame specifically includes:

[0037] S311: Based on the signature comparison result, determine the error type, use a decision algorithm to determine whether it is data corruption or data loss, and apply the error judgment formula: E=α·Δ 3 +β·log(Δ+γ) 2 +δ·Δ

[0038] Calculate erroneous decision indicators and generate error type decision data;

[0039] Where E represents the error decision index, Δ represents the signature discrepancy, and α, β, γ, and δ moderate the impact of the discrepancy on the error decision.

[0040] S312: Based on the error type decision data, select a matching recovery strategy using a selection function formula:

[0041] Calculate the recovery strategy output and generate recovery strategy decision data;

[0042] Where S(x; θ, κ) is the recovery strategy selection function, which is calculated based on the error type decision data x, and ε, θ, and κ are adjustment parameters;

[0043] S313: Perform packet retransmission or local error repair, depending on the recovery strategy decision data, using the operation execution function:

[0044] Calculate and generate operation execution data;

[0045] Where R(d; v, u) is the operation performed based on decision d, and v and u represent the instruction codes for retransmission and repair, respectively;

[0046] S314: Based on the error type decision data, the recovery strategy decision data, and the operation execution data, the completion status formula is adopted: F(r) = hash(r)·(1 + 0.05·r) 2 )

[0047] Calculate the state of the repaired video frames and generate the repaired video frames;

[0048] Where F(r) represents the state of the repaired video frame, and r represents the data state after repair or retransmission.

[0049] As a further aspect of this application, the step of obtaining the adjustment parameters specifically includes:

[0050] S411: Analyze the data integrity of the repaired video frames, collect current network performance metrics, and use the following formula:

[0051] Calculate the network condition index and generate network condition data;

[0052] Where D represents network status data, B, L, and T refer to bandwidth change, packet loss rate, and latency, respectively, and w1, w2, and w3 are adjustment weights;

[0053] S412: Based on the network status data, create adjustment parameters to optimize the parameters of forward error correction and the size of data packets, using the following formula:

[0054] Calculate the current forward error correction parameters and generate the adjusted forward error correction parameters;

[0055] Where P represents the adjusted forward error correction parameter, and κ, λ, δ, and ν are coefficients based on the sensitivity and response range of the actual network management system;

[0056] S413: Adjust the packet size using the adjusted forward error correction parameters, using the formula:

[0057] Calculate the adjusted data packet size and create the adjustment parameters;

[0058] Where S represents the adjusted data packet size, and ξ and μ are adjustment factors.

[0059] As a further aspect of this application, the steps for obtaining the network optimization results are as follows:

[0060] S421: Analyze the adjusted parameters, assess their impact on network performance, and use the performance evaluation formula:

[0061] Calculate the current network performance index and generate network performance evaluation results;

[0062] Where E represents the network performance evaluation result, η and θ are the weighting coefficients of performance measurement, σ is the weighting coefficient of redundancy rate, S is the packet size, P is the forward error correction parameter, and R is the redundancy rate.

[0063] S422: Based on the network performance evaluation results, determine whether iterative optimization of the network configuration is necessary, using the following decision formula:

[0064] Assess the need for network configuration adjustments and generate network adjustment decision results;

[0065] Where C represents the network adjustment decision result, λ is the decision sensitivity adjustment parameter, ε is the performance interval parameter, and δ is the decision adjustment parameter;

[0066] S423: Based on the network adjustment decision results, adjust the network configuration, including bandwidth allocation and data packet transmission interval, using the adjustment formula: N = γ·C 1.5 +δ·|1-C|

[0067] Perform network configuration adjustments and generate network optimization results;

[0068] Where N represents the network optimization result, and γ and δ are the adjustment strategy parameters.

[0069] Compared with the prior art, the advantages and positive effects of this application are as follows:

[0070] In this application, by recalculating the signature of the received video frame and comparing it with the signature of the sender, any data corruption or loss during transmission can be quickly identified, thereby initiating corresponding repair measures. The real-time error detection and response mechanism, through the application of forward error correction technology or the retrieval of redundant data for local repair, reduces the need for retransmission of complete data and improves the effective utilization of network resources. By dynamically adjusting the parameters of forward error correction and the size of data packets, based on the current network state such as bandwidth changes and packet loss rate, overall network performance is optimized, ensuring network connection stability and smooth video transmission. Attached Figure Description

[0071] Figure 1 is a system flowchart of this application;

[0072] Figure 2 is a flowchart of the steps for obtaining the integrity signature information of this application;

[0073] Figure 3 is a flowchart of the steps for obtaining the signature comparison results of this application;

[0074] Figure 4 is a flowchart of the steps for obtaining the video frames after the repair in this application;

[0075] Figure 5 is a flowchart of the steps for obtaining the adjustment parameters in this application;

[0076] Figure 6 is a flowchart of the steps for obtaining the network optimization results of this application. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0078] In the description of this application, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, in the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0079] Example 1

[0080] Please refer to Figure 1. A video doorbell network optimization and fault self-recovery system includes:

[0081] S1: The video frame signature generation module calculates the color distribution, edge information, and texture features of each video frame based on the frame content feature encoding, combines the hash value to establish a composite signature, generates frame signature data, and creates integrity signature information before the video frame is sent based on the frame signature data.

[0082] S2: The video frame signature verification module, based on the integrity signature information, recalculates the signature of the received video frame at the receiving end, compares it with the signature at the sending end, records the information of frames that do not match, and establishes the signature comparison result.

[0083] S3: The network fault repair request module determines the error type as data corruption or data loss based on the signature comparison result, selects to resend the request or apply forward error correction technology for local repair, calls additional redundant data, establishes retransmission or repair instructions, executes data packet retransmission or local error repair, and generates repaired video frames.

[0084] S4: The network status dynamic adjustment module, based on the data integrity of the repaired video frames, combined with bandwidth changes, packet loss rate, and latency indicators, creates adjustment parameters, adjusts the parameters of forward error correction and the size of data packets, optimizes network resource utilization, and obtains network optimization results.

[0085] The frame signature data specifically includes color encoding, edge encoding, and texture encoding. The signature comparison results include frame matching status and difference identifiers. The retransmission or repair instructions specifically refer to error type determination and data repair strategies. The network optimization results include resource allocation efficiency, connection stability, and transmission latency optimization.

[0086] Please refer to Figure 2. The specific steps for obtaining integrity signature information are as follows:

[0087] S111: Collect the raw color data of each video frame, extract color frequencies using the color histogram method, and apply the normalized color frequency formula:

[0088] Calculate the normalized frequency of each color to obtain a normalized color frequency array;

[0089] Among them, C i The normalized frequency of a color, n i α is the number of times color i appears, N is the total number of pixels, α is the weighted color frequency coefficient, β is the normalized offset, and S is the number of times color i appears. i It is the standard deviation of color frequency;

[0090] S112: Analyze the edge information of the video frames, apply the Sobel operator for edge detection, and use the edge strength formula:

[0091] Calculate the edge intensity of pixel (x,y) to obtain the edge intensity matrix;

[0092] Among them, E x,y G represents the edge intensity of pixel (x, y). x G y These represent the gradients in the x and y directions, respectively, and λ and μ are the gradients corresponding to G. x and G y The weighting coefficients of the gradient;

[0093] S113: Extract texture features from video frames, calculate texture contrast using the gray-level co-occurrence matrix, and apply the texture contrast formula:

[0094] Calculate the texture complexity of the entire image and generate texture contrast values;

[0095] Where T represents the contrast of texture features, p i,jγ is the co-occurrence probability of gray levels i and j, γ enhances the sensitivity of contrast to differences, and Ω is used for normalization calculations.

[0096] S114: Combining the normalized color frequency array, edge intensity matrix, and texture contrast values, a feature formula is used:

[0097] Generate frame signature data and establish integrity signature information before sending video frames;

[0098] Where H represents integrity signature information, · represents element-wise multiplication of the array, hash is a hash function used to generate a unique identifier, and ρ is a normalization parameter for overall feature weighting.

[0099] Detailed explanation and derivation of the formula in step 1:

[0100] formula:

[0101] Parameter explanation:

[0102] C i Normalized frequency of color

[0103] n i The number of times color i appears

[0104] N: Total number of pixels

[0105] α: A coefficient for weighted color frequencies, used to adjust the influence of standard deviation; set to 0.5 to reduce noise effects.

[0106] β: Normalization offset, set to 1 to avoid the denominator being zero.

[0107] S i Standard deviation of color frequency

[0108] Calculation example:

[0109] Suppose for a certain color i: n i =150 N=1000

[0110] S i =64 (the squared difference of color frequencies)

[0111] Calculation process:

[0112] calculate

[0113] Insert numbers into formulas:

[0114] Interpretation of results:

[0115] Normalized color frequency C iApproximately 0.154, this means that color accounts for 15.4% of the total pixels. This ratio helps to understand the importance of color in an image.

[0116] Detailed explanation and derivation of the formula in step 2:

[0117] formula:

[0118] Parameter explanation:

[0119] E x,y Edge strength of pixel (x, y)

[0120] G x G y Gradients of the image in the x and y directions

[0121] λ, μ: Gradient weighting coefficients, set to 1, simplifying to the effect of equilibrium.

[0122] Calculation example:

[0123] Suppose that for a certain pixel, the gradient value is: G x =20 G y =15

[0124] Calculation process:

[0125] Interpretation of results:

[0126] Edge strength E x,y A value of 25 indicates that the point is a significant edge region in the image.

[0127] Detailed explanation and derivation of the formula in step 3:

[0128] formula:

[0129] Parameter explanation:

[0130] T: Contrast of texture features

[0131] p i,j The co-occurrence probability of gray levels i and j

[0132] γ: An index emphasizing the degree of difference; set to 2 to enhance the contrast effect.

[0133] Ω: Normalization constant, ensuring the sum is 1, set as the sum of the total probabilities.

[0134] Calculation example:

[0135] Suppose that for gray level 0 and gray level 1, the co-occurrence probability is: p 0,1 =0.05

[0136] Calculation process:

[0137] Explanation of the results:

[0138] The texture contrast T is 0.05, which indicates that the contrast between these two gray levels in the image is not very significant, and the texture area is relatively smooth.

[0139] Please refer to Figure 3. The specific steps for obtaining the signature comparison results are as follows:

[0140] S211: Based on the integrity signature information, after receiving video frame data at the receiving end, extract the normalized color frequency array, edge intensity matrix, and texture contrast value, and use the signature calculation formula:

[0141] The recalculated signature data is obtained;

[0142] Where H′ represents the recalculated signature data, C′, E′, and T′ are the normalized color frequency array, edge intensity matrix, and texture contrast value of the received video frame, respectively, κ, ε, and τ are adjustment constants added for each feature, ρ′ is the normalization constant, and ξ is the adjustment parameter.

[0143] S212: Compare the recalculated signature data with the sender's signature, using the difference measurement formula:

[0144] Calculate the difference between the signatures of the sender and receiver, and generate signature difference data;

[0145] Where Δ represents the signature difference data, and λ is the normalization factor for the difference measure;

[0146] S213: Based on the signature difference data, identify mismatched frames and use the logical filtering formula: R={f∣Δ(f)>θ}

[0147] Generate signature comparison results;

[0148] Where R represents the signature comparison result, f is the frame index, Δ(f) is the signature difference degree of each frame, and θ is the decision threshold.

[0149] Formula derivation and calculation examples in step 1

[0150] formula:

[0151] Parameter details:

[0152] C′: The normalized color frequency array of the video frame at the receiving end, obtained by extracting the color frequencies using the color histogram method and then normalizing them.

[0153] κ: Color frequency adjustment constant, assumed to be 0.1, used to increase flexibility.

[0154] E′: The edge intensity matrix of the received video frame, obtained after edge detection using the Sobel operator.

[0155] ε: Edge strength adjustment constant, assumed to be 0.1, used to increase flexibility.

[0156] T′: The texture contrast value of the received video frame, obtained by calculating using the Gray-Level Co-occurrence Matrix (GLCM).

[0157] τ: Texture contrast adjustment constant, assumed to be 0.1, used to increase flexibility.

[0158] ρ′: Normalization constant, assumed to be 0.1, used for overall feature weighting.

[0159] ξ: Normalization adjustment parameter, assumed to be 0.1, used to increase flexibility.

[0160] Example Derivation:

[0161] Assume the following specific values: C′=[0.2,0.3,0.5] E′=[0.1,0.2,0.3] T′=0.4 κ=0.1 ε=0.1 τ=0.1 ρ′=1.0 ξ=0.1

[0162] The calculation process is as follows:

[0163] The adjusted edge strength matrix: E′+ε=[0.1+0.1,0.2+0.1,0.3+0.1]=[0.2,0.3,0.4]

[0164] Adjusted texture contrast: T′+τ=0.4+0.1=0.5

[0165] Calculate the product: (C′+κ)·(E′+ε)·(T′+τ) =[0.3·0.2·0.5,0.4·0.3·0.5,0.6·0.4·0.5]=[0.03,0.06,0.12] Calculate the denominator: ρ′+ξ=1.0+0.1=1.1

[0166] Final calculation:

[0167] The result is used to calculate the final signature H′ using a hash function.

[0168] Step 2: Formula Derivation and Calculation Examples

[0169] formula:

[0170] Parameter details:

[0171] Δ: Signature difference

[0172] H: Integrity signature information from the sender, obtained through the calculation process in step 1.

[0173] H′: Integrity signature information recalculated by the receiving end, obtained through the calculation process in step 1.

[0174] λ: Normalization factor for the difference measure, assumed to be 1.0, used to reduce measurement error.

[0175] Example Derivation:

[0176] Assume the following specific values: H = [0.03, 0.07, 0.12], H′ = [0.027, 0.055, 0.109], λ = 1.0

[0177] The calculation process is as follows:

[0178] Calculate the difference:

[0179] Calculate the sum of squares:

[0180] Calculate the square root:

[0181] Ultimately, the signature difference Δ≈0.0188.

[0182] Step 3: Formula Derivation and Calculation Examples

[0183] Formula: R={f∣Δ(f)>θ}

[0184] Parameter details:

[0185] R: The final signature comparison result, containing the index of all mismatched frames.

[0186] f: Frame index

[0187] Δ(f): The signature difference of each frame, obtained through the calculation process in step 2.

[0188] θ: Decision threshold, assumed to be 0.02, used to determine when a mismatch is considered.

[0189] Example Derivation:

[0190] Assume the following specific values:

[0191] Signature difference per frame: Δ = [0.0188, 0.025, 0.015, 0.03]

[0192] Judgment threshold: θ = 0.02

[0193] The calculation process is as follows:

[0194] Frame indices with a difference greater than a threshold are selected: R = {f | Δ(f) > 0.02} = {2, 4}

[0195] Ultimately, the signature comparison result R = {2, 4} indicates that there is a mismatch between the 2nd and 4th frames.

[0196] Please refer to Figure 4. The specific steps for obtaining the repaired video frame are as follows:

[0197] S311: Based on the signature comparison results, determine the error type, use a decision algorithm to determine whether it is data corruption or data loss, and apply the error judgment formula: E=α·Δ 3 +β·log(Δ+γ) 2 +δ·Δ

[0198] Calculate erroneous decision indicators and generate error type decision data;

[0199] Where E represents the error decision index, Δ represents the signature discrepancy, and α, β, γ, and δ moderate the impact of the discrepancy on the error decision.

[0200] S312: Based on the error type decision data, select a matching recovery strategy using a selection function formula:

[0201] Calculate the recovery strategy output and generate recovery strategy decision data;

[0202] Where S(x; θ, κ) is the recovery strategy selection function, which is calculated based on the error type decision data x, and ε, θ, and κ are adjustment parameters;

[0203] S313: Perform packet retransmission or local error repair, depending on the recovery strategy decision data, using the operation execution function:

[0204] Calculate and generate operation execution data;

[0205] Where R(d; v, u) is the operation performed based on decision d, and v and u represent the instruction codes for retransmission and repair, respectively;

[0206] S314: Based on error type decision data, recovery strategy decision data, and operation execution data, the completion status formula is adopted: F(r) = hash(r)·(1 + 0.05·r) 2 )

[0207] Calculate the state of the repaired video frames and generate the repaired video frames;

[0208] Where F(r) represents the state of the repaired video frame, and r represents the data state after repair or retransmission.

[0209] Formula analysis and derivation for step 1:

[0210] Formula: E=α·Δ 3 +β·log(Δ+γ) 2 +δ·Δ

[0211] Parameter explanation:

[0212] E: Error Decision Indicator, used to distinguish between data corruption and data loss.

[0213] Δ: Signature discrepancy data, obtained from the signature comparison results of the previous step.

[0214] α, β, γ, δ: Adjustment parameters used to precisely adjust the sensitivity to erroneous decisions.

[0215] Parameter settings and examples:

[0216] Assuming a specific scenario with a difference Δ = 0.5, the adjustment parameters are set as follows: α = 0.8, β = 0.2, γ = 1.0, δ = 0.5.

[0217] Calculation process: E = 0.8 * 0.5 3 +0.2·log(0.5+1.0) 2 +0.5·0.5 E=0.8·0.125+0.2·log(1.5) 2 +0.5·0.5 E=0.1+0.2·0.1761+0.25 (where log(1.5)≈0.1761assuming natural logarithm) E=0.1+0.03522+0.25 E=0.38522

[0218] This error decision indicator E = 0.38522 can be used to identify the error type based on the set threshold.

[0219] Step 2: Formula analysis and derivation:

[0220] formula:

[0221] Parameter explanation:

[0222] x: Error type decision data.

[0223] ε, θ, κ: Adjustment parameters.

[0224] Parameter settings and examples:

[0225] Assume x=0.38522, ε=0.1 θ=5 κ=0.3

[0226] Calculation process: S(x; θ, kappa) = 0.04524

[0227] This value represents the probability of selecting a recovery strategy, which determines whether to request a retransmission or perform local error repair based on the threshold.

[0228] Step 3: Formula analysis and derivation:

[0229] formula:

[0230] Parameter explanation:

[0231] d: Decision instructions obtained from recovery strategy decision data.

[0232] v and u: These are the retransmission and repair commands, respectively.

[0233] Parameter settings and examples:

[0234] Assume d = 0.04524 (obtained from step 2), v = 2, u = 9

[0235] Calculation process: R(d; v,u)=0.04524·4+0.95476·3 R(d;v,u)=0.18096+2.86428 R(d;v,u)=3.04524

[0236] The result R = 3.04524 represents the result of the operation performed based on the decision parameters, indicating a weighted retransmission and repair operation instruction.

[0237] Step 4: Formula analysis and derivation:

[0238] Formula: F(r)=hash(r)·(1+0.05·r) 2 )

[0239] Parameter explanation:

[0240] r: The result of the operation on the data.

[0241] Parameter settings and examples:

[0242] Assume r = 3.04524 (obtained from step 3).

[0243] Calculation process: F(r) = hash(3.04524)·(1+0.05·3.04524) 2) F(r)=hash(3.04524)·(1+0.05·9.273) F(r)=hash(3.04524)·(1+0.46365) F(r)=hash(3.04524)·1.46365

[0244] The hash(3.04524) here needs to be calculated according to the specific hash function. The result F(r) represents the final state of the repaired video frame, ensuring data integrity and correctness.

[0245] Please refer to Figure 5. The specific steps for obtaining the adjustment parameters are as follows:

[0246] S411: Analyze the data integrity of the repaired video frames, collect current network performance metrics, and use the following formula:

[0247] Calculate the network condition index and generate network condition data;

[0248] Where D represents network status data, B, L, and T refer to bandwidth change, packet loss rate, and latency, respectively, and w1, w2, and w3 are adjustment weights;

[0249] S412: Based on network status data, create adjustment parameters to optimize forward error correction parameters and packet size, using the following formula:

[0250] Calculate the current forward error correction parameters and generate the adjusted forward error correction parameters;

[0251] Where P represents the adjusted forward error correction parameter, and κ, λ, δ, and ν are coefficients based on the sensitivity and response range of the actual network management system;

[0252] S413: Adjust packet size using the adjusted forward error correction parameters, using the formula:

[0253] Calculate the adjusted data packet size and create the adjustment parameters;

[0254] Where S represents the adjusted data packet size, and ξ and μ are adjustment factors.

[0255] Detailed explanation and derivation of the formula in step 1

[0256] formula:

[0257] Parameter explanation:

[0258] B: Bandwidth variation (Mbps), assumed to be 100Mbps

[0259] L: Package loss rate (%), assumed to be 2%.

[0260] T: Delay (ms), assumed to be 50ms

[0261] w1, w2, w3: Weighting factors, set as w1 = 0.5, w2 = 0.3, w3 = 0.2 to balance the influence of various indicators.

[0262] Calculation process:

[0263] Bandwidth impact calculation: B 2.5 =100 2.5 =31622.7766Mbps

[0264] Impact of packet loss rate on calculation:

[0265] Delay impact calculation: log(1+T) 2 =log(1+50) 2 =log(51) 2 ≈3.9318 2 =15.459

[0266] Synthetic network state index: D = 0.5·31622.7766 + 0.3·1.2599 + 0.2·15.459 ≈ 15812.91 Interpretation of results:

[0267] The calculated D = 15812.91 is a comprehensive network state index. The higher value reflects that the network state is mainly driven by high bandwidth, indicating the network performance under high transmission capacity.

[0268] Detailed explanation and derivation of the formula in step 2

[0269] formula:

[0270] Parameter settings:

[0271] κ, ν, λ, δ: coefficients, set as κ = 1, ν = 0.01, λ = 0.001, δ = 10000, based on the sensitivity and response range of the actual network management system.

[0272] D: Network state index from step 1

[0273] Calculation process:

[0274] Molecular calculations: (For simplicity, only the formal calculation is performed here; the actual value is much larger.)

[0275] Denominator calculation: 1 + e - 0.001 · (15812.91 - 10000)

[0276] Complete formula calculation:

[0277] Interpretation of results:

[0278] The calculated P-value is used to adjust the forward error correction parameter. This formula takes into account the dynamic adjustment of the error correction parameter due to the significant increase in the network state exponent.

[0279] Detailed explanation and derivation of the formula in step 3

[0280] formula:

[0281] Parameter settings:

[0282] ξ, μ: Adjustment factors, set to ξ = 5000, μ = 0.0001, with typical adjustment ranges based on network data packet size.

[0283] P: Adjustment parameters from step 2

[0284] Calculation process:

[0285] Data packet size adjustment calculation:

[0286] Interpretation of results:

[0287] The calculated S represents the data packet size adjusted according to network conditions to adapt to network load and performance requirements, thereby optimizing transmission efficiency.

[0288] Please refer to Figure 6. The specific steps for obtaining the network optimization results are as follows:

[0289] S421: Analyze and adjust parameters, assess their impact on network performance, and use performance evaluation formulas.

[0290] Calculate the current network performance index and generate network performance evaluation results;

[0291] Where E represents the network performance evaluation result, η and θ are the weighting coefficients of performance measurement, σ is the weighting coefficient of redundancy rate, S is the packet size, P is the forward error correction parameter, and R is the redundancy rate.

[0292] S422: Based on the network performance evaluation results, determine whether iterative optimization of the network configuration is necessary, using the following decision formula:

[0293] Assess the need for network configuration adjustments and generate network adjustment decision results;

[0294] Where C represents the network adjustment decision result, λ is the decision sensitivity adjustment parameter, ε is the performance interval parameter, and δ is the decision adjustment parameter;

[0295] S423: Based on the network adjustment decision, adjust the network configuration, including bandwidth allocation and packet transmission intervals, using the following adjustment formula:

[0296] N=γ·C 1.5 +δ·|1-C|

[0297] Perform network configuration adjustments and generate network optimization results;

[0298] Where N represents the network optimization result, and γ and δ are the adjustment strategy parameters.

[0299] formula:

[0300] Parameter explanation and calculation method:

[0301] E: Network performance evaluation result is an index that comprehensively evaluates network performance.

[0302] η: A weighting factor that measures the impact of packet size on network performance, assumed to be 0.5.

[0303] S: Data packet size, obtained from the previous parameter adjustment steps, assumed to be 1500 bytes.

[0304] θ: A weighting coefficient that measures the impact of forward error correction parameters on network performance, assumed to be 0.3.

[0305] P: Forward error correction parameter, obtained from the previous parameter adjustment steps, assumed to be 2.

[0306] σ: A weighting coefficient that measures the impact of redundant data rate on network performance, assumed to be 0.2.

[0307] R: Redundancy rate, assumed to be 0.05.

[0308] Calculation process:

[0309] First calculate

[0310] Then calculate |P| 1.5 : |P| 1.5 =2 1.5 ≈2.83

[0311] Next calculation

[0312] Finally, combining the above calculation results: E = 0.5 * 4.44 × 10 -7+0.3·2.83+0.2·0.2236 E≈2.83×0.3+0.2236×0.2≈0.849+0.0447≈0.8937

[0313] Interpretation of results:

[0314] E represents the current network performance index, which comprehensively evaluates the impact of packet size, forward error correction parameters, and redundancy rate on network performance. The specific value of 0.8937 reflects the current network performance level.

[0315] formula:

[0316] Parameter explanation and calculation method:

[0317] C: Network adjustment decision output, which determines whether network configuration needs to be adjusted.

[0318] λ: Decision sensitivity adjustment parameter, assumed to be 1.0.

[0319] ε: Performance threshold parameter, assumed to be 0.5.

[0320] δ: Decision adjustment parameter, assumed to be 0.

[0321] E: The network performance evaluation result obtained from step 1 is assumed to be 0.8937.

[0322] Calculation process:

[0323] First, calculate E. 2 -ε: E 2 -ε=(0.8937) 2 -0.5 ≈ 0.7997 - 0.5 = 0.2997

[0324] Then calculate the exponent part:

[0325] Finally, calculate the decision result:

[0326] Interpretation of results:

[0327] C represents the decision output for network configuration adjustment; the specific value of 0.5744 indicates that the current network configuration needs to be adjusted.

[0328] Formula: N=γ·C 1.5 +δ·|1-C|

[0329] Perform network configuration adjustments and generate the final network optimization result N.

[0330] Parameter explanation and calculation method:

[0331] N: The final state after network configuration adjustments.

[0332] γ: Adjust the strategy parameter, assuming it is 10.

[0333] δ: Adjust the strategy parameter, let's assume it's 5.

[0334] C: The network adjustment decision result calculated from step 2 is assumed to be 0.5744.

[0335] Calculation process:

[0336] First, calculate C. 1.5 C 1.5 =(0.5744) 1.5 ≈0.435

[0337] Then calculate |1-C|: |1-C|=|1-0.5744|=0.4256

[0338] Finally, combining the above calculation results: N = 10 * 0.435 + 5 * 0.4256 N = 4.35 + 2.128 ≈ 6.478

[0339] Interpretation of results:

[0340] N represents the final state after network configuration adjustment, and the specific value 6.478 represents the result of the adjusted network configuration.

[0341] The above are merely embodiments of this application and are not intended to limit this application in any other way. Any person skilled in the art may use the above-disclosed technical content to make changes or modifications to equivalent embodiments and apply them to other fields. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of this application without departing from the content of the technical solution of this application shall still fall within the protection scope of the technical solution of this application.

Claims

1. A video doorbell network optimization and fault self-recovery system, wherein, The system includes: The video frame signature generation module calculates the color distribution, edge information, and texture features of each video frame based on frame content feature encoding, combines the hash value to establish a composite signature, generates frame signature data, and creates integrity signature information before the video frame is sent based on the frame signature data. Based on the integrity signature information, the video frame signature verification module recalculates the signature of the received video frame at the receiving end, compares it with the signature at the sending end, records the information of mismatched frames, and establishes the signature comparison result. Based on the signature comparison result, the network fault repair request module determines whether the error type is data corruption or data loss, selects to resend the request or apply forward error correction technology for local repair, calls additional redundant data, establishes retransmission or repair instructions, executes data packet retransmission or local error repair, and generates a repaired video frame. The network status dynamic adjustment module, based on the data integrity of the repaired video frames, and combined with bandwidth changes, packet loss rate, and latency indicators, creates adjustment parameters, adjusts the parameters of forward error correction and the size of data packets, optimizes network resource utilization, and obtains network optimization results.

2. The video doorbell network optimization and fault self-recovery system according to claim 1, wherein, The specific steps for obtaining the integrity signature information are as follows: Raw color data for each video frame is collected, and color frequencies are extracted using the color histogram method, followed by a normalized color frequency formula: Calculate the normalized frequency of each color to obtain a normalized color frequency array; Among them, C i The normalized frequency of a color, n i α is the number of times color i appears, N is the total number of pixels, α is the weighted color frequency coefficient, β is the normalized offset, and S is the number of times color i appears. i It is the standard deviation of color frequency; Analyze the edge information of video frames, apply the Sobel operator for edge detection, and use the edge strength formula: Calculate the edge intensity of pixel (x,y) to obtain the edge intensity matrix; Among them, E x,y G represents the edge intensity of pixel (x, y). x G y These represent the gradients in the x and y directions, respectively, and λ and μ are the gradients corresponding to G. x and G y The weighting coefficients of the gradient; Extract texture features from video frames, calculate texture contrast using the gray-level co-occurrence matrix, and apply the texture contrast formula: Calculate the texture complexity of the entire image and generate texture contrast values; Where T represents the contrast of texture features, p i,j γ is the co-occurrence probability of gray levels i and j, γ enhances the sensitivity of contrast to differences, and Ω is used for normalization calculations. Combining the normalized color frequency array, the edge intensity matrix, and the texture contrast value, the feature formula is adopted: Generate frame signature data and establish integrity signature information before sending video frames; Where H represents integrity signature information, · represents element-wise multiplication of the array, hash is a hash function used to generate a unique identifier, and ρ is a normalization parameter for overall feature weighting.

3. The video doorbell network optimization and fault self-recovery system according to claim 2, wherein, The specific steps for obtaining the signature comparison result are as follows: Based on the integrity signature information, after receiving video frame data at the receiving end, the normalized color frequency array, edge intensity matrix, and texture contrast value are extracted, and the signature calculation formula is used: The recalculated signature data is obtained; Where H′ represents the recalculated signature data, C′, E′, and T′ are the normalized color frequency array, edge intensity matrix, and texture contrast value of the received video frame, respectively, κ, ε, and τ are adjustment constants added for each feature, ρ′ is the normalization constant, and ξ is the adjustment parameter. The recalculated signature data is compared with the sender's signature using a difference measurement formula: Calculate the difference between the signatures of the sender and receiver, and generate signature difference data; Where Δ represents the signature difference data, and λ is the normalization factor for the difference measure; Based on the signature difference data, mismatched frames are identified using a logical filtering formula: R = {f | Δ(f) > θ} Generate signature comparison results; Where R represents the signature comparison result, f is the frame index, Δ(f) is the signature difference degree of each frame, and θ is the decision threshold.

4. The video doorbell network optimization and fault self-recovery system according to claim 3, wherein, The specific steps for obtaining the repaired video frame are as follows: Based on the signature comparison results, the error type is determined, and a decision algorithm is used to determine whether it is data corruption or data loss. The error judgment formula is applied as follows: E=a·D 3 +β·log(Δ+γ) 2 +d·D Calculate erroneous decision indicators and generate error type decision data; Where E represents the error decision index, Δ represents the signature discrepancy, and α, β, γ, and δ moderate the impact of the discrepancy on the error decision. Based on the error type decision data, a matching recovery strategy is selected using a selection function formula: Calculate the recovery strategy output and generate recovery strategy decision data; Where S(x; θ, κ) is the recovery strategy selection function, which is calculated based on the error type decision data x, and ε, θ, and κ are adjustment parameters; Whether to perform packet retransmission or local error repair depends on the recovery strategy decision data, and is executed using the following function: Calculate and generate operation execution data; Where R(d; v, u) is the operation performed based on decision d, and v and u represent the instruction codes for retransmission and repair, respectively; Based on the error type decision data, the recovery strategy decision data, and the operation execution data, the completion status formula is adopted: F(r)=hash(r)·(1+0.05·r 2 ) Calculate the state of the repaired video frames and generate the repaired video frames; Where F(r) represents the state of the repaired video frame, and r represents the data state after repair or retransmission.

5. The video doorbell network optimization and fault self-recovery system according to claim 4, wherein, The specific steps for obtaining the adjustment parameters are as follows: Analyze the data integrity of the repaired video frames, collect current network performance metrics, and use the following formula: Calculate the network condition index and generate network condition data; Where D represents network status data, B, L, and T refer to bandwidth change, packet loss rate, and latency, respectively, and w1, w2, and w3 are adjustment weights; Based on the network status data, adjust parameters to optimize the forward error correction parameters and packet size, using the following formula: Calculate the current forward error correction parameters and generate the adjusted forward error correction parameters; Where P represents the adjusted forward error correction parameter, and κ, λ, δ, and ν are coefficients based on the sensitivity and response range of the actual network management system; The packet size is adjusted using the modified forward error correction parameters, according to the formula: Calculate the adjusted data packet size and create the adjustment parameters; Where S represents the adjusted data packet size, and ξ and μ are adjustment factors.

6. The video doorbell network optimization and fault self-recovery system according to claim 5, wherein, The specific steps for obtaining the network optimization results are as follows: Analyze the adjusted parameters, assess their impact on network performance, and use the performance evaluation formula: Calculate the current network performance index and generate network performance evaluation results; Where E represents the network performance evaluation result, η and θ are the weighting coefficients of performance measurement, σ is the weighting coefficient of redundancy rate, S is the packet size, P is the forward error correction parameter, and R is the redundancy rate. Based on the network performance evaluation results, a decision is made as to whether iterative optimization of the network configuration is necessary, using the following decision formula: Assess the need for network configuration adjustments and generate network adjustment decision results; Where C represents the network adjustment decision result, λ is the decision sensitivity adjustment parameter, ε is the performance interval parameter, and δ is the decision adjustment parameter; Based on the network adjustment decision results, network configuration adjustments are made, including adjusting bandwidth allocation and data packet transmission intervals, using the following adjustment formula: N=γ·C 1.5 +δ·|1-C| Perform network configuration adjustments and generate network optimization results; Where N represents the network optimization result, and γ and δ are the adjustment strategy parameters.

Images