A neural signal conditioning circuit design (1)

Human neural signals direct characterization of the human self, of neural signals in order to understand and identify the body provides a way.

Over the years. At present, the study includes neural electrodes and neural signal conditioning circuits in two parts. Neural electrodes can be neural signals from the human body, extracted and neural signal conditioning circuits, neural signal denoising, amplification, recognition, etc.

Neural signals and other human biological signals have the same features, but also has its unique features.

According to research in neurobiology, nerve signals a takes the pulse of signals, frequencies around General be 1 kHz, up to a high of 10kHz. For example, a bundle of muscle control, when the motor nerve impulse voltage signal comes, muscle fiber contraction occurs, the intensity of contraction, according to the different nerve impulse frequency and intensity. Therefore, as long as the pulse voltage identification, handling into digital control signals to control prosthetics such as specific application.

Of course, neural signal detection has its difficulties.

Human nerve signals is a powerful low frequency noise interference, due to their weak signal is very weak, only the microvolt, at anomaly interference and therefore effective signal strong, tend to be submerged. Interference signal generally include high-frequency electromagnetic interference, power frequency interference and 50Hz polarization voltage, etc. Power-frequency interference to exist in the form of common-mode signal, usually amplitude up to a few volts to dozens of volts. And polarization voltage is measured between the electrode and biological tissues constitute half battery chemistry to produce DC voltage, generally scores of millivolt, up to 300mV. In addition, because of the complexity and specificity of the organism, its equivalent signal source output impedance generally is very large, you can have tens of thousands of Europe, it is also necessary to consider. According to the characteristics of neural signals above, a targeted, performance superior, reliable neural signal conditioning circuits.

1 circuit system structure and principle

According to the characteristics of nerve signals, and the General characteristics of electrodes, conditioning circuit must have some must-have for performance.

First, the circuit must have a high common-mode rejection ratio, the ratio suppression frequency interference and other measurement parameters of the physiological effects of interference. If the circuit cmrr is 120dB, enter the signal effect of the common mode signal will weaken a million times, the common mode signal, 1V effect 1 μ v differential-mode signals. At the same circuit input impedance is also a very important parameter. High input impedance can effectively reduce the signal source high internal resistance. Mentioned organisms of equivalent signal source output impedance generally tens thousand euro, this requires the design of circuit input impedance is larger than 100 megabytes of Europe. In addition, because each electrode contact with human tissue is different, thus demonstrating the high internal resistance of the unstable nature of the source will cause electrode impedance unbalanced, so the differential mode EMI interference. Improve amplifier input impedance to reduce the impact of this transformation. At the same time, relative to the amplitude of the microvolt, neural signal conditioning circuits of low noise, low drift, and other indicators also is extremely important. This article by neural signal conditioning circuit structure as shown in Figure 1. Circuit system is divided into three parts: the front panel input amplifier, intermediate-level signal processing circuits and subsequent transmission of the signal recognition circuit.

2 circuit design

Early biological signal circuit design and use of discrete components, along with the continuous development of micro-electronics technology, a number of high-performance integrated instrumentation amplifier, because this class of device performance and to avoid the installation and commissioning work in biomedical instrumentation design was generally welcomed.

This article was previously published under design fully apply this type of diodes.

2.1 predecessor input amplifier

Predecessor-the main considerations of noise, input impedance and three common-mode rejection ratio.

Here the design of the circuit is composed of three parts: an input buffer, high-frequency filtering and instrument amplifier. Circuit diagram shown in Figure 2.

Input buffer with direct voltage-feedback design, theoretical input impedance is infinite, effectively isolating human and circuit system, remove the signal source resistance, high and unstable.

By R1a, R1b, C1b, C2 consisting of low-pass filter network can effectively remove the impact of high frequency electromagnetic noise.

Circuit of differential-mode signal cut-off frequency common mode signal as BWDDFF and frequency BWCM if type (1), (2) below, where R1a, R1b, C1a, C1b must precisely equal, C2 > 10C1. In General, instrument amplifier of signal above 20kHz has no common mode rejection, use of the networkYou can make the instrument amplifier work more efficiently.

Instrument amplifier because its classic three operational structure has a high input impedance and common-mode rejection ratio, and simply add a resistance to set the gain in the area of biological signal processing is widely used.

This selection of AD company AD8221 is the latest models, than common AD620 in areas of high performance to an order of magnitude. Because of the existence of polarization voltage, in order to avoid the circuit of saturated, preamplifier gain must be dozens of times, not too large.

2.2 intermediate processing circuit

Intermediate-level processing circuit is divided into band pass frequency selection network II, amplifier, 50Hz notch filter and gain regulation circuit, etc.

Band pass frequency selection network composed by RC passive network, simple and reliable, the band's maximum range set to 0.05Hz ~ 10kHz.

Depending on individual differences, the network can be used by digital control circuit for a combination of different frequency bands to choose to comply with the best signal status.

Secondary amplification circuit in structure and gain regulation circuit is similar to that received by a voltage-feedback operational amplifier.

Former signal amplification, which control the overall circuit gain up to 120dB. Its structure diagram shown in Figure 3. Here, the operational amplifier selection OP27 and applying voltage series negative feedback structures. Its advantage is simple in structure, has the following irreplaceable superior performance: (1) input equivalent impedance, Ri = (1 + AF) rid, output equivalent impedance small, Ro = ro/(1 + AF), rid as amplifier input impedance, ro for output impedance. Not only do the signal amplification effect, but also played the role of buffer, effectively isolating the before and after the class module, without additional impedance converter and matching module; (2) the use of the capacitor C so that the entire module has a low pass function, not only can you get rid of signals in high frequency interference, also because of its advanced compensatory role effective signal of high frequency parts of a phase compensation. Through reasonable design, circuit frequency section of phase will change flat. Upper mentioned neural signal is a kind of pulse shape signal, signal shape without noticeable distortion, in the time domain processing has a positive sense.