Design of EEG signal acquisition circuit based on universal serial bus

0 Introduction
The brain is the center of human thinking activities. It is the command center for receiving external signals, generating sensations, forming consciousness, conducting logical thinking, issuing commands, and generating behaviors. By studying electroencephalogram (EEG) signals, we can understand the mechanism of brain activity and human cognitive process, and it is also an important means of diagnosing brain diseases.
Most of the existing EEG signal acquisition systems use special machines, which are inconvenient to use, move, and repair. In addition, the anti-interference ability is poor, and monitoring must be carried out in a specific environment (shielded room). Moreover, the amount of detection data recorded is small, and long-term, large-capacity recording and analysis cannot be achieved. In addition, the cost is expensive, and the collected EEG signals are not accurate enough. It is often necessary to combine the doctor's experience to make a diagnosis, so it has a certain degree of subjectivity. This paper designs a new EEG signal acquisition circuit based on USB 2.0.

1 Principle of EEG signal acquisition
Analyzing EEG signals and mastering their characteristics are crucial to designing accurate and effective EEG signal acquisition circuits. First, we can select the signal that carries the most information; second, we can study the corresponding signal processing algorithm for specific signals.
1.1 Characteristics and basic components of EEG waves
There are three main characteristics of EEG signals: the frequency is mainly concentrated in the low frequency band of 100 Hz; the signal is weak, generally 50μV or less; the source impedance of the signal is high and easily interfered by external signals. The amplitude range of EEG signals of adults is generally between 10 and 50μV, and the frequency range is between 0.5 and 30 Hz.
1.2 Principle and method of EEG signal acquisition
At present, for the measurement of EEG signals, a very high resolution can be obtained in the dimension of time. However, in the dimension of space, the resolution obtained is very low, which depends on the number of electrodes placed on the scalp. This system uses 16 electrodes to extract EEG signals with a sampling frequency of 1 000 Hz. Due to the inherent characteristics of EEG signals and environmental factors, the background noise of EEG signals is relatively complex, including 50 Hz power frequency interference, ECG artifacts, myoelectric interference, baseline drift, contact noise between electrodes and skin, and electromagnetic interference from other surrounding instruments. Therefore, the acquisition system is required to have high input impedance, high common mode rejection ratio, low noise amplification, and high-quality filtering measures that can extract weak signals from strong noise.
EEG signals are usually collected by placing some electrodes on the surface of the scalp. Common types of electrodes include silver tube electrodes, needle electrodes, and adhesion electrodes. This system uses silver tube electrodes to achieve the connection between the scalp and the EEG measurement equipment. In order to enhance connectivity and conductivity, some physiological saline is applied between the electrode and the skin. The placement of the electrodes adopts the international 10-20 system electrode placement method.

2 Design of EEG signal acquisition circuit
The EEG signal acquisition circuit includes four parts: EEG signal amplification, filtering, A/D conversion and USB interface circuit. The overall structure is shown in Figure 1.

2.1 Design of preamplifier circuit for EEG signal
The signal amplification detection circuit is a very important part of this system. It makes hardware preparation for subsequent data acquisition and processing analysis. This system uses high-precision instrument amplifier AD8221 as the preamplifier circuit, which has the characteristics of high input impedance, high common mode rejection ratio, low noise and strong anti-interference ability. When the electrode contacts the skin, a polarization potential of tens of millivolts can be generated, so the amplification factor of the preamplifier cannot be too large to avoid saturation of the circuit. [page]

0 Introduction
The brain is the center of human thinking activities. It is the command center for receiving external signals, generating sensations, forming consciousness, conducting logical thinking, issuing commands, and generating behaviors. By studying electroencephalogram (EEG) signals, we can understand the mechanism of brain activity and human cognitive process, and it is also an important means of diagnosing brain diseases.
Most of the existing EEG signal acquisition systems use special machines, which are inconvenient to use, move, and repair. In addition, the anti-interference ability is poor, and monitoring must be carried out in a specific environment (shielded room). Moreover, the amount of detection data recorded is small, and long-term, large-capacity recording and analysis cannot be achieved. In addition, the cost is expensive, and the collected EEG signals are not accurate enough. It is often necessary to combine the doctor's experience to make a diagnosis, so it has a certain degree of subjectivity. This paper designs a new EEG signal acquisition circuit based on USB 2.0.

1 Principle of EEG signal acquisition
Analyzing EEG signals and mastering their characteristics are crucial to designing accurate and effective EEG signal acquisition circuits. First, we can select the signal that carries the most information; second, we can study the corresponding signal processing algorithm for specific signals.
1.1 Characteristics and basic components of EEG waves
There are three main characteristics of EEG signals: the frequency is mainly concentrated in the low frequency band of 100 Hz; the signal is weak, generally 50μV or less; the source impedance of the signal is high and easily interfered by external signals. The amplitude range of EEG signals of adults is generally between 10 and 50μV, and the frequency range is between 0.5 and 30 Hz.
1.2 Principle and method of EEG signal acquisition
At present, for the measurement of EEG signals, a very high resolution can be obtained in the dimension of time. However, in the dimension of space, the resolution obtained is very low, which depends on the number of electrodes placed on the scalp. This system uses 16 electrodes to extract EEG signals with a sampling frequency of 1 000 Hz. Due to the inherent characteristics of EEG signals and environmental factors, the background noise of EEG signals is relatively complex, including 50 Hz power frequency interference, ECG artifacts, myoelectric interference, baseline drift, contact noise between electrodes and skin, and electromagnetic interference from other surrounding instruments. Therefore, the acquisition system is required to have high input impedance, high common mode rejection ratio, low noise amplification, and high-quality filtering measures that can extract weak signals from strong noise.
EEG signals are usually collected by placing some electrodes on the surface of the scalp. Common types of electrodes include silver tube electrodes, needle electrodes, and adhesion electrodes. This system uses silver tube electrodes to achieve the connection between the scalp and the EEG measurement equipment. In order to enhance connectivity and conductivity, some physiological saline is applied between the electrode and the skin. The placement of the electrodes adopts the international 10-20 system electrode placement method.

2 Design of EEG signal acquisition circuit
The EEG signal acquisition circuit includes four parts: EEG signal amplification, filtering, A/D conversion and USB interface circuit. The overall structure is shown in Figure 1.

2.1 Design of preamplifier circuit for EEG signal
The signal amplification detection circuit is a very important part of this system. It makes hardware preparation for subsequent data acquisition and processing analysis. This system uses high-precision instrument amplifier AD8221 as the preamplifier circuit, which has the characteristics of high input impedance, high common mode rejection ratio, low noise and strong anti-interference ability. When the electrode contacts the skin, a polarization potential of tens of millivolts can be generated, so the amplification factor of the preamplifier cannot be too large to avoid saturation of the circuit.

[page]

The sampling accuracy of the A/D conversion circuit (see Figure 6) is 16 bits, and the sampling frequency is 1 000 Hz. The USB controller controls the AD7675 to convert each EEG signal through the CNVST# pin (pin 35 of AD7675), and also obtains the current status of the AD7675 by querying the BUSY pin (pin 29 of AD7675). The USB controller also responds to the host's request and transmits the EEG data to the host in a timely and accurate manner. In order to improve the accuracy of the conversion, a sample-and-hold circuit is designed to latch the 16 EEG signals at the same time and then convert them one by one. Since 32 KB (16×16×1 000) of EEG data is generated per second, in order to ensure that there is enough space to temporarily store these EEG data, an SDRAM with a capacity of 256K×16 b is expanded outside the CY7C68013 chip.

4 Conclusion
Electroencephalogram (EEG) is an important bioelectric signal of the human body. At present, the role of EEG signal acquisition and processing in the treatment of patients with certain diseases has been increasingly valued by medical institutions. This paper discusses in detail the design process of USB 2.0 interface EEG signal acquisition circuit. The system realizes the acquisition, high-speed transmission and real-time processing of EEG data, effectively solving the defects of traditional EEG data acquisition system such as slow speed, simple processing function, small data storage capacity and complex connection, and meets the actual needs.