Novel brain-computer interface secret communication method

Abstract

The invention discloses a novel brain-computer interface secret communication method, relates to the technical field of secret communication, and provides a DCSK brain-computer interface secret communication scheme based on a non-autonomous chaotic system. A non-autonomous chaotic system (CCNN) is adopted as a chaotic signal generator, the CCNN is a non-autonomous chaotic system and can simulate neurons of a primary visual cortex, and due to the fact that a vector field of the non-autonomous chaotic system is time-varying, the non-autonomous chaotic system shows complex dynamic behaviors and has high randomness and unpredictability. Therefore, the scheme is simple in structure and high in safety.

Description

Technical Field

[0001] The present invention relates to the field of secure communication technology, and in particular to a novel brain-computer interface secure communication method. Background Art

[0002] The electroencephalogram (EEG)-based brain-computer interface (BCI) is a technical system that enables direct communication between the brain and the external environment. It can read EEG signals in the brain and convert them into control commands, thereby facilitating interaction between people and computers or other electronic devices. BCI has a wide range of uses, such as using BCI systems to control wheelchairs, prostheses, cursors, etc. It can also be used in the field of neurorehabilitation to help patients with stroke or brain damage regain some control and motor function. As a relatively new field, BCI's communication system has more serious security issues. Unlike other user data, the user's brain activity data in this field can be used to infer cognitive, mental and physical health, etc. Therefore, the communication data of BCI needs to be encrypted to ensure data security.

[0003] In recent years, there are two main routes for secure encryption solutions. The first route is data encryption, which focuses on encrypting the data itself. Only by holding the corresponding key can the original data be restored. For example, Ein Shoka et al. (AAEin Shoka, MM Dessouky, A. El-Sayed, and E. El-Din Hemdan, “An efficientCNN based epileptic seizures detection framework using encrypted EEG signalsfor secure telemedicine applications,” Alex. Eng. J.) used the Arnold transform algorithm and chaotic Baker mapping technology to encrypt EEG data. Rajasekar and Pushpalatha (P. Rajasekar and M.Pushpalatha, “Huffman quantization approach for optimized EEG signalcompression with transformation technique,” ​​Soft Comput.) proposed using discrete cosine transform to decompose EEG signals, thereby improving data security. Liu et al. (J. Liu, J. Lian, JC Sprott, Q.Liu, and Y. Ma, “The Butterfly Effect in Primary Visual Cortex,” IEEE Trans. Comput.) used the Paillier encryption algorithm to encrypt EEG data and converted the data into integers by appropriate scaling to solve the problem of floating-point operations. Karakış et al. (R. Karakış, İ. Güler, İ. Çapraz, and E. Bilir, “A novel fuzzy logic-based image steganography method to ensure medical data security,” Comput. Biol. Med.) used the Huffman and LZW algorithms to losslessly compress EEG signals and then used the Rijndael symmetric algorithm with a 128-bit key to encrypt the compressed EEG signals.Akgul et al. (A.Akgul, OF Boyraz, K. Rajagopal, E. Guleryuz, MZ Yildiz, and M. Kutlu, "An unforced megastable chaotic oscillator and its application on protecting electrophysiological signals," Z. Für Naturforschung A.) proposed adding privacy information to EEG signals and encrypting EEG signals through a 3D chaotic system. However, data encryption methods often require complex data processing, which increases the requirements for device performance, which is fatal to the high real-time performance required by the BCI system. Therefore, secure channel encryption is a better choice.

[0004] The second route is channel encryption, which focuses on encrypting data on the data transmission channel. Compared with data encryption, secure channel encryption directly encrypts the transmission channel, so it has a wider range of applications. For example, Mezzina et al. (G. Mezzina, VF Annese, and D. De Venuto, “A Cybersecure P300-Based Brain-to-Computer Interface against Noise-Based and Fake P300 Cyberattacks,” Sensors.) proposed a system for transmitting data on randomly selected and dynamically changing channels, which can identify damage to the BCI system. Shahzadi et al. (R. Shahzadi, SM Anwar, F. Qamar, M. Ali, JJPC Rodrigues, and M. Alnowami, “Secure EEG Signal Transmission for Remote Health Monitoring Using Optical Chaos,” IEEE Access.) proposed the use of additive chaos masking to securely transmit EEG signals through optical fiber media. Zhang et al. (X. Zhang, X. Ding, D. Tong, P. Chang, and J. Liu, “Secure Communication Scheme for Brain-Computer Interface Systems Based on High-Dimensional Hyperbolic Sine Chaotic System,” Front. Phys.) proposed a new BCI communication scheme that uses a hyperbolic sine chaotic system as a chaotic signal generator in a differential chaotic shift keying (DCSK) framework to achieve multi-channel transmission of information. Lau et al. (FCM Lau, MM Yip, CK Tse, and SF Hau, “A multiple-access technique for differential chaos-shift keying,” IEEE Trans. Circuits Syst. Fundam. Theory Appl.) proposed a variable-delay multi-user DCSK system that achieves multi-user transmission by using different delay times for each time slot.Compared with traditional channel encryption methods, the DCSK framework is simpler and can be implemented without high-performance computing equipment. However, most of the existing DCSK schemes are based on autonomous chaotic systems, and the chaotic behavior of autonomous chaotic systems is generated by initial conditions and parameter settings, which means that if the initial conditions and parameters are leaked, the chaotic behavior may be predicted, resulting in the risk of data leakage. Summary of the invention

[0005] The purpose of the present invention is to provide a novel brain-computer interface secure communication method in order to solve the above-mentioned problems.

[0006] To achieve the above object, the technical solution adopted by the present invention is:

[0007] According to one aspect of the present invention, a novel brain-computer interface secure communication method is provided, comprising the following contents:

[0008] The invention is used to protect the user's EEG data in the brain-computer interface system to ensure the privacy and security of the data. The transmitting end includes using a non-autonomous chaotic system as a chaotic signal generator, introducing EEG data as an external driving signal of the non-autonomous chaotic system, so that the generated chaotic carrier signal is directly related to the EEG data. The chaotic signal generated by the chaotic signal generator is modulated into two groups, one group as a reference signal and the other group as an information-bearing signal, and then sequentially transmitted in two equal time slots; at the receiving end, the received reference signal and the information-bearing signal are correlated, and then the information carried by the signal is recovered according to the symbol output by the correlator. The specific steps are as follows:

[0009] Step 1: Collect raw EEG data;

[0010] Step 2: Apply the non-autonomous chaotic system; CCNN; to the solution. The CCNN formula is described as follows:

[0011]

[0012] The CCNN model consists of five main parts: coupled connections , Feed input , modulation products , Dynamic Activities and output , Indicates the external drive signal received, , , is the exponential decay factor, and is the weighting factor, and in addition, and denote the feeding and connecting synaptic weights, respectively, represents the connection strength, which directly determines the modulation product In ;

[0013] Step 3: In order to apply CCNN to the BCI secure communication scheme, the uncoupled connected CCNN model is used as the chaotic signal generator. The modified formula is described as follows:

[0014]

[0015] Assumptions , , , the initial conditions are set to , , . It is an external driving signal;

[0016] Step 4: Collect the EEG data As an external driving signal in the CCNN model , the modified formula is described as follows:

[0017]

[0018] Step 5: Take As the output sequence of the CCNN model;

[0019] Step 6: To eliminate the output sequence The physical characteristics of the , and the decorrelation operation is performed, the formula is described as follows:

[0020]

[0021] Step 7: Transmit the signal, and each bit duration is divided into two equal time slots. The first time slot transmits a length of Decorrelated chaotic sequence As a reference signal, the second time slot transmits a chaotic sequence with the same length as the first time slot as an information-bearing signal, so it takes If the bit to be transmitted is +1, the reference signal and the information-bearing signal transmitted in the two time slots are the same; if the bit to be transmitted is -1, the information-bearing signal is the negative value of the reference signal. The time transmission signal can be described as follows:

[0022]

[0023] in is a decorrelated chaotic sequence generated by the CCNN non-autonomous chaotic system. Represents the EEG data to be transmitted in the BCI system, is the sequence length of the reference signal or information-bearing signal, represents the iteration time;

[0024] Step 8: To recover EEG data , the received signal is and its delay The signal after Performing correlation operation, the output of the correlator is:

[0025]

[0026] Step 9: EEG Data This can be restored by determining the symbol of the variable:

[0027]

[0028] If the transmitted information bit is +1, then If the transmitted information bit is -1, then .

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] (1) This scheme introduces a non-autonomous chaotic system (CCNN) into DCSK, forming a novel BCI secure communication scheme. In particular, EEG data is used as the external driving signal of the non-autonomous chaotic system, so that the generated chaotic carrier signal is directly related to the EEG data and is highly random and unpredictable.

[0031] (2) In order to further enhance the pseudo-randomness of the chaotic signal generated by the non-autonomous chaotic system, a decorrelation operation method is adopted, thereby increasing the security of the BCI secure communication scheme.

[0032] (3) Security analysis shows that the chaotic carrier signal of this scheme has successfully passed the fifteen tests of the NIST suite. In terms of anti-interference ability, when the signal-to-noise ratio (SNR) is 100dB, the BER is approximately In addition, it also has good performance in terms of time efficiency. Therefore, this scheme has high real-time performance, is suitable for BCI systems, and can effectively protect user privacy. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a block diagram of the NA-DCSK scheme of the present invention;

[0034] Figure 2 It is a schematic diagram of CCNN of the present invention;

[0035] Figure 3 It is a block diagram of a chaotic signal generator of the present invention;

[0036] Figure 4 The waveform of the output sequence of the present invention and the corresponding phase space diagram;

[0037] Figure 5 Schematic diagram of probability density distribution of the output sequence of the present invention;

[0038] Figure 6 Schematic diagram of probability density distribution of output sequence after decorrelation operation of the present invention;

[0039] Figure 7 It is a schematic diagram of the bit error rate (BER) performance under the additive white Gaussian noise (AWGN) channel of the present invention. DETAILED DESCRIPTION

[0040] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0041] The NA-DCSK scheme transmits a reference signal and an information-bearing signal in different time slots. At the transmitter, a non-autonomous chaotic system (CCNN) is used as a chaotic signal generator, and then EEG data is used as an external driving signal for the non-autonomous chaotic system, resulting in more complex dynamic behaviors. Since the vector fields of non-autonomous chaotic systems are time-varying, they exhibit complex dynamic behaviors. In addition, a decorrelation operation is used to enhance the pseudo-randomness of the output sequence of the CCNN model. At the receiver, the received reference signal and the information-bearing signal are correlated, and the EEG data is recovered based on the symbols output by the correlator. This scheme uses an iterated function system and has good time efficiency.

[0042] The block diagram of the NA-DCSK scheme is as follows Figure 1 shown.

[0043] Step 1: Collect raw EEG data.

[0044] Step 2: Apply the non-autonomous chaotic system (CCNN) to the solution. The CCNN formula is described as follows:

[0045]

[0046] The CCNN model consists of five main parts: coupled connections , Feed input , modulation products , Dynamic Activities and output . Indicates the external drive signal received. , , is an exponential decay factor. and is the weighting factor. In addition, and denote the feeding and connecting synaptic weights, respectively. represents the connection strength, which directly determines the modulation product In The five main parts of CCNN are as follows: Figure 2 shown.

[0047] Step 3: In order to apply CCNN to the BCI secure communication scheme, the CCNN model with uncoupled connection is used as the chaotic signal generator. The block diagram of the chaotic signal generator is shown in the figure. Figure 3 The modified formula is described as follows:

[0048]

[0049] Assumptions , , , the initial conditions are set to , , . It is the external driving signal.

[0050] Step 4: Collect the EEG data As an external driving signal in the CCNN model , the modified formula is described as follows:

[0051]

[0052] Step 5: Take As the output sequence of the CCNN model, the waveform and phase space diagram are as follows Figure 4 As shown, Figure 4 Middle (A) (B) is the waveform of EY, and (B) is the phase space diagram of EY.

[0053] Step 6: Figure 5 is the output sequence The probability density distribution diagram shows that the output sequence has physical characteristics. The physical characteristics of the , and the decorrelation operation is performed, the formula is described as follows:

[0054]

[0055] After decorrelation operation, the probability density distribution of the above output sequence is as follows: Figure 6 shown.

[0056] Step 7: Transmit the signal, and divide each bit duration into two equal time slots. The first time slot transmits a length of Decorrelated chaotic sequence As a reference signal, the second time slot transmits a chaotic sequence with the same length as the first time slot as an information-bearing signal, so it takes If the bit to be transmitted is +1, the reference signal and the information-bearing signal transmitted in the two time slots are the same. If the bit to be transmitted is -1, the information-bearing signal is the negative value of the reference signal. The time transmission signal can be described as follows:

[0057]

[0058] in is a decorrelated chaotic sequence generated by the CCNN non-autonomous chaotic system (chaotic signal generator), Represents the EEG data to be transmitted in the BCI system, is the sequence length of the reference signal or information-bearing signal (the length of a time slot is ), Represents the iteration time.

[0059] Step 8: To recover EEG data , the received signal is and its delay The signal after Performing correlation operation, the output of the correlator is:

[0060]

[0061] Step 9: EEG Data This can be restored by determining the symbol of the variable:

[0062]

[0063] If the transmitted information bit is +1, then If the transmitted information bit is -1, then .

[0064] Experimental data:

[0065] The output sequence after decorrelation operation is tested using the Standard Statistical Test Suite for Random Number Generators (SP 800-22). Pseudo-randomness verification is performed. The test results are shown in Table 1.

[0066] Table 1 SP 800-22 test results

[0067]

[0068] The test results show that the output sequence after decorrelation operation is It can pass fifteen NIST tests and therefore generate highly random sequences.

[0069] The bit error rate (BER) performance is evaluated using an additive white Gaussian noise (AWGN) channel. Figure 7 The results show that when the signal-to-noise ratio (SNR) is 100dB, the BER is approximately . Therefore, it has good anti-interference ability.

[0070] right Bit data was tested to evaluate the time efficiency performance. The results are shown in Table 2.

[0071] Table 2 Time and speed test results of the scheme

[0072]

[0073] The results show that the scheme has good time efficiency performance.

[0074] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0075] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A novel brain-computer interface confidential communication method for protecting the user's EEG data in a brain-computer interface system to ensure the privacy and security of the data, characterized by: At the transmitting end, a non-autonomous chaotic system is used as a chaotic signal generator, and EEG data is introduced as an external driving signal of the non-autonomous chaotic system, so that the generated chaotic carrier signal is directly related to the EEG data. The chaotic signal generated by the chaotic signal generator is modulated into two groups, one group is used as a reference signal, and the other group is used as an information-bearing signal, and then sequentially transmitted in two equal time slots; at the receiving end, the received reference signal and the information-bearing signal are correlated, and then the information carried by the signal is restored according to the symbol output by the correlator. The specific steps are as follows: Step 1: Collect raw EEG data; Step 2: Apply the non-autonomous chaotic system; CCNN; to the solution. The CCNN formula is described as follows: The CCNN model consists of five main parts Composition: coupling connection , Feed input , modulation products , Dynamic Activities and output , Indicates the external drive signal received, , , is the exponential decay factor, and is the weighting factor, and in addition, and denote the feeding and connecting synaptic weights, respectively, represents the connection strength, which directly determines the modulation product In ; Step 3: In order to apply CCNN to the BCI secure communication scheme, the uncoupled connected CCNN model is used as the chaotic signal generator. The modified formula is described as follows: Assumptions , , , the initial conditions are set to , , ; It is an external driving signal; Step 4: Collect the EEG data As the external driving signal in the CCNN model, the modified formula is described as follows: Step 5: Take As the output sequence of the CCNN model; Step 6: To eliminate the output sequence The physical characteristics of the , and the decorrelation operation is performed, the formula is described as follows: Step 7: Transmit the signal, and each bit duration is divided into two equal time slots. The first time slot transmits a length of Decorrelated chaotic sequence As a reference signal, the second time slot transmits a chaotic sequence with the same length as the first time slot as an information-bearing signal, so it takes If the bit to be transmitted is +1, the reference signal and the information-bearing signal transmitted in the two time slots are the same; if the bit to be transmitted is -1, the information-bearing signal is the negative value of the reference signal. The time transmission signal can be described as follows: in is a decorrelated chaotic sequence generated by the CCNN non-autonomous chaotic system. Represents the EEG data to be transmitted in the BCI system, is the sequence length of the reference signal or information-bearing signal, represents the iteration time; Step 8: To recover EEG data , the received signal is and its delay The signal after Performing correlation operation, the output of the correlator is: Step 9: EEG Data This can be restored by determining the symbol of the variable: If the transmitted information bit is +1, then ; If the transmitted information bit is -1, then .

Images