CN118075914A - NVR and IPC automatic wireless code matching connection method - Google Patents

Abstract

The application belongs to the technical field of video network application, and particularly relates to an NVR and IPC automatic wireless code matching connection method; according to the application, the equipment identification module of the NVR scans surrounding wireless signals, the signal characteristics are analyzed based on the deep learning model to identify the signals of the IPC equipment, then a preliminary communication link is established with the identified IPC equipment, the generated dynamic key is exchanged based on the preliminary communication link, information of both sides is authenticated, and after the authentication is successful, the optimal connection frequency band is intelligently selected based on the characteristics of the current network environment, the transmission power and the communication parameters are adjusted, so that the automatic wireless code matching of the NVR and the IPC equipment is realized, the problem of time and labor consumption in manual configuration is effectively avoided, and the optimal connection frequency band, the transmission power and the communication parameters are intelligently selected, so that the method can adapt to complex and changeable network environments and ensure the communication quality.

Description

NVR and IPC automatic wireless code matching connection method

Technical Field

The application belongs to the technical field of video network application, and particularly relates to an NVR and IPC automatic wireless code matching connection method.

Background

In the current digital age, a video monitoring system becomes a key technology in the fields of safety and supervision, and is widely applied to various fields of public safety, traffic monitoring, industrial monitoring, commercial and residential safety and the like; wherein, the combination of the Network Video Recorder (NVR) and the network video camera (IPC) forms the core of the modern video monitoring system; with the rapid development of wireless communication technology, wireless NVR and IPC systems are increasingly adopted due to the advantages of flexible installation, low cost, easy expansion and the like.

In a wireless video monitoring system, it is important to ensure stable, reliable and efficient communication connection between NVR and IPC, which requires that the system not only be able to automatically identify and pair devices, but also to intelligently manage communication links, including selecting the best radio frequency band and automatically adjusting communication parameters to accommodate dynamically changing network environments.

The traditional wireless video monitoring system has a plurality of limitations in terms of equipment code matching, signal transmission and communication quality management; for example, the device pairing and communication link establishment procedures in early systems often required manual configuration, which is not only time-consuming and labor-consuming, but also prone to error; in addition, the fixed frequency band selection and the communication parameter setting are difficult to ensure optimal communication quality in complex and changeable network environments, which may cause video transmission interruption, image quality degradation or delay increase, thereby affecting the monitoring effect and the practicability of the system.

Disclosure of Invention

The invention provides an NVR and IPC automatic wireless code matching connection method, which aims to solve the technical problems that the manual configuration is time-consuming, labor-consuming and error-prone at present, and the fixed frequency band selection and the communication parameters are set in a complex and changeable network environment, so that the optimal communication quality is difficult to ensure.

An NVR and IPC automatic wireless code matching connection method comprises the following steps:

the NVR scans surrounding wireless signals based on the equipment identification module and analyzes the signal characteristics based on the deep learning model to identify the signals of the IPC equipment;

based on the identified signal of the IPC device, the NVR establishes a preliminary communication link with the IPC device;

the NVR and IPC equipment authenticates information of both parties based on the dynamic key generated by the primary communication link exchange;

After the information authentication of the two parties is successful, the optimal connection frequency band is intelligently selected based on the characteristics of the current network environment, and the transmission power and the communication parameters are automatically adjusted.

According to the invention, the equipment identification module of the NVR scans surrounding wireless signals, the signal characteristics are analyzed based on the deep learning model to identify the signals of the IPC equipment, then a preliminary communication link is established with the identified IPC equipment, the generated dynamic key is exchanged based on the preliminary communication link, information of both sides is authenticated, and after the authentication is successful, the optimal connection frequency band is intelligently selected based on the characteristics of the current network environment, the transmission power and the communication parameters are adjusted, so that the automatic wireless code matching of the NVR and the IPC equipment is realized, the problem of time and labor consumption in manual configuration is effectively avoided, and the optimal connection frequency band, the transmission power and the communication parameters are intelligently selected, so that the method can adapt to complex and changeable network environments and ensure the communication quality.

Preferably, the analyzing the signal characteristics based on the deep learning model to identify the signal of the IPC device includes:

and a band-pass filter is adopted to reserve signals in the operating frequency range of the IPC equipment:

；

Wherein: And/> Representing the low and high cut-off frequencies of the band-pass filter, respectively;

and carrying out noise suppression on the signals after band-pass filtering by adopting spectral subtraction to improve the signal-to-noise ratio of the signals:

；

Wherein: representing a noise suppressed version of the signal; /(I) Representing a signal spectrum; /(I)Representing noise suppression intensity; Representing the noise spectrum;

extracting features of the signals subjected to noise suppression, extracting features which are helpful for identifying IPC equipment, and constructing feature vectors;

analyzing the feature vector based on the deep learning model, and identifying signals of the IPC equipment;

The deep learning model is a mixed model combining CNN and LSTM, and the output of the mixed model is the probability of the signal being IPC equipment ：

；

Wherein: Representing a sigmoid function, converting the output into a probability; /(I) 、/>、/>AndNetwork parameters respectively representing LSTM and CNN; /(I)Representing CNN processing input characteristics and extracting spatial characteristics; Representing LSTM processing sequence features, capturing time dependence.

In the invention, the signal quality is optimized through the band-pass filter and the spectral subtraction, so that the signal-to-noise ratio of the signal is improved, and the accuracy of signal processing is further ensured; finally, the spatial characteristics can be extracted by combining the mixed model of the CNN and the LSTM, the time dependence of the signals can be captured, and the recognition accuracy is improved.

Preferably, the training step of the hybrid model combining CNN and LSTM is as follows:

collecting wireless signal data in an actual environment, including signals from IPC equipment and signals of other interference sources;

labeling the collected data, and distinguishing signals from IPC equipment and signals from non-IPC equipment;

Preprocessing the marked data, including signal filtering and denoising;

extracting signal characteristics from the preprocessed data, and constructing characteristic vectors to obtain a data set;

Dividing the data set into a training set and a verification set by adopting cross verification, and using the training set for training the mixed model; training the mixed model by adopting an Adam optimizer, and adjusting the super parameters of the model; the training of the mixed model is stopped when the loss reaches the preset requirement by monitoring the accuracy and the loss of the training process;

and evaluating the performance of the model based on the test set, and adjusting and optimizing the structure and parameters of the hybrid model based on the performance evaluation result.

Preferably, the NVR and IPC devices exchange generated dynamic keys based on the preliminary communication link, and the authentication information includes:

the NVR and IPC devices start a TLS handshake based on the established preliminary communication link;

NVR sends a supported encryption algorithm list to IPC equipment; the IPC device selects a matched encryption algorithm and sends a digital certificate to the NVR, wherein the digital certificate comprises a public key of the IPC device, and the digital certificate is signed by a trusted certificate authority;

NVR verifies the certificate validity of the IPC device, including whether the signature, certificate chain, and public key in the certificate actually represent the IPC device; after verification is successful, ECDH key exchange is carried out;

the NVR and the IPC equipment respectively generate respective ECDH public-private key pairs, public key exchange is carried out on the NVR and the IPC equipment after generation, and the same shared secret key is calculated through an ECDH protocol by respectively using the public key of the opposite party and the private key of the opposite party;

The NVR and IPC devices derive a session key using the shared secret key for subsequent communications encryption and message authentication.

In the invention, based on TLS handshake and ECDH key exchange mechanism, the security of data transmission is ensured, data leakage and man-in-the-middle attack are prevented, the security of the network is further enhanced by verifying the digital certificate of the IPC equipment, and the authenticity and reliability of the equipment are ensured.

Preferably, the intelligent selecting the optimal connection frequency band includes:

Testing each available radio frequency band, and transmitting and receiving a preset data packet on each frequency band by NVR and IPC equipment;

Monitoring signal strength, noise level and interference index on each frequency band;

calculating a frequency band utility value based on the monitored signal strength, noise level, and interference index:

；

Wherein: a utility value representing the frequency band f; /(I) Signal strength representing frequency band f; /(I)A noise level representing frequency band f; /(I)An interference index representing the frequency band f; /(I)、/>/>Representing the weight factor;

and calculating the frequency band utility value of each acquired available frequency band, and selecting the frequency band with the highest score.

Preferably, the automatically adjusting the transmission power includes:

adjusting the transmission power according to the selected frequency band and the current network condition:

；

Wherein: representing the transmission power at a distance d and a frequency f; /(I) Representing a baseline power at a reference distance; /(I)Representing a frequency-dependent environment-specific signal attenuation factor; /(I)Representing a distance between the NVR device and the IPC device; /(I)Representing the center frequency corresponding to the currently used frequency band; wherein/>Based on field test under different frequency bands, signal strength under different distances is measured, and then signal attenuation factors/>, are determined。

Preferably, the communication parameters include a channel bandwidth and a modulation scheme, wherein the channel bandwidth is adjusted as follows:

；

Wherein: a specific value representing bandwidth adjustment; /(I) Is the minimum of available bandwidth; /(I)Is the maximum value of available bandwidth; /(I)Representing the current signal to noise ratio; /(I)Representing a preset minimum signal-to-noise ratio threshold; /(I)Representing a preset highest signal-to-noise ratio threshold;

the modulation scheme determines a specific modulation scheme according to the current signal-to-noise ratio traversal mapping table; and constructing modulation schemes corresponding to different signal-to-noise ratio intervals in the mapping table.

Preferably, the intelligent selection of the optimal connection frequency band, and automatic adjustment of transmission power and communication parameters are realized through a parameter prediction model arranged in the NVR;

The parameter prediction model adopts a combination of LSTM and CNN models; wherein LSTM is used to analyze time series data and CNN is used to analyze features in the frequency domain;

The training steps of the parameter prediction model are as follows:

The NVR and IPC devices collect network environment and device status data, including: signal strength, noise level, interference index, signal to noise ratio, currently used frequency band, transmission power, channel bandwidth, modulation scheme, battery power of the device and power consumption;

Pretreatment: cleaning and standardizing the collected network environment and equipment state data;

Taking the preprocessed network environment and the preprocessed device state data as a training set, wherein the training set comprises the device state data, the network environment data and the corresponding optimal communication parameter configuration as labels;

Training the parameter prediction model by using a training set, and optimizing model parameters of the parameter prediction model through back propagation and gradient descent;

wherein the output of the parameter prediction model includes the connection frequency band, the transmission power, and the communication parameters.

The beneficial effects of the invention include:

According to the invention, the equipment identification module of the NVR scans surrounding wireless signals, the signal characteristics are analyzed based on the deep learning model to identify the signals of the IPC equipment, then a preliminary communication link is established with the identified IPC equipment, the generated dynamic key is exchanged based on the preliminary communication link, information of both sides is authenticated, and after the authentication is successful, the optimal connection frequency band is intelligently selected based on the characteristics of the current network environment, the transmission power and the communication parameters are adjusted, so that the automatic wireless code matching of the NVR and the IPC equipment is realized, the problem of time and labor consumption in manual configuration is effectively avoided, and the optimal connection frequency band, the transmission power and the communication parameters are intelligently selected, so that the method can adapt to complex and changeable network environments and ensure the communication quality.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a block diagram of overall steps provided in an embodiment of the present invention.

Fig. 2 is a block diagram of steps for identifying IPC device signals according to an embodiment of the present invention.

Fig. 3 is a block diagram of steps for selecting an optimal frequency band according to an embodiment of the present invention.

Detailed Description

In order to make the technical problems, technical schemes and beneficial effects to be solved more clear, the application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

Examples

Referring to fig. 1, an automatic wireless code matching connection method for NVR and IPC includes the following steps:

the NVR scans surrounding wireless signals based on the equipment identification module and analyzes the signal characteristics based on the deep learning model to identify the signals of the IPC equipment;

based on the identified signal of the IPC device, the NVR establishes a preliminary communication link with the IPC device;

the NVR and IPC equipment authenticates information of both parties based on the dynamic key generated by the primary communication link exchange;

After the information authentication of the two parties is successful, the optimal connection frequency band is intelligently selected based on the characteristics of the current network environment, and the transmission power and the communication parameters are automatically adjusted.

According to the invention, the equipment identification module of the NVR scans surrounding wireless signals, the signal characteristics are analyzed based on the deep learning model to identify the signals of the IPC equipment, then a preliminary communication link is established with the identified IPC equipment, the generated dynamic key is exchanged based on the preliminary communication link, information of both sides is authenticated, and after the authentication is successful, the optimal connection frequency band is intelligently selected based on the characteristics of the current network environment, the transmission power and the communication parameters are adjusted, so that the automatic wireless code matching of the NVR and the IPC equipment is realized, the problem of time and labor consumption in manual configuration is effectively avoided, and the optimal connection frequency band, the transmission power and the communication parameters are intelligently selected, so that the method can adapt to complex and changeable network environments and ensure the communication quality.

Preferably, referring to fig. 2, the analyzing the signal of the signal feature recognition IPC device based on the deep learning model includes:

and a band-pass filter is adopted to reserve signals in the operating frequency range of the IPC equipment:

；

Wherein: And/> Representing the low and high cut-off frequencies of the band-pass filter, respectively;

and carrying out noise suppression on the signals after band-pass filtering by adopting spectral subtraction to improve the signal-to-noise ratio of the signals:

；

Wherein: representing a noise suppressed version of the signal; /(I) Representing a signal spectrum; /(I)Representing noise suppression intensity; Representing the noise spectrum;

extracting features of the signals subjected to noise suppression, extracting features which are helpful for identifying IPC equipment, and constructing feature vectors;

analyzing the feature vector based on the deep learning model, and identifying signals of the IPC equipment;

The deep learning model is a mixed model combining a Convolutional Neural Network (CNN) and a cyclic neural network (LSTM), and the output of the mixed model is the probability of the signal being IPC equipment ：

；

Wherein: Representing a sigmoid function, converting the output into a probability; /(I) 、/>、/>AndNetwork parameters respectively representing LSTM and CNN; /(I)Representing CNN processing input characteristics and extracting spatial characteristics; Representing LSTM processing sequence features, capturing time dependence.

In the embodiment, the signal quality is optimized through the band-pass filter and the spectral subtraction, so that the signal-to-noise ratio of the signal is improved, and the accuracy of signal processing is further ensured; finally, the spatial characteristics of the mixed model combining the CNN and the LSTM can be extracted, the time dependence of the signals can be captured, and the recognition accuracy is improved.

Illustratively, the structure of the hybrid model includes:

Input layer: for receiving feature vectors of the original signal, such as signal data subjected to preprocessing and feature extraction; the dimension of the input data is Wherein: /(I)Indicating the batch size; Representing the length of the time series; /(I) Representing the number of features per time step;

CNN layer: the CNN layer comprises a plurality of convolution layers and a plurality of pooling layers; wherein the convolution layer is used for identifying a spatial mode in the signal, and the pooling layer is used for reducing characteristic dimension and increasing spatial invariance of the model; linking the pooling layers after each convolution layer;

Layer of flat: constructing a flattening layer (flat) between the CNN and the LSTM layer, converting the two-dimensional feature map into one dimension for LSTM layer processing;

LSTM layer: receiving characteristics of a flattening layer, and processing time sequence data to obtain a long-term dependency relationship; wherein the LSTM layer has a plurality of LSTM layers for enhancing the time series learning ability of the model.

Output layer: comprises a neuron, which uses a sigmoid activation function for outputting the probability that the signal belongs to the IPC device.

Preferably, the training step of the hybrid model combining CNN and LSTM is as follows:

collecting wireless signal data in an actual environment, including signals from IPC equipment and signals of other interference sources;

labeling the collected data, and distinguishing signals from IPC equipment and signals from non-IPC equipment;

Preprocessing the marked data, including signal filtering and denoising;

extracting signal characteristics from the preprocessed data, and constructing characteristic vectors to obtain a data set;

Dividing the data set into a training set and a verification set by adopting cross verification, and using the training set for training the mixed model; training the mixed model by adopting an Adam optimizer, and adjusting the super parameters of the model; the training of the mixed model is stopped when the loss reaches the preset requirement by monitoring the accuracy and the loss of the training process;

and evaluating the performance of the model based on the test set, and adjusting and optimizing the structure and parameters of the hybrid model based on the performance evaluation result.

In one possible implementation manner, the NVR and IPC devices exchange generated dynamic keys based on a preliminary communication link, and the authentication information includes:

the NVR and IPC devices start a TLS handshake based on the established preliminary communication link;

NVR sends a supported encryption algorithm list to IPC equipment; the IPC device selects a matched encryption algorithm and sends a digital certificate to the NVR, wherein the digital certificate comprises a public key of the IPC device, and the digital certificate is signed by a trusted certificate authority;

NVR verifies the certificate validity of the IPC device, including whether the signature, certificate chain, and public key in the certificate actually represent the IPC device; after verification is successful, ECDH (elliptic curve Diffie-Hellman key exchange) key exchange is carried out;

the NVR and the IPC equipment respectively generate respective ECDH public-private key pairs, public key exchange is carried out on the NVR and the IPC equipment after generation, and the same shared secret key is calculated through an ECDH protocol by respectively using the public key of the opposite party and the private key of the opposite party;

The NVR and IPC devices derive a session key using the shared secret key for subsequent communications encryption and message authentication.

The ECDH exchange comprises the following specific steps:

Elliptic curves and base points are selected:

The NVR and IPC devices use a common elliptic curve parameter (NIST curve) and a base point (G).

NVR generates a public-private key pair:

a random number (a) is generated as its private key.

The public key (a=ag), i.e. the elliptic curve point multiplication of the private key (a) and the base point (G), is calculated.

The IPC device generates a public-private key pair:

a random number (b) is generated as its private key.

The public key (b=bg) is calculated, likewise by elliptic curve point multiplication.

Exchanging public keys:

the NVR sends its public key (A) to the IPC device.

The IPC device sends its public key (B) to the NVR.

Calculating a shared secret key:

at the NVR end, the shared secret (s=ab) is calculated using its private key (a) and the public key (B) received from the IPC device. The property according to the elliptic curve is equivalent to a (bG).

On the IPC device side, using its private key (b) and the public key (a) received from the NVR, computing the shared secret (s=ba) is equivalent to b (aG).

Verify and use the shared secret key:

Both sides get the same shared secret (S) because (abG = baG).

The party can derive a session key using this shared secret (S) for encrypting and authenticating messages in the communication process.

In the embodiment, by combining the ECDH key exchange protocol and the TLS protocol, the encryption scheme not only effectively solves the problem of secure automatic wireless code matching connection between different signals and IPC devices of different manufacturers, but also provides high-strength security guarantee. And based on TLS handshake and ECDH key exchange mechanism, the security of data transmission is ensured, data leakage and man-in-the-middle attack are prevented, the security of the network is further enhanced by verifying the digital certificate of the IPC equipment, and the authenticity and reliability of the equipment are ensured.

Preferably, referring to fig. 3, the intelligently selecting the optimal connection frequency band includes:

Testing each available radio frequency band, and transmitting and receiving a preset data packet on each frequency band by NVR and IPC equipment;

Monitoring signal strength, noise level and interference index on each frequency band;

calculating a frequency band utility value based on the monitored signal strength, noise level, and interference index:

；

Wherein: a utility value representing the frequency band f; /(I) Signal strength representing frequency band f; /(I)A noise level representing frequency band f; /(I)An interference index representing the frequency band f; /(I)、/>/>Representing the weight factor; the weight factors are set according to the actual application scene.

And calculating the frequency band utility value of each acquired available frequency band, and selecting the frequency band with the highest score.

Preferably, the automatically adjusting the transmission power includes:

adjusting the transmission power according to the selected frequency band and the current network condition:

；

Wherein: representing the transmission power at a distance d and a frequency f; /(I) Representing a baseline power at a reference distance; /(I)Representing a frequency-dependent environment-specific signal attenuation factor; /(I)Representing a distance between the NVR device and the IPC device; /(I)Representing the center frequency corresponding to the currently used frequency band; wherein/>Based on field test under different frequency bands, signal strength under different distances is measured, and then signal attenuation factors/>, are determined。

In one possible implementation manner, the communication parameters include a channel bandwidth and a modulation scheme, wherein the channel bandwidth is adjusted as follows:

；

Wherein: a specific value representing bandwidth adjustment; /(I) Is the minimum of available bandwidth; /(I)Is the maximum value of available bandwidth; /(I)Representing the current signal to noise ratio; /(I)Representing a preset minimum signal-to-noise ratio threshold; /(I)Representing a preset highest signal-to-noise ratio threshold;

The modulation scheme determines a specific modulation scheme according to the current signal-to-noise ratio traversal mapping table; constructing modulation schemes corresponding to different signal-to-noise ratio intervals in the mapping table;

Illustratively, the selection of the modulation scheme is dynamically adjusted based on the current signal-to-noise ratio, e.g., when the signal-to-noise ratio is high, a higher order modulation scheme is selected to increase the data transmission rate; when the signal-to-noise ratio is low, a low modulation scheme is selected, so that the reliability of transmission is improved; the mapping table is shown in the following table 1:

TABLE 1

Through the technical scheme, the NVR and IPC equipment can intelligently adjust transmission power, channel bandwidth and modulation scheme according to real-time network state and data requirements so as to adapt to complex and changeable wireless network environments. The dynamic adjustment strategy is beneficial to improving the stability, transmission efficiency and energy efficiency of wireless connection and meeting the requirements of a high-quality video monitoring system.

Example 2

The difference between the embodiment 2 and the embodiment 1 is only that the technical schemes of selecting the best connection frequency band and automatically adjusting the transmission power and the communication parameters are different, and the specific embodiments are as follows:

The optimal connection frequency band is intelligently selected, the transmission power and the communication parameters are automatically adjusted, and the intelligent connection frequency band is realized through a parameter prediction model arranged in the NVR;

The parameter prediction model adopts a combination of LSTM and CNN models; wherein LSTM is used to analyze time series data and CNN is used to analyze features in the frequency domain;

The training steps of the parameter prediction model are as follows:

The NVR and IPC devices collect network environment and device status data, including: signal strength, noise level, interference index, signal to noise ratio, currently used frequency band, transmission power, channel bandwidth, modulation scheme, battery power of the device and power consumption;

Pretreatment: cleaning and standardizing the collected network environment and equipment state data;

Taking the preprocessed network environment and the preprocessed device state data as a training set, wherein the training set comprises the device state data, the network environment data and the corresponding optimal communication parameter configuration as labels;

Training the parameter prediction model by using a training set, and optimizing model parameters of the parameter prediction model through back propagation and gradient descent;

wherein the output of the parameter prediction model includes the connection frequency band, the transmission power, and the communication parameters.

The parameter prediction model is designed by referring to the structure of the hybrid model, and the difference is that the structure and the super parameters of the model are different according to the training condition; but these differences are based on conventional adjustments that can be made by those skilled in the art; therefore, corresponding details are not repeated in this embodiment.

For the different advantages of the technical solutions of embodiment 2 and embodiment 1 of the present application, that is, the implementation manner of embodiment 1 is simpler than that of embodiment 2, and is easy to implement and maintain, and it can adapt to the common occasion with less extreme environmental change; and the calculation amount of the embodiment 2 is far larger than that of the embodiment 1; therefore, whether the technical scheme of the embodiment 1 or the technical scheme of the embodiment 2 is deployed can be selected according to the application scene, and the embodiment 1 is suitable for relatively stable or infrequent-change scenes; whereas embodiment 2 is more suitable for scenes where the environmental dynamics are complex or where long-term optimization is required; and embodiment 1 has lower demand for resources, and is suitable for resource-constrained devices; whereas example 2 requires stronger hardware support and more maintenance work; the technical solutions of the above-described embodiment 1 and embodiment 2 are therefore each advantageous; the deployment can be performed according to the actual application scene.

The foregoing description of the preferred embodiments of the application is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the application.

Claims (8)

1. An NVR and IPC automatic wireless code matching connection method is characterized by comprising the following steps:

the NVR scans surrounding wireless signals based on the equipment identification module and analyzes the signal characteristics based on the deep learning model to identify the signals of the IPC equipment;

based on the identified signal of the IPC device, the NVR establishes a preliminary communication link with the IPC device;

the NVR and IPC equipment authenticates information of both parties based on the dynamic key generated by the primary communication link exchange;

After the information authentication of the two parties is successful, the optimal connection frequency band is intelligently selected based on the characteristics of the current network environment, and the transmission power and the communication parameters are automatically adjusted.

2. The automatic wireless code alignment connection method for NVR and IPC according to claim 1, wherein the analyzing the signal characteristic based on the deep learning model to identify the signal of the IPC device comprises:

and a band-pass filter is adopted to reserve signals in the operating frequency range of the IPC equipment:

；

Wherein: And/> Representing the low and high cut-off frequencies of the band-pass filter, respectively;

and carrying out noise suppression on the signals after band-pass filtering by adopting spectral subtraction to improve the signal-to-noise ratio of the signals:

；

Wherein: representing a noise suppressed version of the signal; /(I) Representing a signal spectrum; /(I)Representing noise suppression intensity; /(I)Representing the noise spectrum;

extracting features of the signals subjected to noise suppression, extracting features which are helpful for identifying IPC equipment, and constructing feature vectors;

analyzing the feature vector based on the deep learning model, and identifying signals of the IPC equipment;

The deep learning model is a mixed model combining CNN and LSTM, and the output of the mixed model is the probability of the signal being IPC equipment ：

；

Wherein: Representing a sigmoid function, converting the output into a probability; /(I) 、/>、/>/>Network parameters respectively representing LSTM and CNN; /(I)Representing CNN processing input characteristics and extracting spatial characteristics; /(I)Representing LSTM processing sequence features, capturing time dependence.

3. The automatic wireless code alignment connection method for NVR and IPC according to claim 2, wherein the training step of combining the hybrid model of CNN and LSTM is as follows:

collecting wireless signal data in an actual environment, including signals from IPC equipment and signals of other interference sources;

labeling the collected data, and distinguishing signals from IPC equipment and signals from non-IPC equipment;

Preprocessing the marked data, including signal filtering and denoising;

extracting signal characteristics from the preprocessed data, and constructing characteristic vectors to obtain a data set;

Dividing the data set into a training set and a verification set by adopting cross verification, and using the training set for training the mixed model; training the mixed model by adopting an Adam optimizer, and adjusting the super parameters of the model; the training of the mixed model is stopped when the loss reaches the preset requirement by monitoring the accuracy and the loss of the training process;

and evaluating the performance of the model based on the test set, and adjusting and optimizing the structure and parameters of the hybrid model based on the performance evaluation result.

4. The automatic wireless code alignment connection method of NVR and IPC according to claim 1, wherein the NVR and IPC device exchanges generated dynamic keys based on a preliminary communication link, and the authentication of both party information includes:

the NVR and IPC devices start a TLS handshake based on the established preliminary communication link;

NVR sends a supported encryption algorithm list to IPC equipment; the IPC device selects a matched encryption algorithm and sends a digital certificate to the NVR, wherein the digital certificate comprises a public key of the IPC device, and the digital certificate is signed by a trusted certificate authority;

NVR verifies the certificate validity of the IPC device, including whether the signature, certificate chain, and public key in the certificate actually represent the IPC device; after verification is successful, ECDH key exchange is carried out;

the NVR and the IPC equipment respectively generate respective ECDH public-private key pairs, public key exchange is carried out on the NVR and the IPC equipment after generation, and the same shared secret key is calculated through an ECDH protocol by respectively using the public key of the opposite party and the private key of the opposite party;

The NVR and IPC devices derive a session key using the shared secret key for subsequent communications encryption and message authentication.

5. The automatic wireless code alignment connection method for NVR and IPC according to claim 1, wherein said intelligently selecting the best connection band comprises:

Testing each available radio frequency band, and transmitting and receiving a preset data packet on each frequency band by NVR and IPC equipment;

Monitoring signal strength, noise level and interference index on each frequency band;

calculating a frequency band utility value based on the monitored signal strength, noise level, and interference index:

；

Wherein: a utility value representing the frequency band f; /(I) Signal strength representing frequency band f; /(I)A noise level representing frequency band f; /(I)An interference index representing the frequency band f; /(I)、/>/>Representing the weight factor;

and calculating the frequency band utility value of each acquired available frequency band, and selecting the frequency band with the highest score.

6. The method for automatically wireless code alignment connection between NVR and IPC of claim 1, wherein said automatically adjusting transmission power comprises:

adjusting the transmission power according to the selected frequency band and the current network condition:

；

Wherein: representing the transmission power at a distance d and a frequency f; /(I) Representing a baseline power at a reference distance; /(I)Representing a frequency-dependent environment-specific signal attenuation factor; /(I)Representing a distance between the NVR device and the IPC device; /(I)Representing the center frequency corresponding to the currently used frequency band; wherein/>Based on field test under different frequency bands, signal strength under different distances is measured, and then signal attenuation factors/>, are determined。

7. The automatic wireless code alignment connection method of claim 1, wherein the communication parameters include a channel bandwidth and a modulation scheme, and the channel bandwidth is adjusted as follows:

；

Wherein: a specific value representing bandwidth adjustment; /(I) Is the minimum of available bandwidth; /(I)Is the maximum value of available bandwidth; Representing the current signal to noise ratio; /(I) Representing a preset minimum signal-to-noise ratio threshold; /(I)Representing a preset highest signal-to-noise ratio threshold;

the modulation scheme determines a specific modulation scheme according to the current signal-to-noise ratio traversal mapping table; and constructing modulation schemes corresponding to different signal-to-noise ratio intervals in the mapping table.

8. The automatic wireless code matching connection method for NVR and IPC according to claim 1, wherein the intelligent selection of the optimal connection frequency band and the automatic adjustment of the transmission power and the communication parameters are realized through a parameter prediction model arranged in the NVR;

The parameter prediction model adopts a combination of LSTM and CNN models; wherein LSTM is used to analyze time series data and CNN is used to analyze features in the frequency domain;

The training steps of the parameter prediction model are as follows:

The NVR and IPC devices collect network environment and device status data, including: signal strength, noise level, interference index, signal to noise ratio, currently used frequency band, transmission power, channel bandwidth, modulation scheme, battery power of the device and power consumption;

Pretreatment: cleaning and standardizing the collected network environment and equipment state data;

Taking the preprocessed network environment and the preprocessed device state data as a training set, wherein the training set comprises the device state data, the network environment data and the corresponding optimal communication parameter configuration as labels;

Training the parameter prediction model by using a training set, and optimizing model parameters of the parameter prediction model through back propagation and gradient descent;

wherein the output of the parameter prediction model includes the connection frequency band, the transmission power, and the communication parameters.