Integrated multi-network integration satellite portable station

Abstract

The invention discloses an integrated multi-network integration satellite portable station, which belongs to the technical field of wireless communication networks, and comprises the following steps: a radio frequency module acquires satellite link signal quality and ad hoc network node distribution parameters, and generates a network original parameter set; the parameter processing module generates a network state feature vector through multi-dimensional feature extraction, and dynamically calculates a fusion weight value of the satellite and the ad hoc network based on a multi-target optimization model; the data receiving unit generates a multi-network modal scheduling instruction and a coding path according to the weight value, and receives production data and an emergency video stream; the data encapsulation unit encapsulates data in a mixed manner of framing compression and quantum key encryption; and the routing distribution module dynamically distributes the transmission proportion according to the weight. According to the invention, dynamic fusion weight calculation, a closed-loop power adaptive algorithm and a multi-protocol coding path selection mechanism are adopted, so that real-time optimal distribution of network resources, dynamic adaptation of transmitting power and efficient and safe transmission of mixed data streams can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless communication networks, and in particular to an integrated multi-network fusion satellite portable station. Background Art

[0002] In emergency communications, field operations, and military fields, portable satellite stations, as critical communications equipment, shoulder the crucial task of ensuring information transmission. This role is particularly prominent in environments where ground communication infrastructure is lacking or damaged. To meet diverse communication needs, portable satellite stations typically need to support hybrid networking, combining satellite links with ad hoc networks. However, traditional portable satellite stations often use independent modules to handle communication tasks for different networks, rely on manual network switching, and have rigid data encapsulation and routing strategies. These make it difficult to flexibly adapt to business needs in dynamic and complex environments, limiting their effectiveness in practical applications.

[0003] In existing technologies, some solutions achieve multi-network access through an overlay design, such as physically integrating satellite communication modules with ad hoc network modules and using static weight allocation to mix data traffic. Other solutions trigger link switching based on fixed thresholds, such as selecting the primary link based solely on a signal strength threshold. Regarding data security, conventional solutions typically use the same encryption algorithm or a single key distribution mechanism for both satellite and ad hoc network links. Power control often employs open-loop compensation, such as presetting the transmit power level based on historical experience. While these technologies have improved the feasibility of multi-network integration to a certain extent, they still have significant drawbacks: static weight allocation cannot respond in real time to fluctuations in satellite link signal quality and changes in ad hoc network node topology, resulting in low network resource utilization; module-based data encryption mechanisms increase processing delays and are prone to security vulnerabilities due to key synchronization errors; and preset transmit power modes make it difficult to eliminate the dynamic effects of precipitation attenuation and multipath scattering in the millimeter wave band, resulting in reduced communication link stability.

[0004] To sum up, the existing technology has obvious deficiencies in terms of adaptability to dynamic environments, data security, and power control flexibility. There is an urgent need for an integrated multi-network integrated satellite portable station solution that can achieve real-time response, efficient resource utilization, reliable security, and dynamic power adjustment to meet communication needs in complex scenarios. Summary of the Invention

[0005] To solve the above problems, the present invention provides an integrated multi-network fusion satellite portable station, which adopts dynamic fusion weight calculation, closed-loop power adaptation algorithm and multi-protocol coding path selection mechanism to achieve real-time optimal allocation of network resources, dynamic adaptation of transmission power and efficient and secure transmission of mixed data streams.

[0006] The above objectives can be achieved through the following solutions:

[0007] An integrated multi-network fusion satellite portable station, comprising a radio frequency module, a parameter processing module, a transmission data processing module, a power control module, and a routing distribution module, wherein the transmission data processing module comprises a data receiving unit and a data encapsulation unit; wherein the radio frequency module is used to obtain the signal quality parameters of the satellite communication link and the distribution density parameters of the ad hoc network nodes to generate a network original parameter set; the parameter processing module is used to perform multi-dimensional feature extraction on the network original parameter set to generate a network state feature vector; and calculate the dynamic fusion weight value of the satellite link and the ad hoc network link based on the network state feature vector; the data receiving unit is used to generate a multi-network module based on the dynamic fusion weight value. dynamic scheduling instructions, and trigger a preset modulation and coding strategy selection operation to establish a multi-protocol dynamic coding path; and receive the production data stream and the emergency video stream through the multi-protocol dynamic coding path; the data encapsulation unit is used to securely encapsulate the production data stream and the emergency video stream to generate a mixed transmission data frame; the routing allocation module is used to allocate the transmission ratio of the mixed transmission data frame to the satellite communication link and the ad hoc network link according to the dynamic fusion weight value; the power control module is used to adjust the transmission power of the radio frequency module according to a preset closed-loop power adaptive algorithm; the radio frequency module is also used to distribute the mixed transmission data frame according to the transmission ratio and the transmission power.

[0008] Optionally, performing multi-dimensional feature extraction on the network original parameter set to generate a network status feature vector includes: extracting the signal-to-noise ratio mean of the satellite link and the hop count distribution matrix of the ad hoc network node from the network original parameter set; inputting the signal-to-noise ratio mean and the hop count distribution matrix into a preset spatiotemporal correlation neural network model to extract a transmission delay sensitivity evaluation index; constructing a space-frequency joint mapping relationship table based on the transmission delay sensitivity evaluation index to generate a network status feature vector.

[0009] Optionally, the extraction of the transmission delay sensitivity evaluation index includes: establishing a spatiotemporal correlation neural network model, the spatiotemporal correlation neural network model including a first feature channel, a second feature channel, and a hidden layer; inputting the signal-to-noise ratio mean into the first feature channel to extract frequency domain periodic features to obtain a satellite link frequency domain feature vector; inputting the hop count distribution matrix into the second feature channel to perform node relationship feature mining to obtain a self-organizing network topology feature vector; and fusing the satellite link characteristics and the self-organizing network topology features through a space-frequency joint convolution kernel preset in the hidden layer to obtain a transmission delay sensitivity evaluation index.

[0010] Optionally, the calculation of the dynamic fusion weight value of the satellite link and the ad hoc network link based on the network status feature vector includes: extracting the satellite link attenuation factor and the ad hoc network path redundancy index from the network status feature vector; calculating the attenuation factor and the redundancy index based on a pre-trained multi-objective optimization model, and outputting a multi-network bearer priority sequence; generating and updating the dynamic fusion weight value according to the multi-network bearer priority sequence.

[0011] Optionally, the training process of the multi-objective optimization model is: constructing a loss function including satellite link capacity, ad hoc network throughput and end-to-end delay indicators; using a non-dominated sorting genetic algorithm to perform a Pareto front search on the loss function to generate an optimal weight distribution under multi-parameter constraints.

[0012] Optionally, the data encapsulation unit includes: a production data processing subunit, used to perform adaptive frame compression on the production data stream to generate a compressed data packet set; an emergency video processing subunit, used to input the emergency video stream into a preset quantum key distribution edge terminal to generate a dynamic encryption factor; and a hybrid encapsulation subunit, used to securely encapsulate the compressed data packet set and the dynamic encryption factor to generate a hybrid transmission data frame.

[0013] Optionally, the secure encapsulation of the compressed data packet set and the dynamic encryption factor to generate a mixed transmission data frame includes: performing XOR interleaving encoding on the compressed data packet set and the dynamic encryption factor output by the quantum key distribution edge terminal to generate an intermediate data stream; performing secondary encryption using the intermediate data stream to generate a scrambling code sequence; and embedding the scrambling code sequence into a preset Beidou positioning synchronization information field to form a mixed transmission data frame.

[0014] Optionally, the secondary encryption using the intermediate data stream to generate a scrambling code sequence includes: selecting an encryption mode from a preset national encryption algorithm library according to the resolution of the current emergency video stream; adjusting encryption parameters according to the encapsulation progress; encrypting the intermediate data stream according to the selected encryption mode and encryption parameters to generate a scrambling code sequence.

[0015] Optionally, adjusting the transmission power of the RF module according to a preset closed-loop power adaptive algorithm includes: real-time monitoring of the equivalent isotropic radiated power of the satellite link and the path loss gradient of the ad hoc network link; generating a RF dynamic gain adjustment signal based on the equivalent isotropic radiated power and the path loss gradient; synchronizing the RF dynamic gain adjustment signal to the bias circuit of the RF module and outputting a polarization matching power signal.

[0016] Optionally, generating a radio frequency dynamic gain adjustment signal based on the equivalent isotropic radiated power and the path loss gradient includes: deploying a dynamic channel estimator at the receiving end of the satellite link to extract the scattering fading component of the propagation path in real time; performing nonlinear fitting on the scattering fading component with a preset precipitation attenuation model to generate a power compensation coefficient for the millimeter wave band; and generating a radio frequency dynamic gain adjustment signal using the power compensation coefficient, the equivalent isotropic radiated power and the path loss gradient.

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] 1. Improve network resource utilization and transmission efficiency. By dynamically merging weights to adjust the transmission ratio of satellite links and ad hoc network links in real time, combined with multi-dimensional feature extraction and multi-objective optimization models, it can adapt to channel fluctuations and node topology changes, eliminating resource idleness and congestion in traditional static allocation modes, and significantly improving hybrid network throughput.

[0019] 2. Enhance data security and packaging efficiency; adopt a differentiated packaging mechanism that combines frame compression of production data streams with dynamic encryption of emergency video traffic subkeys. Cross-protocol joint security protection is achieved through XOR interleaving encoding and secondary encryption, avoiding key synchronization errors in module-based encryption and shortening security processing delays.

[0020] 3. Optimize power control stability and link reliability. Dynamically adjust RF power using a closed-loop adaptive algorithm based on equivalent isotropic radiated power and path loss gradient. Combined with millimeter-wave band precipitation attenuation compensation and path scattering suppression, this reduces the probability of communication interruption and ensures continuous communication capabilities in complex environments.

[0021] 4. Support multi-protocol dynamic encoding and real-time business carrying; through the coordinated control of modulation and coding strategies and multi-network mode scheduling, establish encoding paths that adapt to different network characteristics, realize low-latency parallel transmission of production data and emergency video streams, and meet the high timeliness requirements of field operations and emergency communications.

[0022] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 This is a framework diagram of an integrated multi-network fusion satellite portable station according to an embodiment of the present invention.

[0025] Figure 2 It is a structural diagram of an integrated multi-network fusion satellite portable station according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0027] Reference Figure 1 One embodiment of the present invention proposes an integrated multi-network fusion satellite portable station, which adopts dynamic fusion weight calculation, closed-loop power adaptation algorithm and multi-protocol coding path selection mechanism to achieve real-time optimization allocation of network resources, dynamic adaptation of transmission power and efficient and secure transmission of mixed data streams.

[0028] The portable station of this embodiment specifically includes: a radio frequency module, a parameter processing module, a transmission data processing module, a power control module, and a routing distribution module. Figure 2 As shown, the transmission data processing module includes a data receiving unit and a data encapsulation unit; wherein,

[0029] The radio frequency module is used to obtain the signal quality parameters of the satellite communication link and the distribution density parameters of the ad hoc network nodes to generate the network original parameter set;

[0030] Specifically, the RF module uses a built-in spectrum analyzer to collect signal quality parameters of the satellite communication link in real time. It also uses a distributed node detector to obtain the distribution density parameters of the ad hoc network nodes. These two parameters are aligned by timestamp and stored as a set of network primitive parameters. For example, signal quality parameters include the satellite link's signal-to-noise ratio (SNR), bit error rate (BER), and carrier power fluctuation. The distribution density parameter is calculated by counting the number of active ad hoc network nodes and the average number of hops between adjacent nodes within a unit area. The spectrum analyzer uses a sweep mode to capture the frequency domain characteristics of the satellite communication link, while the distributed node detector periodically sends probe messages to collect the response delays and topological relationships of the ad hoc network nodes. The set of network primitive parameters is stored in a matrix format, with rows corresponding to sampling times and columns recording the specific values of the satellite signal quality parameters and the ad hoc network distribution density parameters, each with a geographic location tag. Signal quality parameters are technical indicators obtained by measuring the physical layer transmission characteristics of the satellite communication link, including quantized root mean square (RMS) signal-to-noise ratio (SNR) and modulation error rate (MER). The distribution density parameter is a statistical quantity reflecting the spatial distribution of ad hoc network nodes, specifically a joint representation of the number of communicable nodes per square kilometer and the average number of path hops. The original network parameter set is a structured data set generated by the radio frequency module and contains satellite communication link and ad hoc network node parameters, which is used for subsequent feature extraction.

[0031] The parameter processing module is used to extract multi-dimensional features from the original network parameter set to generate a network state feature vector; and calculate the dynamic fusion weight value of the satellite link and the ad hoc network link based on the network state feature vector;

[0032] Optionally, performing multi-dimensional feature extraction on the original network parameter set to generate a network state feature vector includes:

[0033] Extracting the signal-to-noise ratio mean of the satellite link and the hop count distribution matrix of the ad hoc network nodes from the network original parameter set;

[0034] Specifically, the parameter processing module separates the satellite link's signal-to-noise ratio sequence from each row of data in the network's original parameter set, and calculates the signal-to-noise ratio mean using a sliding window averaging method. Simultaneously, it traverses the hop count statistics of the ad hoc network nodes, constructs a hop count distribution matrix with the source node and the target node as row and column indexes, and performs normalization. For example, the network's original parameter set contains satellite signal-to-noise ratio data for 100 consecutive moments. The parameter processing module calculates the moving average signal-to-noise ratio using a 10-second window to reduce short-term channel noise interference. The diagonal elements in the hop count distribution matrix represent situations where the number of direct hops between two nodes is 1, and the non-diagonal elements record the maximum number of hops in the multi-hop path between nodes. The signal-to-noise ratio mean is a satellite link quality indicator that has been smoothed over time, and its calculation formula is:

[0035]

[0036] Among them, μ SNR is the mean signal-to-noise ratio of the satellite link, SNR t represents the signal-to-noise ratio value measured at the tth second, and T is the sliding window duration. The hop count distribution matrix is a two-dimensional square matrix generated by the statistics of the communication hop counts of the ad hoc network nodes, and the element H i,j It represents the cumulative number of hops from source node i to target node j, which is divided by the global maximum hop number H max Normalize to get H i,j / H max Normalization refers to mapping the original hop count to the interval [0,1] to eliminate statistical deviations of networks of different sizes.

[0037] Inputting the signal-to-noise ratio mean and the hop count distribution matrix into a preset spatiotemporal correlation neural network model to extract a transmission delay sensitivity evaluation index;

[0038] Optionally, extracting a transmission delay sensitivity evaluation index includes:

[0039] Establishing a spatiotemporal correlation neural network model, wherein the spatiotemporal correlation neural network model includes a first feature channel, a second feature channel, and a hidden layer;

[0040] Inputting the signal-to-noise ratio mean into the first feature channel to extract frequency domain periodic features, thereby obtaining a satellite link frequency domain feature vector;

[0041] Inputting the hop count distribution matrix into the second feature channel to perform node relationship feature mining to obtain a self-organizing network topology feature vector;

[0042] The satellite link characteristics and the ad hoc network topology features are fused through the space-frequency joint convolution kernel preset in the hidden layer to obtain a transmission delay sensitivity evaluation index.

[0043] Specifically, the mean signal-to-noise ratio and the hop count distribution matrix are input into a spatiotemporal correlation neural network model. The model input layer includes two parallel feature channels: the first channel performs Fourier transform on the mean signal-to-noise ratio sequence to extract frequency-domain periodic features to obtain the satellite link frequency-domain feature vector, and the second channel uses a graph convolutional network to mine node relationship features on the hop count distribution matrix to obtain the ad hoc network topology feature vector. In the hidden layer, the spatiotemporal correlation neural network model fuses the satellite link frequency-domain feature vector and the ad hoc network topology feature vector using a preset space-frequency joint convolution kernel to obtain the transmission delay sensitivity evaluation index F. fusion , and its operation process is:

[0044]

[0045] Among them, F sat is the satellite link frequency domain eigenvector, F meshis the topology feature vector of the ad hoc network, represents the feature concatenation operation, W f is the trainable parameter matrix of the joint space-frequency convolution kernel, and σ is the activation function. The cross-protocol delay sensitivity feature is an end-to-end transmission delay sensitivity evaluation indicator generated by integrating the differences between satellite links and ad hoc network protocols. It reflects the hybrid network's ability to carry real-time services.

[0046] A space-frequency joint mapping relationship table is constructed according to the transmission delay sensitivity evaluation index to generate a network state feature vector.

[0047] Specifically, a joint space-frequency mapping table is established based on these features. The network state at each moment is encoded in vector form as a combination of the satellite link frequency band index, the ad hoc network topology feature dimension, and the delay sensitivity evaluation value, ultimately forming a network state feature vector. The joint space-frequency mapping table is a relational data structure that associates spectrum resources with network topology status. Each row records the optimal matching pattern between spectrum utilization efficiency and node distribution at the current moment.

[0048] For example, the mean signal-to-noise ratio (SNR) extracted from the original network parameter set is 18dB, and the hop count distribution matrix contains communication hop count data for five nodes. After normalization, the maximum element in the matrix is 0.8. In the spatiotemporal correlation neural network model, the mean SNR sequence undergoes Fourier transform to obtain a frequency-domain feature vector with a dominant frequency component of 2Hz. The hop count distribution matrix is then processed through a graph convolutional network to extract the critical path features of nodes 3 and 5. A joint space-frequency convolution kernel concatenates these two features to generate a fused vector, and the activation function outputs a cross-protocol delay sensitivity feature value of 0.73. The corresponding rows in the joint space-frequency mapping table store the satellite link Ku-band frequency, the path features of nodes 3-5, and the delay sensitivity of 0.73, forming the three core dimensions of the current network state feature vector.

[0049] Optionally, the calculating the dynamic fusion weight value of the satellite link and the ad hoc network link according to the network state feature vector includes:

[0050] Extracting a satellite link attenuation factor and an ad hoc network path redundancy index from the network state feature vector;

[0051] Specifically, after obtaining the network state eigenvector, the dynamic fusion weights of the satellite link and the ad hoc network link are calculated. First, the satellite link attenuation factor and the ad hoc network path redundancy index are parsed from the network state eigenvector. The satellite link attenuation factor is calculated based on the frequency band index and space-frequency joint mapping table recorded in the eigenvector, representing the current satellite link signal attenuation. The ad hoc network path redundancy index is determined through connectivity analysis of the hop count distribution matrix and reflects the number of available alternative paths between nodes.

[0052] Calculating the attenuation factor and redundancy index based on a pre-trained multi-objective optimization model, and outputting a multi-network bearer priority sequence;

[0053] Optionally, the training process of the multi-objective optimization model is:

[0054] Construct a loss function that includes satellite link capacity, ad hoc network throughput, and end-to-end delay metrics;

[0055] A non-dominated sorting genetic algorithm is used to perform Pareto front search on the loss function to generate the optimal weight distribution under multi-parameter constraints.

[0056] Specifically, the satellite link attenuation factor and the MANET path redundancy metric were input into a pre-trained multi-objective optimization model. The model learned Pareto optimal solutions from a historical dataset and output a sequence containing the priorities of satellite and MANET links. During the training process, a set of loss functions for satellite link capacity, MANET throughput, and end-to-end latency was constructed. A non-dominated sorting genetic algorithm was used to perform multi-objective trade-offs and generate an optimal weight distribution that satisfied multiple parameter constraints.

[0057] Specifically, to achieve multi-objective optimization, the non-dominated sorting genetic algorithm (NSGA-II) was employed during training. This algorithm, a multi-objective optimization algorithm based on a genetic algorithm, effectively balances multiple objectives to find a set of optimal solutions. The Pareto front is a set of optimal solutions in a multi-objective optimization problem, where any improvement in any objective inevitably leads to degradation of other objectives. During training, the NSGA-II algorithm performs a Pareto front search on the loss function, aiming to find a set of weight parameters that achieves the optimal combination of satellite link capacity, ad hoc network throughput, and end-to-end delay under given conditions. During training, a set of loss functions is first constructed that encompasses satellite link capacity, ad hoc network throughput, and end-to-end delay. The NSGA-II algorithm is then used to optimize the loss functions, and a set of weight parameters is found through an iterative search. These parameters are Pareto-optimal, meaning that no objective can be further improved without sacrificing the others. Ultimately, the resulting optimal weight distribution under multi-parameter constraints is used to guide portable stations in dynamically allocating weights for integrating satellite and ad hoc links in actual operation.

[0058] Exemplarily, the portable station calculates and updates the dynamic fusion weight value based on the current network state feature vector (including the satellite link attenuation factor and the ad hoc network path redundancy index) and the pre-trained multi-objective optimization model. For example, if the satellite link attenuation factor extracted from the network state feature vector is 0.35, the ad hoc network path redundancy index is 0.82, the satellite link capacity in the pre-trained model is set to 150Mbps, and the ad hoc network throughput is 75Mbps, then according to the formula, the satellite link weight is calculated to be 0.82, and the ad hoc network weight is 0.18. Through this training process, the multi-objective optimization model can dynamically adjust the fusion weights of the satellite link and the ad hoc network link to adapt to the complex and changing network environment and achieve efficient utilization of network resources.

[0059] A dynamic fusion weight value is generated and updated according to the multi-network bearer priority sequence.

[0060] Specifically, the fusion weights of satellites and ad hoc networks are dynamically allocated according to the multi-network bearer priority sequence. The calculation formula is:

[0061]

[0062] w mesh =1-w sat ,

[0063] Among them, w sat and w mesh are the fusion weights of satellite and ad hoc network, C sat represents the satellite link capacity, R red is the path redundancy index, C mesh is the throughput of the ad hoc network, A att is the attenuation factor. This formula eliminates dimensional differences through normalization to ensure the rationality of weight distribution.

[0064] For example, the satellite link attenuation factor extracted from the network state feature vector is 0.35, and the ad hoc network path redundancy index is 0.82. The satellite link capacity in the pre-trained model is set to 150 Mbps, and the ad hoc network throughput is 75 Mbps. The satellite link weight w is calculated according to the formula sat =0.82, ad hoc network weight w mesh =0.18. The multi-network bearer priority sequence determines that the satellite link is the preferred channel. After the dynamic fusion weight is updated, the routing allocation module allocates 82% of the mixed transmission data frame traffic to the satellite link, and the remaining 18% is transmitted through the ad hoc network.

[0065] The data receiving unit is configured to generate a multi-network modality scheduling instruction based on the dynamic fusion weight value, trigger a preset modulation and coding strategy selection operation, establish a multi-protocol dynamic coding path, and receive a production data stream and an emergency video stream through the multi-protocol dynamic coding path;

[0066] Specifically, when the data receiving unit generates a multi-network modal scheduling instruction based on the dynamic fusion weight value, it first determines the dual-network concurrent transmission ratio according to the satellite link weight and the ad hoc network link weight in the dynamic fusion weight value. For example, if the dynamic fusion weight value is 0.7 for satellite and 0.3 for ad hoc network, the scheduling instruction allocates 70% of the service flow to the satellite link and 30% to the ad hoc network link. Then the preset modulation and coding strategy selection operation is triggered. The strategy library stores the modulation level and coding rate combinations that match different network quality levels. The optimal strategy is indexed in real time based on the satellite link signal-to-noise ratio mean value and the ad hoc network hop count distribution matrix in the current network state feature vector. When establishing a multi-protocol dynamic coding path, the protocol stack reconstruction technology is used to synchronously adapt the convolutional code interleaving method of satellite communication and the LDPC code block structure of the ad hoc network, and configure the hybrid ARQ retransmission mechanism. The formula is expressed as follows:

[0067]

[0068] Among them, MCS index is the modulation and coding strategy index, H avg is the average number of hops per node in the ad hoc network, and argmax is the function that maximizes the independent variable, i.e., returns the input value (independent variable) that maximizes the function. This formula eliminates dimensional differences in the number of hops through an exponential function, ensuring the mathematical compatibility of the joint indicator.

[0069] For example, the dynamic fusion weight value sets the satellite link weight to 0.7 and the ad hoc network weight to 0.3. In the network status feature vector output by the parameter processing module, the satellite link signal-to-noise ratio average is 20dB and the average number of hops of the ad hoc network node is 3. The MCS is calculated according to the formula index = 5, corresponding to the QPSK modulation and 1 / 2 bitrate combination at index 5 in the policy library. In the multi-protocol dynamic coding path established by the data receiving unit, the satellite link uses QPSK modulation and deep interleaving coding of the DVB-S2X protocol stack, while the ad hoc network link uses LDPC coding of the IEEE802.11s protocol. Sensor messages from the production data stream and the H.265 bitstream of the emergency video stream are synchronously received through a dynamic priority queue.

[0070] The data encapsulation unit is used to securely encapsulate the production data stream and the emergency video stream to generate a mixed transmission data frame;

[0071] Alternatively, as Figure 2 As shown, the data encapsulation unit includes:

[0072] A production data processing subunit, configured to perform adaptive framing compression on the production data stream to generate a set of compressed data packets;

[0073] An emergency video processing subunit, configured to input the emergency video stream into a preset quantum key distribution edge terminal to generate a dynamic encryption factor;

[0074] Specifically, the data encapsulation unit performs differentiated and secure encapsulation on the production data stream and emergency video stream received by the multi-protocol dynamic encoding path. First, the production data processing subunit performs adaptive frame compression based on the sampling rate and priority tag of the data stream, dividing the continuous data stream into a frame sequence of fixed length. The quantization coefficient of the discrete cosine transform is dynamically adjusted according to the preset compression ratio threshold to generate a set of compressed data packets. The compression ratio is calculated as follows:

[0075]

[0076] Among them, R comp is the actual compression ratio, B orig is the average bit rate of the original data stream, B thresh is the link capacity threshold, R max is the preset maximum compression rate. This formula ensures that the compression strength dynamically matches the current network load. At the same time, the emergency video processing subunit separates the I-frame and P-frame of each video clip in the emergency video stream, inputs them into the quantum key distribution edge terminal, and uses a time-phase coded quantum random number generator to generate a dynamic encryption factor. The length of the encryption factor is proportional to the video resolution, and the formula is:

[0077]

[0078] Among them, L key is the length of the dynamic encryption factor, W frame With H frame is the width and height of the video frame in pixels, T seg The length of the video clip. Each encryption factor is bound to the precise timestamp output by the BeiDou timing module to achieve spatiotemporal synchronization of the encryption key.

[0079] For example, the production data stream is industrial sensor data with a sampling rate of 1kHz and an original bit rate of B orig =8Mbps, when the link capacity threshold B thresh =10Mbps, substitute into the formula to calculate R comp =0.16, then a compression rate of 16% is used to perform frame compression, generating a compressed data packet of 150ms duration. The resolution of the emergency video stream is 1920×1080, and the video clip duration is T seg = 2 seconds, then the dynamic encryption factor length Lkey = 4050 bytes. The quantum key distribution edge terminal generates a 4050-byte quantum random number sequence based on this length and binds it to the 2023-08-15T14:30:00Z timestamp of the Beidou timing module.

[0080] The hybrid encapsulation subunit is used to securely encapsulate the compressed data packet set and the dynamic encryption factor to generate a hybrid transmission data frame.

[0081] Optionally, securely encapsulating the compressed data packet set and the dynamic encryption factor to generate a mixed transmission data frame includes:

[0082] Performing XOR interleaving encoding on the compressed data packet set and the dynamic encryption factor output by the quantum key distribution edge terminal to generate an intermediate data stream;

[0083] Specifically, when the hybrid encapsulation subunit securely encapsulates a set of compressed data packets and a dynamic encryption factor, it first divides the binary stream of the compressed data packet into preset byte blocks and performs a bit-by-bit XOR operation with the quantum random number sequence of the dynamic encryption factor. For example, the compressed data packet is divided into 512-byte data blocks, and the dynamic encryption factor is sliced into pieces of the same length. The two are then XORed by position to generate encrypted data segments. At the same time, an interleaving coefficient matrix is introduced to adjust the code element distribution of the XOR result to generate an intermediate data stream. The calculation formula is:

[0084]

[0085] Among them, C i is the i-th codeword after encryption, D j is the jth byte of the compressed data packet, K is the dynamic encryption factor sequence, L key is the encryption factor length, M is an n×m-dimensional interleaving coefficient matrix, and each element is an integer multiple of the interleaving offset. This formula uses modular operations to ensure data alignment and eliminate the impact of length differences. The matrix element values are set to the geographic grid code extracted from the Beidou positioning synchronization information field.

[0086] For example, the size of a data packet in the compressed data packet set is 512 bytes, and the length of the dynamic encryption factor is 4050 bytes. The encryption factor is used cyclically by bytes during the XOR operation. When j=500, K (500mod4050) =500, and D 500 XOR generates the intermediate code element. The interleaving coefficient matrix M is an 8×8 matrix, and the element values are taken from the last two decimal places of the Beidou positioning coordinates, for example, M (500mod8,500mod8) =M (4,4) = 12. The final generated C 500 The code element is the value of the XOR result shifted left by 12 bits, forming an intermediate data stream.

[0087] Performing secondary encryption using the intermediate data stream to generate a scrambling sequence;

[0088] Optionally, the performing secondary encryption using the intermediate data stream to generate a scrambling sequence includes:

[0089] According to the resolution of the current emergency video stream, select the encryption mode from the preset national encryption algorithm library;

[0090] Adjust encryption parameters according to packaging progress;

[0091] The intermediate data stream is encrypted according to the selected encryption mode and encryption parameters to generate a scrambling code sequence.

[0092] The scrambling code sequence is embedded in a preset Beidou positioning synchronization information field to form a mixed transmission data frame.

[0093] Specifically, the intermediate data stream after XOR interleaving enters the dynamic key encryption protocol processor, where a nonlinear scrambling sequence is generated according to the following steps: First, the SM4 encryption mode is dynamically selected based on the resolution of the emergency video stream. When the resolution reaches 1920×1080 or above, the hardware acceleration engine in SM4-CTR mode is invoked. SM4-CTR mode is a counter-based national secret algorithm operating mode that converts the block cipher into a stream cipher, enabling parallel encryption processing. When the resolution falls below this threshold, the engine switches to SM4-CBC mode. For example, a 2560×1440 emergency video stream transmitted by a drone triggers the selection of CTR mode, utilizing a parallel pipeline to reduce encryption latency. The encryption parameters are then adjusted based on the encapsulation progress of the mixed data frame. If the current encapsulation progress is less than 50% and the data frame length exceeds 1024 bytes, 32 rounds of encryption are used with an all-zero initialization vector. When the encapsulation progress reaches or exceeds 50%, the encryption switches to 16 rounds and the PKCS7 padding strategy is applied. The PKCS7 padding strategy is a standardized padding scheme where each padding byte is the number of bytes to be padded, ensuring that the data length aligns with the packet size. The filling length calculation formula is:

[0094] Pad len =Block size -(Data len mod Block size ),

[0095] Among them, Pad len The number of bytes to fill, Block size Fixed 16-byte block size for SM4 algorithm, Data len is the byte length of the current data segment. After encryption is completed, the generated scrambling sequence is reorganized according to the Beidou positioning synchronization information format, and the embedding formula is defined as:

[0096] Frame=Scrambled data ‖UTC time ‖CRC32(Scrambled data ),

[0097] Among them, Scrambled data is a nonlinear scrambling sequence, UTC time This is the 8-byte BCD timestamp output by the Beidou timing module. ‖ indicates byte-level concatenation. A CRC32 checksum is appended to the end of the frame. The BCD timestamp is in binary-coded decimal format, with each decimal digit represented by 4 bits. For example, 14:30 is encoded as 0001010000110000. The CRC32 checksum is a 32-bit cyclic redundancy check (CRC) code, calculated using the IEEE 802.3 standard polynomial, and is used to detect bit errors in data transmission.

[0098] For example, the length of the XOR-interleaved encoded data stream is 1500 bytes, and the emergency video stream resolution is 2560×1440, triggering SM4-CTR mode. The dynamic key encryption protocol processor detects that the current encapsulation progress is 30% and the frame length exceeds the threshold, enabling 32 rounds of encryption and zero-padding the initial vector. The first 16 bytes of the encrypted scrambling sequence are 0x3A5F…E9. The Beidou time output of 2023-08-15T14:30:00Z corresponds to the BCD code 0x20230815143000. The combined data frame is constructed according to the formula: the first 1500 bytes of scrambled data, the middle 8 bytes of timestamp, and the last 4 bytes of CRC32 checksum 0xA1B2C3D4.

[0099] The routing allocation module is used to allocate the transmission ratio of the mixed transmission data frame to the satellite communication link and the ad hoc network link according to the dynamic fusion weight value;

[0100] Specifically, the routing distribution module receives the dynamic fusion weight value generated by the parameter processing module, which includes the satellite communication link weight and the ad hoc network link weight. According to the proportional relationship between the satellite communication link weight and the ad hoc network link weight, the mixed transmission data frame is split into satellite data fragments and ad hoc network data fragments. The specific execution process is as follows: first, the frame length of the mixed transmission data frame is analyzed, and the satellite link weight w in the dynamic fusion weight value is used to calculate the mixed transmission data frame length. sat and the link weight w of the ad hoc network mesh The ratio of the two is used to calculate the number of fragmented bytes. For the satellite link byte number T sat and the number of bytes of the ad hoc network link T mesh ,have:

[0101] T sat =w sat *Ltotal ,

[0102] T mesh =L total -T sat ,

[0103] Among them, L total is the total length of the mixed transmission data frame. The dual-channel queue scheduler then sends the satellite data fragments to the protocol encapsulation interface of the satellite communication link, distributing the ad hoc network data fragments to the multi-hop routing forwarding module of the ad hoc network nodes. The dynamic fusion weight value is the weight parameter output by the parameter processing module based on the satellite link attenuation factor and the ad hoc network path redundancy index through a multi-objective optimization model. It represents the load capacity of the satellite communication link and the ad hoc network link under the current network conditions. The mixed transmission data frame is a complete transmission unit generated by the data encapsulation unit, containing compressed production data and encrypted emergency video. Its frame structure embeds the Beidou positioning synchronization information field. The dual-channel queue scheduler is the execution unit for traffic segmentation and uses a dynamic token bucket algorithm to limit the burst traffic peak of the two types of fragments.

[0104] For example, if the satellite link weight w in the dynamic fusion weight value sat =0.82 and the link weight of the ad hoc network is w mesh =0.18, total length of mixed transmission data frame L total =1500 bytes. The routing module is based on T sat =0.82×1500=1230 bytes, T mesh =1500-1230=270 bytes allocated fragments. The dual-channel queue scheduler marks the 1230-byte satellite data fragment as high priority and assigns it to the DVB-S2X protocol stack's transmit buffer using a weighted fair queue scheduling algorithm. Simultaneously, the 270-byte ad hoc network data fragment is encapsulated as an IEEE802.11s mesh packet and forwarded via nodes 5 and 3 specified in the multi-hop routing table. The fragmentation process preserves the Beidou timestamp field of the original mixed transmission data frame, ensuring time synchronization and reconstitution between the receiving devices at both ends.

[0105] The power control module is used to adjust the transmit power of the radio frequency module according to a preset closed-loop power adaptive algorithm;

[0106] Optionally, adjusting the transmit power of the radio frequency module according to a preset closed-loop power adaptive algorithm includes:

[0107] Real-time monitoring of the equivalent isotropic radiated power of satellite links and the path loss gradient of ad hoc network links;

[0108] Specifically, the power control module continuously collects the equivalent isotropically radiated power (EIRP) measurement value of the satellite link through the feedback circuit embedded in the RF front end, and calculates the path loss gradient through the interactive channel sounding signal of the ad hoc network node. Among them, the EIRP is obtained by integrating the spectrum analysis data output by the satellite signal receiver with the carrier power. For the EIRP, we have:

[0109]

[0110] Where, P rx (t) is the received power at the t-th sampling moment (unit: dBm), L fs (t) is the free space path loss, which can be, for example: Where d(t) is the dynamic distance between the satellite and the terminal, and λ is the carrier wavelength. The path loss gradient is calculated by linearly fitting the received signal strength indicator (RSSI) sequence between the nodes of the ad hoc network. curr ,have:

[0111]

[0112] Where RSSI(x) is the received signal strength indicator.

[0113] generating a radio frequency dynamic gain adjustment signal according to the equivalent isotropic radiated power and the path loss gradient;

[0114] Optionally, the generating a radio frequency dynamic gain adjustment signal according to the equivalent isotropically radiated power and the path loss gradient includes:

[0115] Deploy a dynamic channel estimator at the receiving end of the satellite link to extract the scattering fading component of the propagation path in real time;

[0116] Performing nonlinear fitting on the scattering fading component and a preset precipitation attenuation model to generate a power compensation coefficient for the millimeter wave band;

[0117] A radio frequency dynamic gain adjustment signal is generated by using the power compensation coefficient, the equivalent isotropically radiated power, and the path loss gradient.

[0118] The radio frequency dynamic gain adjustment signal is synchronized to the bias circuit of the radio frequency module to output a polarization matching power signal.

[0119] Specifically, the dynamic channel estimator analyzes the phase noise variance of the satellite signal multipath component in real time to obtain the scattering fading component Input this into the precipitation attenuation model:

[0120] A rain =k*R α *fβ *L,

[0121]

[0122] Where R is the rainfall rate, f is the millimeter wave frequency, L is the propagation path length, k, α, and β are model coefficients, which are empirical parameters related to rainfall type and frequency. comp is the power compensation coefficient, γ is the normalization factor, and together with the above parameters, they generate the gain adjustment signal G:

[0123] G=EIRP ref +B1*C comp *ΔPL curr +B2,

[0124] Where B1 and B2 are gain adjustment coefficients, which control the ratio of the influence of precipitation compensation and path loss gradient on gain. This signal is output to the gate bias voltage controller of the RF power amplifier through the digital-to-analog converter to adjust the polarization matching characteristics of the output power to ensure that the transmission power meets the following requirements:

[0125]

[0126] Among them, P tx is the RF module transmission power, P nom is the nominal power value, B3 is the unit conversion factor, which is the proportional factor required to convert the gain G from the logarithmic domain to the exponential base ψ, where ψ is the exponential base, usually 10, corresponding to ten times the power gain.

[0127] The radio frequency module is further configured to distribute mixed transmission data frames according to the transmission ratio and the transmission power.

[0128] Specifically, the RF module distributes the mixed transmission data frames to the satellite communication link and the ad hoc network link based on the transmission ratio determined by the routing allocation module and the transmit power adjusted by the power control module. The RF module receives satellite data fragments and ad hoc network data fragments from the routing allocation module via a dual-channel queue scheduler and adjusts the RF gain parameters of each fragment based on the transmit power polarization matching power signal. The length of the satellite data fragments and the ad hoc network data fragments is determined by the transmission ratio, and the transmit power is dynamically optimized in real time using a closed-loop power adaptation algorithm. For example, if the transmission ratio is 0.82 for the satellite link and 0.18 for the ad hoc network link, the RF module distributes the first 82% of the bytes of the mixed transmission data frame to the satellite link and the remaining 18% to the ad hoc network link. During the distribution process, the RF module amplifies the satellite link fragments based on the power compensation coefficient in the polarization matching power signal and performs multi-hop path gain compensation on the ad hoc network link fragments based on the path loss gradient to ensure that the equivalent isotropically radiated power of both links meets the preset threshold. Finally, the RF module sends two fragments synchronously through the encapsulation interface of the satellite communication protocol and the multi-hop routing forwarding module, achieving efficient distribution of mixed transmission data frames.

[0129] The transmission ratio refers to the data allocation ratio between the satellite communication link and the ad hoc network link, calculated by the routing allocation module based on the dynamic fusion weight. For example, a satellite link weight of 0.82 corresponds to 82% of the data allocation. Transmit power refers to the RF signal output strength optimized in real time by the power control module using a closed-loop power adaptive algorithm. Its value is determined by the equivalent isotropic radiated power and the path loss gradient. The hybrid transmission data frame refers to the secure data unit generated by the data encapsulation unit, containing a compressed production data stream and an encrypted emergency video stream. Its structure consists of a Beidou positioning synchronization information field, a scrambling code sequence, and a CRC checksum. The dual-channel queue scheduler is the execution unit within the RF module that implements data fragment priority scheduling and flow control. It uses a dynamic token bucket algorithm to limit burst traffic on both links. The polarization-matched power signal refers to the transmit power signal adjusted by the RF module based on the power compensation coefficient. Its parameters dynamically match the polarization characteristics of the satellite channel.

[0130] For example, assuming the routing distribution module outputs a transmission ratio of 0.82 for the satellite link and 0.18 for the ad hoc network link, the power control module adjusts the transmit power to 43dBm for the satellite link and 28dBm for the ad hoc network link. After receiving a mixed transmission data frame with a total length of 1500 bytes, the RF module allocates 1230 bytes to the satellite link and 270 bytes to the ad hoc network link. After encapsulation using the DVB-S2X protocol, the satellite data fragments are boosted to 43dBm by a gain amplifier and sent to the satellite communication link. The ad hoc network data fragments are sent at 28dBm through the multi-hop routing forwarding module, which dynamically compensates for inter-node transmission losses based on the path loss gradient. During the distribution process, the dual-channel queue scheduler monitors the real-time throughput of both links. If the satellite link rate drops due to rain attenuation, the token bucket rate is dynamically adjusted to ensure the coordinated optimization of transmission ratio and power allocation.

[0131] It should be noted that the electrical connections between the above-mentioned units do not necessarily mean direct connections of lines. Indirect connections are applicable to the embodiments of the present invention as long as the purpose of the present invention is achieved. The above description is only an exemplary embodiment of the present invention and is not intended to limit the scope of the present invention.

[0132] That is, any equivalent changes and modifications made according to the teachings of the present invention are still within the scope of the present invention. Those skilled in the art will readily conceive of other embodiments of the present invention after considering the disclosure of the specification and practical truths. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary technical means in the art not described herein.

Claims

1. An integrated multi-network integrated satellite portable station, characterized in that: The portable station includes: a radio frequency module, a parameter processing module, a transmission data processing module, a power control module, and a routing distribution module. The transmission data processing module includes a data receiving unit and a data encapsulation unit. The radio frequency module is used to obtain the signal quality parameters of the satellite communication link and the distribution density parameters of the ad hoc network nodes to generate the network original parameter set; The parameter processing module is used to extract multi-dimensional features from the original network parameter set to generate a network state feature vector; and calculate the dynamic fusion weight value of the satellite link and the ad hoc network link based on the network state feature vector; The data receiving unit is configured to generate a multi-network modality scheduling instruction based on the dynamic fusion weight value, trigger a preset modulation and coding strategy selection operation, establish a multi-protocol dynamic coding path, and receive a production data stream and an emergency video stream through the multi-protocol dynamic coding path; The data encapsulation unit is used to securely encapsulate the production data stream and the emergency video stream to generate a mixed transmission data frame; The routing allocation module is used to allocate the transmission ratio of the mixed transmission data frame to the satellite communication link and the ad hoc network link according to the dynamic fusion weight value; The power control module is used to adjust the transmit power of the radio frequency module according to a preset closed-loop power adaptive algorithm; The radio frequency module is further configured to distribute mixed transmission data frames according to the transmission ratio and the transmission power.

2. The integrated multi-network fusion satellite portable station according to claim 1, characterized in that: The performing multi-dimensional feature extraction on the original network parameter set to generate a network state feature vector comprises: Extracting the signal-to-noise ratio mean of the satellite link and the hop count distribution matrix of the ad hoc network nodes from the network original parameter set; Inputting the signal-to-noise ratio mean and the hop count distribution matrix into a preset spatiotemporal correlation neural network model to extract a transmission delay sensitivity evaluation index; A space-frequency joint mapping relationship table is constructed according to the transmission delay sensitivity evaluation index to generate a network state feature vector.

3. The integrated multi-network fusion satellite portable station according to claim 2, characterized in that: The extraction of the transmission delay sensitivity evaluation index includes: Establishing a spatiotemporal correlation neural network model, wherein the spatiotemporal correlation neural network model includes a first feature channel, a second feature channel, and a hidden layer; Inputting the signal-to-noise ratio mean into the first feature channel to extract frequency domain periodic features, thereby obtaining a satellite link frequency domain feature vector; Inputting the hop count distribution matrix into the second feature channel to perform node relationship feature mining to obtain a self-organizing network topology feature vector; The satellite link characteristics and the ad hoc network topology features are fused through the space-frequency joint convolution kernel preset in the hidden layer to obtain a transmission delay sensitivity evaluation index.

4. The integrated multi-network fusion satellite portable station according to claim 2, characterized in that: The calculating of the dynamic fusion weight value of the satellite link and the ad hoc network link according to the network state feature vector includes: Extracting a satellite link attenuation factor and an ad hoc network path redundancy index from the network state feature vector; Calculating the attenuation factor and redundancy index based on a pre-trained multi-objective optimization model, and outputting a multi-network bearer priority sequence; A dynamic fusion weight value is generated and updated according to the multi-network bearer priority sequence.

5. The integrated multi-network integrated satellite portable station according to claim 4, characterized in that: The training process of the multi-objective optimization model is: Construct a loss function that includes satellite link capacity, ad hoc network throughput, and end-to-end delay metrics; A non-dominated sorting genetic algorithm is used to perform Pareto front search on the loss function to generate the optimal weight distribution under multi-parameter constraints.

6. The integrated multi-network integrated satellite portable station according to claim 1, characterized in that: The data encapsulation unit includes: A production data processing subunit, configured to perform adaptive framing compression on the production data stream to generate a set of compressed data packets; An emergency video processing subunit, configured to input the emergency video stream into a preset quantum key distribution edge terminal to generate a dynamic encryption factor; The hybrid encapsulation subunit is used to securely encapsulate the compressed data packet set and the dynamic encryption factor to generate a hybrid transmission data frame.

7. The integrated multi-network integrated satellite portable station according to claim 6, characterized in that: The securely encapsulating the compressed data packet set and the dynamic encryption factor to generate a mixed transmission data frame includes: Performing XOR interleaving encoding on the compressed data packet set and the dynamic encryption factor output by the quantum key distribution edge terminal to generate an intermediate data stream; Performing secondary encryption using the intermediate data stream to generate a scrambling sequence; The scrambling code sequence is embedded in a preset Beidou positioning synchronization information field to form a mixed transmission data frame.

8. The integrated multi-network integrated satellite portable station according to claim 7, characterized in that: The performing secondary encryption using the intermediate data stream to generate a scrambling sequence includes: According to the resolution of the current emergency video stream, select the encryption mode from the preset national encryption algorithm library; Adjust encryption parameters according to packaging progress; The intermediate data stream is encrypted according to the selected encryption mode and encryption parameters to generate a scrambling code sequence.

9. The integrated multi-network integrated satellite portable station according to claim 1, characterized in that: The adjusting the transmit power of the radio frequency module according to the preset closed-loop power adaptive algorithm includes: Real-time monitoring of the equivalent isotropic radiated power of satellite links and the path loss gradient of ad hoc network links; generating a radio frequency dynamic gain adjustment signal according to the equivalent isotropic radiated power and the path loss gradient; The radio frequency dynamic gain adjustment signal is synchronized to the bias circuit of the radio frequency module to output a polarization matching power signal.

10. The integrated multi-network integrated satellite portable station according to claim 9, characterized in that: Generating a radio frequency dynamic gain adjustment signal according to the equivalent isotropically radiated power and the path loss gradient includes: Deploy a dynamic channel estimator at the receiving end of the satellite link to extract the scattering fading component of the propagation path in real time; Performing nonlinear fitting on the scattering fading component and a preset precipitation attenuation model to generate a power compensation coefficient for the millimeter wave band; A radio frequency dynamic gain adjustment signal is generated by using the power compensation coefficient, the equivalent isotropically radiated power, and the path loss gradient.

Images