Radio frequency transceiver based on AI secret communication
By improving the filter structure of the RF transceiver and introducing an AI secure communication protocol, the problems of low intermediate frequency signal sampling rate and communication security were solved, achieving stable data transmission and enhanced security.
Patent Information
- Application Number
- CN202520216416.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-11
- Publication Date
- 2025-12-05
- Estimated Expiration
- 2035-02-11
AI Technical Summary
Traditional superheterodyne RF transceivers have low intermediate frequency signal sampling rates, resulting in significant data loss at the receiving end. Furthermore, existing secure communication algorithms pose security threats during key distribution and updates, failing to effectively guarantee the security and reliability of communication.
An AI-based secure communication RF transceiver is adopted. By improving the RF module and frequency synthesizer module, using a 7th-order LC bandpass filter and PLL chip ADF4351, combined with the AI secure communication protocol, the sampling rate of the intermediate frequency signal is matched, and a secure transmission protocol is set in the baseband and intermediate frequency modules.
It achieves accurate sampling of intermediate frequency signals and stable data transmission, improves the security and reliability of communication, prevents data from being stolen or tampered with, and enhances the anti-interference capability of communication.
Smart Images
Figure CN223639262U_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The utility model relates to wireless communication technical field, concretely relates to a radio frequency transceiver based on AI secret communication. BACKGROUND
[0002] The superheterodyne radio frequency transceiver becomes the most widely used architecture in today's communication system due to its ultra-high reliability, taking the transmitter as an example, the transmitter framework it adopts is as shown in the accompanying Figure 1 The framework has the advantages of reducing direct current bias and local oscillator leakage, and the performance is improved by using a large number of filters, and the centralized control of gain can be realized.
[0003] Under the background of rapid development of the Internet of Things, communication security problems are increasingly prominent, security incidents are constantly exploding, and the Internet of Things technology has security risks, so secret communication technology has a deep significance for the development of the Internet of Things. The devices and sensors in the Internet of Things and artificial intelligence transmit data through wireless communication, which brings the risk of data being stolen or tampered with.
[0004] However, based on the above status, there are some problems in the traditional secret communication algorithm in the superheterodyne radio frequency transceiver: key management is more difficult, because the key needs to be securely distributed to each participant of the communication, and needs to be updated regularly to ensure security; the distribution of the key in wireless communication may be attacked, such as man-in-the-middle attack or eavesdropping; if the key is intercepted or tampered with during transmission, the security of the communication will be threatened. In this case, the existing signal transmission method is: randomly select a data matrix with the input baseband data symbol characteristics from the database and send it into the trained neural network model for comparison, and if the actual input baseband data is consistent with the selected data matrix, it is sent out through the radio frequency module. If the existing framework as shown in the accompanying Figure 1 is directly used, it is found that the sampling rate of the intermediate frequency signal is low, which leads to serious loss of data at the receiving end and cannot recover the original transmitted data.
[0005] Therefore, it is necessary to provide a radio frequency transceiver capable of ensuring correct sampling of intermediate frequency signals for the signal transmission method, so as to reduce privacy leakage while ensuring the reliability of signal transmission. INVENTION CONTENTS
[0006] 1. The technical problem to be solved is:
[0007] In view of the above technical problems, the utility model provides a radio frequency transceiver based on AI secret communication, which improves the radio frequency module and the frequency synthesis module of the existing superheterodyne radio frequency transceiver, so that the sampling rate of the intermediate frequency signal matches the sampling rate required by the secret communication algorithm.
[0008] 2. Technical scheme:
[0009] The radio frequency transceiver based on AI secret communication includes a baseband & intermediate frequency module, a frequency synthesizer module and a radio frequency module which are connected with each other; characterized in that: the band pass filters in the input and output signal channels of the radio frequency module are all 7-order LC band pass filters; the band pass filters in the oscillation circuit of the frequency synthesizer module are all 7-order LC band pass filters, and the PLL chip adopts a model ADF4351 phase-locked loop chip; the baseband & intermediate frequency module pre-sets an AI secret communication protocol to realize safe and secret transmission of input and output signals.
[0010] Further, the input signal channel of the radio frequency module specifically includes an input signal antenna, a first 7-order LC band pass filter, a low noise amplifier circuit LAN and a second 7-order LC band pass filter which are connected in sequence, and the output end of the second 7-order LC band pass filter is connected to the input mixer of the baseband & intermediate frequency module; the output signal channel of the radio frequency module includes a third 7-order LC band pass filter, a power amplifier, a fourth 7-order LC band pass filter and an output antenna which are connected in sequence, wherein the third 7-order LC band pass filter BPF3 receives the signal output by the output mixer of the baseband & intermediate frequency module.
[0011] Further, the input signal channel of the baseband & intermediate frequency module includes an input mixer, a signal processing circuit and an ADC conversion circuit which are connected in sequence; the output signal of the ADC conversion circuit is transmitted to the FPGA chip; the FPGA chip sends the signal which has passed through the preset AI secret communication protocol to the third 7-order LC band pass filter of the radio frequency module through the DAC conversion circuit, the analog modulator and the output mixer of its output signal channel in sequence; the baseband & intermediate frequency module further includes an ARM module; the ARM module receives the control instruction from the PC end and transmits it to the FPGA chip.
[0012] Further, the frequency synthesizer module includes a temperature compensation type crystal oscillation circuit, an ARM chip, a PLL chip and fifth and sixth 7-order LC band pass filters; the output ends of the temperature compensation type crystal oscillation circuit, the ARM chip and the PLL chip are connected to the PLL chip; the PLL chip outputs the signal to the fifth and sixth 7-order LC band pass filters; the fifth and sixth 7-order LC band pass filters respectively output corresponding local oscillation signals to the corresponding input or output mixers of the baseband & intermediate frequency module.
[0013] Further, in the frequency synthesizer module, the temperature compensation type crystal oscillation circuit outputs a 60MHz sine signal; the communication between the ARM chip and the PLL chip is SPI communication; the fifth 7-order LC band pass filter is a passive LC band pass filter with a center frequency of 422.3MHz; and the sixth 7-order LC band pass filter is a passive LC band pass filter with a center frequency of 503MHz.
[0014] Further, the first to fourth 7th order LC bandpass filters in the radio frequency module are all passive LC bandpass filters with a center frequency of 433MHz; and the input and output antennas are both 433MHz radio frequency antennas.
[0015] Further, the preset AI secret communication protocol in the baseband & intermediate frequency module is that the FPGA selects a data matrix with input baseband data symbol characteristics from a stored database at random and sends it into a trained neural network model, and if the baseband data symbol selected by the FPGA is consistent with the actual input data symbol, the radio frequency module is used for sending.
[0016] 3. Beneficial effects:
[0017] (1) The radio frequency transceiver based on AI secret communication provided by the utility model is improved on the superheterodyne radio frequency transceiver under the existing AI secret communication protocol, and first to fourth 7th order passive LC bandpass filters are arranged in the radio frequency module, the center frequency of the four 7th order passive LC bandpass filters is 433MHz, the passband bandwidth is 70MHz, the input and output impedances are both 50Ω, and the insertion loss is greater than -1.5dB. The first BPF can be used for filtering out signals lower than 398MHz or higher than 468MHz from the signals received by the 433MHz input antenna, so that the signals sent into the low noise amplifier LNA are within 398MHz~468MHz, and the signal with the highest power is 433MHz; the second BPF is used for filtering out the spurious signals lower than 398MHz or higher than 468MHz generated after the LNA amplification, so that the signals sent into the input mixer are more stable and do not fluctuate greatly in power, which is convenient for processing the digital signals in the later stage. The third BPF is used for filtering out the spurious signals lower than 398MHz or higher than 468MHz in the output signal of the output mixer Mixer2, mainly the upper sideband signal of 573MHz, and the lower sideband signal of 433MHz is reserved. The fourth BPF is used for filtering out the spurious signals of the 433MHz signal after the amplification of the power amplifier PA, so that the output signal is stable at 433MHz, a signal carrier with strong power is generated for the data to be transmitted, so that the transmitted data is more stable and the transmission distance is farther.
[0018] (2) The radio frequency transceiver based on AI secret communication provided by the utility model, two seventh-order passive LC band pass filters are also used in the frequency synthesis module, namely the fifth and sixth seventh-order passive LC band pass filters; the center frequency of the fifth BPF is 422.3MHz, the passband bandwidth is 40MHz, the input and output impedances are both 50Ω, the insertion loss is greater than -1.5dB, the signals higher than 442.3MHz or lower than 402.3MHz in the PLL output signal can be filtered out, so that the frequency of the output local oscillation signal LO1 signal is stabilized at 422.3MHz. Similarly, the sixth BPF with the center frequency of 503MHz makes the frequency of the PLL output local oscillation signal LO2 signal stabilized at 503MHz. Through multiple verification examples, it can be verified that when the sampling rate of the intermediate frequency signal is 1.6MHz, the transmitted data can be correctly recovered, and the external environment cannot intercept the transmitted data through the traditional eavesdropping technology. BRIEF DESCRIPTION OF DRAWINGS
[0019] Figure 1 It is a principle block diagram of the superheterodyne radio frequency transmitter of the prior art;
[0020] Figure 2 It is a whole block diagram of the radio frequency transceiver based on AI secret communication of the utility model;
[0021] Figure 3 It is a circuit diagram of the seventh-order passive LC band pass filter in the utility model;
[0022] Figure 4 It is a flow chart of the preset AI secret communication algorithm in the embodiment;
[0023] Figure 5 It is a frequency spectrum diagram of the tested local oscillator LO1 in the embodiment;
[0024] Figure 6 It is a frequency spectrum diagram of the tested local oscillator LO2 in the embodiment;
[0025] Figure 7 It is an ADS simulation result diagram of the BPF1~4 filters in the embodiment.
[0026] The letter explanation of the drawing: BPF1~BPF6 are the first~sixth seventh-order passive LC band pass filters; SMA is a high-frequency connector; UART is a serial communication; SMA is a high-frequency connector; LAN is a low-noise amplifier; LO1~2 are local oscillation signals output by the PLL in the frequency synthesis module, and also serve as input local oscillation signals of the input and output mixers in the radio frequency module; PA is a power amplifier; SPI is an SPI communication protocol; TCXO is a temperature compensated crystal oscillator circuit. DETAILED DESCRIPTION
[0027] The utility model is specifically explained below in combination with the drawings.
[0028] As shown in the accompanying Figure 2 , 3 , a radio frequency transceiver based on AI secret communication, comprising a baseband & intermediate frequency module, a frequency synthesis module and a radio frequency module connected with each other; characterized in that: the bandpass filters in the input and output signal channels of the radio frequency module are all 7-order LC bandpass filters; the bandpass filters in the oscillation circuit of the frequency synthesis module are all 7-order LC bandpass filters, and the PLL chip adopts a phase-locked loop chip with a model number of ADF4351; the baseband & intermediate frequency module is pre-set with an AI secret communication protocol to realize safe and secret transmission of input and output signals.
[0029] Further, the input signal channel of the radio frequency module specifically comprises an input signal antenna, a first 7-order LC bandpass filter, a low-noise amplification circuit LAN and a second 7-order LC bandpass filter connected in sequence, and the output end of the second 7-order LC bandpass filter is connected to an input mixer of the baseband & intermediate frequency module; the output signal channel of the radio frequency module comprises a third 7-order LC bandpass filter, a power amplifier, a fourth 7-order LC bandpass filter and an output antenna connected in sequence, wherein the third 7-order LC bandpass filter receives signals output by an output mixer of the baseband & intermediate frequency module.
[0030] Further, the input signal channel of the baseband & intermediate frequency module comprises an input mixer, a signal processing circuit and an ADC conversion circuit connected in sequence; the output signal of the ADC conversion circuit is transmitted to an FPGA chip; the FPGA chip sends signals that have passed through the preset AI secret communication protocol to the third 7-order LC bandpass filter of the radio frequency module through a DAC conversion circuit, an analog modulator and an output mixer of its output signal channel in sequence; the baseband & intermediate frequency module further comprises an ARM module; the ARM module receives control instructions from a PC end and transmits them to the FPGA chip.
[0031] Further, the frequency synthesis module comprises a temperature-compensated crystal oscillation circuit, an ARM chip, a PLL chip and fifth and sixth 7-order LC bandpass filters; the temperature-compensated crystal oscillation circuit, the ARM chip and the PLL chip are connected to the PLL chip at the output end; the PLL chip outputs signals to the fifth and sixth 7-order LC bandpass filters; the fifth and sixth 7-order LC bandpass filters respectively output corresponding local oscillation signals to the corresponding input or output mixers of the baseband & intermediate frequency module.
[0032] Further, in the frequency synthesizer module, the temperature compensation type crystal oscillator circuit outputs a 60MHz sine signal; the communication between the ARM chip and the PLL chip is SPI communication; the fifth 7th order LC bandpass filter is a passive LC bandpass filter with a center frequency of 422.3MHz; and the sixth 7th order LC bandpass filter is a passive LC bandpass filter with a center frequency of 503MHz.
[0033] Further, in the radio frequency module, the first to fourth 7th order LC bandpass filters are all passive LC bandpass filters with a center frequency of 433MHz; and the input and output antennas are both 433MHz radio frequency antennas.
[0034] Further, in the baseband & intermediate frequency module, the preset AI secret communication protocol is that the FPGA selects a data matrix with input baseband data symbol characteristics from a stored database at random and sends it into a trained neural network model, and if the baseband data symbol selected by the FPGA is consistent with the actual input data symbol, the signal is sent out through the radio frequency module.
[0035] As shown in the accompanying Figure 2 , the principle of the radio frequency transceiver for realizing secret communication / the working mode of the radio frequency transceiver is as follows:
[0036] The FPGA chip performs QPSK modulation and AI encryption on the original baseband signal to obtain a 70MHz intermediate frequency signal, mixes the obtained intermediate frequency signal with LO2 (503MHz) as shown in the accompanying Figure 6 , filters it through a 433MHz 7th order passive LC bandpass filter to obtain a 433MHz radio frequency signal, amplifies the signal through a power amplifier, processes it through a 7th order passive LC bandpass filter, and transmits it through Antenna2; Antenna1 receives the 433MHz radio frequency signal, filters it through a 433MHz 7th order passive LC bandpass filter, processes it through a low noise amplifier, filters it again through a passive LC bandpass filter, mixes it with LO1 (422.3MHz) as shown in the accompanying Figure 5 , and obtains a 10.7MHz intermediate frequency signal, which is processed through a filter, an ADC, etc., and then enters the FPGA chip; after the FPGA demodulates, samples, and decrypts the signal, the original input data can be seen on the PC end through the ARM. Specific embodiments:
[0038] In order to verify the effect of the radio frequency transceiver, the local oscillator signals LO1 and LO2 as shown in the accompanying Figure 5 , 6 are used for testing.
[0039] In this embodiment, the temperature compensation type crystal oscillator circuit TCXO of the frequency synthesizer module is a 60MHz sine signal, which provides a stable oscillation source for the PLL chip, and SPI communication is used between the ARM and the PLL; the model of the PLL chip is ADF4351, and the spectrum diagram of LO1 and LO2 is as shown in Figure 5 、 Figure 6 The power of the LO1 sine signal is -6.06dBm, and the power of the LO2 sine signal is -3.21dBm. Figure 5 、 6 In the figure, the horizontal axis represents the frequency value of the signal, and the unit is MHz, and the vertical axis represents the power value of the signal, and the unit is dBm. Figure 5 The position of "1" in the figure is the frequency spectrum of the measured signal, and the value is shown as M1 in the upper right corner of the figure, the frequency is about 422.3MHz, and the power is -6.06dBm. Figure 6 The position of "1" in the figure is the frequency spectrum of the measured signal, and the value is shown as M1 in the upper right corner of the figure, the frequency is about 422.3MHz, and the power is -6.06dBm.
[0040] In this embodiment, the center frequency of BPF1-4 is 433MHz, the passband bandwidth is 70MHz, the input and output impedance is 50Ω, and the insertion loss is greater than -1.5dB.
[0041] At the same time, the AI-based secure communication algorithm used in this embodiment is as shown in the attached Figure 4 , which specifically includes the following steps:
[0042] S1. Process the input baseband data to transmit it in the form of ASCII code in the channel;
[0043] S2. Select a data matrix with the data symbol characteristics corresponding to the input data from the database;
[0044] S3. Send the selected data matrix into the trained neural network model to verify whether the selected data is consistent with the input symbol of step S1;
[0045] S4. If the selected data model of step S2 is consistent with the input data of step S1, then the data is converted by DA and sent out through the radio frequency module; if not, return to step S2;
[0046] S5. The data sent by step S4 is received by the radio frequency module and converted by AD to obtain the sent data;
[0047] S6. Put the data obtained in step S5 into the trained model for data processing to obtain the symbol;
[0048] S7. The symbols obtained in step S6 are restored to obtain the data body in step S1.
[0049] In step S2, the selected data matrix type is consistent with the ASCII encoding type, and there are 256 ASCII encoding character types, and there are 256 data matrix types. In order to meet the requirements of neural network training and data selection in step S3, the initial matrix array of the corresponding character type is obtained, and the number of matrices in each two-dimensional matrix array is equal.
[0050] And step S2 further comprises:
[0051] S21. The data matrix is a two-dimensional matrix array.
[0052] S22. Randomly select a data matrix corresponding to the ASCII code of the input data in step S1 from the database using a random function;
[0053] S23. Add random noise interference to the data matrix, and the random noise interference is realized by randomly generating a function in the data matrix array, and the random number is distributed at each position of the data matrix.
[0054] As shown in the accompanying Figure 7 As shown in the accompanying As shown in the accompanying
[0055] As shown in the accompanying Figure 2 As shown in the accompanying
[0056] At the same time, the AI secret communication sending and receiving data of the radio frequency transceiver are realized by using the UART serial communication of ARM, the visualization of the data transmission is realized, and the data transmission rate is 115200 baud.
[0057] While the utility model has disclosed as above with preferable embodiments, they are not used to limit the utility model, any skilled person, without departing from the spirit and scope of the utility model, should be able to make various changes or refinements, therefore the protection scope of the utility model should be limited by the protection scope of the claims of the present application.
Claims
1. An AI-based secure communication radio frequency transceiver, comprising a baseband & intermediate frequency module, a frequency synthesis module, and a radio frequency module connected to each other; characterized in that: The band pass filter in the input and output signal channel of the radio frequency module is a 7-order LC band pass filter; the band pass filter in the oscillation circuit of the frequency synthesizer module is a 7-order LC band pass filter, and the PLL chip adopts a model ADF4351 phase-locked loop chip; the baseband & intermediate frequency module pre-set AI secret communication protocol realizes the safe and secret transmission of the input and output signals.
2. The radio frequency transceiver based on AI secret communication according to claim 1, wherein: The input signal channel of the radio frequency module specifically comprises an input signal antenna, a first 7-order LC band pass filter, a low noise amplifier circuit LAN, a second 7-order LC band pass filter connected in sequence, and the output end of the second 7-order LC band pass filter is connected to the input mixer of the baseband & intermediate frequency module; the output signal channel of the radio frequency module comprises a third 7-order LC band pass filter, a power amplifier, a fourth 7-order LC band pass filter and an output antenna connected in sequence, wherein the third 7-order LC band pass filter receives the signal output by the output mixer of the baseband & intermediate frequency module.
3. The radio frequency transceiver based on AI secret communication according to claim 1, wherein: The input signal channel of the baseband & intermediate frequency module comprises an input mixer, a signal processing circuit and an ADC conversion circuit connected in sequence; the output signal of the ADC conversion circuit is transmitted to the FPGA chip; the FPGA chip sends the signal passing through the DAC conversion circuit, the analog modulator and the output mixer of the output signal channel in sequence to the third 7-order LC band pass filter of the radio frequency module through the pre-set AI secret communication protocol; the baseband & intermediate frequency module further comprises an ARM module; the ARM module receives the control instruction from the PC end and transmits it to the FPGA chip.
4. The radio frequency transceiver based on AI secret communication according to claim 1, wherein: The frequency synthesizer module comprises a temperature compensation type crystal oscillation circuit, an ARM chip, a PLL chip and fifth and sixth 7-order LC band pass filters; the output end of the temperature compensation type crystal oscillation circuit, the ARM chip and the PLL chip is connected to the PLL chip; the PLL chip outputs the signal to the fifth and sixth 7-order LC band pass filters; the fifth and sixth 7-order LC band pass filters respectively output corresponding local oscillation signals to the corresponding input or output mixers of the baseband & intermediate frequency module.
5. The radio frequency transceiver based on AI secret communication according to claim 4, characterized in that: In the frequency synthesizer module, the output signal of the temperature compensation type crystal oscillation circuit is a 60MHz sine signal; the communication between the ARM chip and the PLL chip is SPI communication; the fifth 7-order LC band pass filter is a passive LC band pass filter with a center frequency of 422.3MHz; the sixth 7-order LC band pass filter is a passive LC band pass filter with a center frequency of 503MHz.
6. The radio frequency transceiver based on AI secret communication according to claim 2, wherein: The first to fourth 7-order LC band pass filters in the radio frequency module are all passive LC band pass filters with a center frequency of 433MHz; the input and output antennas are both 433MHz radio frequency antennas.
7. The radio frequency transceiver based on AI secret communication according to claim 3, characterized in that: The pre-set AI secret communication protocol in the baseband & intermediate frequency module is that the FPGA selects a data matrix with input baseband data symbol characteristics from the stored database at random and sends it into the trained neural network model, and if the baseband data symbol selected by the FPGA is consistent with the actual input data symbol, it is sent out through the radio frequency module.