CN105726023B - Electroencephalogram signal quality real-time judging system - Google Patents

Abstract

The invention discloses an electroencephalogram signal quality real-time judgment system, which relates to the field of electroencephalogram signal digital signal processing. The analysis data generation module is used for: acquiring an electroencephalogram signal segment corresponding to the time window by using the cached electroencephalogram signal data and the generated time window; the standard deviation calculation module is used for: calculating the average value of the electroencephalogram signal segments, and calculating the standard deviation of the electroencephalogram signal segments by using the average value; the signal quality determination module is used for: and obtaining the volatility judgment result of the current electroencephalogram signal segment through the calculated standard deviation, and jointly judging the quality of the current electroencephalogram signal by utilizing the volatility judgment result of the current electroencephalogram signal segment and the volatility judgment result of the previous electroencephalogram signal segment. The method can quickly and effectively make real-time judgment and analysis on the quality of the electroencephalogram signal, and effectively guarantee the accuracy of subsequent electroencephalogram application and analysis.

Description

A real-time judgment system of EEG signal quality

technical field

The invention relates to the field of digital signal processing of electroencephalogram signals, in particular to a real-time judgment system for the quality of electroencephalogram signals.

Background technique

EEG (electroencephalogram) is an electrical signal sent by the human brain. This signal is extremely weak, generally only a few microvolts to hundreds of microvolts. Therefore, EEG signals need to be captured by a dedicated EEG acquisition device. The EEG sensor of the EEG acquisition device generally consists of an electrode and a reference electrode or a plurality of electrodes and a reference electrode. Due to the different electrode standard potentials of the electrode materials, there will be a voltage difference between the electrode and the reference electrode after touching the human skin. Because these voltage differences tend to be in an unstable state within a few seconds, and these voltage differences are often much larger or smaller than the fluctuations of the EEG signal, so the electrical signal (we call it This section of electrical signal is a bad EEG signal) and cannot be used for various application analysis of subsequent EEG signals.

At the same time, due to the muscle activity of the human body, the contact between the human skin and the electrodes may be temporarily detached or displaced. This situation will also cause sudden instability of the electrical signal captured by the EEG acquisition device, and it will take a period of time before the electrical signal can be passively stabilized by the electrode drive circuit. In addition, when an inappropriate electrode material is selected or the surface of the electrode material is oxidized, the electrode will be polarized. At this time, the electrodes will not be able to capture EEG signals normally unless the electrodes are replaced or repaired.

If the bad EEG signals generated in the above situations are mistakenly involved in brain emotion analysis, it may produce wrong emotion judgment results that are completely inconsistent with reality. These wrong emotional judgment results may further cause wrong guidance for psychological counseling, personal recuperation, etc. Therefore, whether it is in the actual application process of the EEG acquisition device or the product testing process, it is necessary to judge and analyze the quality of the EEG signal.

However, there is currently no effective method for determining and analyzing the quality of EEG signals, and it is impossible to effectively identify the EEG signals captured by the EEG acquisition device to distinguish good EEG signals from bad EEG signals. A large number of misjudgments occurred in the analysis of EEG applications.

Contents of the invention

Aiming at the defects existing in the prior art, the technical problem solved by the present invention is: quickly and effectively make real-time judgment and analysis on the quality of EEG signals, and effectively guarantee the accuracy of subsequent EEG application analysis.

In order to achieve the above object, the technical solution adopted by the present invention is: provide a kind of EEG signal quality real-time judging system, this system comprises the analysis data generating module connected in sequence, the standard deviation calculation module and the signal quality judging module; The generation module is used to: use the buffered EEG signal data and the generated time window to obtain the EEG signal segment corresponding to the time window; the standard deviation calculation module is used to: calculate the EEG signal segment in the current time window The average value of the average value is used to calculate the standard deviation of the EEG signal segment in the current time window; the signal quality determination module is used to: obtain the EEG signal segment in the current time window through the calculated standard deviation. Volatility determination result: the quality of the current EEG signal is jointly determined by using the volatility determination result of the current EEG signal segment and the volatility determination result of the previous EEG signal segment.

On the basis of the above technical solution, the analysis data generating module includes an EEG signal data receiving buffer unit and an EEG analysis time window generation unit; the EEG signal data receiving buffer unit is used to: The EEG signal data from the acquisition device; the EEG analysis time window generation unit is used to: generate a time window for EEG analysis according to the specified time interval cycle and time window length; receive the buffer unit from the EEG signal data In , extract the EEG signal segment corresponding to the time window length of the time window.

On the basis of the above technical solution, the time interval period is denoted as ΔT, the length of the time window is denoted as L, and the generated time window meets the following requirements:

ΔT=T _(n+1) -T _(n) ;

Wh _(n+1) = Wh _(n) +ΔT;

Wt _(n+1) = Wt _(n) +ΔT;

L>ΔT;

Wherein, T _(n) is the analysis time of the last EEG signal segment, T _(n+1) is the analysis time of the current EEG signal segment, Wh _(n+1) is the start time of the generated time window, Wh _(n) is the start time of the last time window, Wt _(n+1) is the end time of the generated time window, and Wt _(n) is the end time of the last time window.

On the basis of the above technical solution, the standard deviation calculation module includes an average value calculation unit and a standard deviation calculation unit: the average value calculation unit is used to: according to the value of each signal data piece contained in the current EEG signal segment, Calculate the average value of the current EEG signal segment through the average value calculation formula; the standard deviation calculation unit is used for: the average value calculated by the average value calculation unit and the value of each signal data piece contained in the EEG signal segment , calculate the standard deviation of the EEG signal segment in the current time window through the standard deviation calculation formula.

On the basis of the above-mentioned technical scheme, the formula for calculating the average value is:

In the formula, is the average value of the current EEG signal segment, E ₁ , E ₂ and E _L are the first signal data piece, the second signal data piece and the Lth signal data piece of the EEG signal piece respectively, and L is the EEG The length of the signal segment, that is, the length of the time window.

On the basis of the above-mentioned technical scheme, the formula for calculating the standard deviation is:

In the formula, SD is the standard deviation of the EEG signal segment in the current time window, is the average value of the current EEG signal segment, E _i is the i-th signal data segment of the EEG signal segment, i is a positive integer, and L is the length of the EEG signal segment, that is, the length of the time window.

On the basis of the above technical solution, the signal quality judging module includes an EEG signal fluctuation judging unit, a signal fluctuation judging historical result queue storage unit, and a signal quality correlation judging unit; the EEG signal fluctuating judging unit is used for: The standard deviation obtained is compared with the upper limit of the threshold value and the lower limit of the threshold value to obtain the volatility determination result of the EEG signal segment in the current time window; the signal fluctuation determination history result queue storage unit is used for: storing in the current time window The determination result of the volatility of the EEG signal segment; and the determination result of the volatility of the EEG signal segment of the previous time window is provided to the signal quality correlation determination unit; the signal quality correlation determination unit is used for: using the current time window The determination result of the volatility of the EEG signal segment and the determination result of the volatility of the EEG signal segment in the previous time window jointly determine the quality of the current EEG signal.

On the basis of the above technical solution, the specific process for the signal quality correlation determination unit to determine the quality of the current EEG signal is: compare the volatility determination result of the EEG signal segment in the current time window with the EEG signal in the previous time window The result of the volatility determination of the signal segment is logically ANDed to obtain the quality of the current EEG signal.

On the basis of the above technical solution, when the EEG signal fluctuation determination unit obtains the volatility determination result of the EEG signal segment in the current time window, it follows the following derivation process:

If SD≥Lower and SD≤Upper, then MD is 1;

If SD<Lower or SD>Upper, then MD is 0;

Among them, SD is the standard deviation of the EEG signal segment in the current time window, Lower is the lower limit of the threshold, Upper is the upper limit of the middle threshold, and MD is the determination result of the volatility of the EEG signal segment.

The beneficial effects of the present invention are:

(1) The present invention is an EEG signal quality judgment system added after the EEG signal data collection and before the EEG application analysis. The system includes an analysis data generation module for obtaining the EEG signal segment corresponding to the time window, a standard deviation calculation module for calculating the standard deviation of the EEG signal segment in the current time window, and a module for judging the current EEG signal quality The signal quality judgment module. Through the cooperation of the above three modules, the system can quickly and effectively make real-time judgment and analysis on the quality of EEG signals. Once bad signals are identified, subsequent EEG applications will actively discard the bad signals based on the recognition results. , or synchronously give a prompt operation of poor signal quality on the user interface, thereby effectively ensuring the accuracy of subsequent EEG application analysis.

(2) An EEG analysis time window generation unit is set in the analysis data generation module of the present invention. The unit adopts the form of overlapping time windows, so that newly received EEG signals can quickly participate in the calculation process of signal quality judgment, which not only reduces the delay of the time window for signal quality judgment, but also helps EEG analysis applications to timely target bad The signal is further manipulated; and the experience of using the EEG analysis application and the accuracy of the analysis results are improved.

(3) In the present invention, the standard deviation of the EEG signal segment in the current time window is used to determine the fluctuation range of the electrical signal, the calculation is fast, and the identification of the random jitter and non-random jitter of the signal has a good effect . Moreover, the standard deviation is not sensitive to the drift of the EEG signal, and it will not cause misjudgment of poor signal quality due to the drift of the EEG signal when the normal EEG signal is captured, and the stability is good.

(4) In order to make up for the shortcomings of directly using the standard deviation to measure the signal quality, the present invention optimizes it on the basis of the standard deviation algorithm. Specifically, a signal quality correlation judgment unit is provided in the signal quality judgment module, and the signal quality correlation judgment unit judges the threshold value of the current standard deviation (fluctuation judgment result) with the threshold judgment result of the previous standard deviation. The results are logically ANDed, and finally the quality judgment result of the current EEG signal is obtained, thereby reducing the shortage of directly using the standard deviation to measure the signal quality, improving the overall judgment accuracy rate of the EEG signal quality judgment, and reducing the EEG signal quality in most cases. Misjudgments occur.

Description of drawings

Fig. 1 is the structural block diagram of the real-time judgment system of EEG signal quality in the embodiment of the present invention;

Fig. 2 is the flowchart of the real-time judgment method of EEG signal quality in the embodiment of the present invention;

FIG. 3 is a schematic diagram of a waveform of a bad signal.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

Referring to FIG. 1 , an embodiment of the present invention provides a real-time EEG signal quality determination system, which determines the quality of EEG signals by analyzing the degree of dispersion of EEG signals captured by EEG acquisition equipment. In this embodiment, the EEG acquisition device can be used in homes, mental health clinics, yoga classrooms and other places that need to collect and analyze human brain EEG signals to determine human emotions.

The real-time judgment system of EEG signal quality includes an analysis data generation module, a standard deviation calculation module and a signal quality judgment module connected in sequence, and the analysis data generation module is also connected with an external EEG acquisition device.

Wherein, the analysis data generating module is used for: using the buffered EEG signal data and the generated time window to obtain the EEG signal segment corresponding to the time window.

Specifically, the analysis data generation module includes an EEG signal data receiving buffer unit and an EEG analysis time window generation unit. The EEG signal data receiving buffer unit is used for: receiving and storing the EEG signal data transmitted by the EEG acquisition device in real time; the EEG analysis time window generation unit is used for: according to the specified time interval period and time window The length generates a time window for EEG analysis; from the EEG signal data receiving buffer unit, the EEG signal segment corresponding to the time window length of the time window is taken out for use in the analysis and determination of the EEG signal quality.

In this embodiment, the time interval cycle is denoted as ΔT, and the length of the time window is denoted as L, then it can be known that the length of the EEG signal segment is the time window length L (because the EEG signal segment taken out is the same as the time window length of the time window Corresponding); and the time interval cycle ΔT=T _(n+1) -T _(n) , wherein, T _(n) is the analysis time of the last EEG signal segment, T _(n+1) is the current EEG The analysis time of the signal fragment. In addition, it can be understood that in the real-time judgment system of EEG signal quality, except for the EEG signal segment defined by the first time window, the EEG signal segments defined by other time windows are all consistent with the EEG signal segment of the last time window. Fragments partially overlap. In order to ensure that the EEG signal segment corresponding to the time window length L of the time window can be obtained each time the EEG signal segment analysis is performed, the latest received EEG signal data can also be analyzed in real time. The start time Wh _(n+1) of the time window generated by the EEG analysis time window generation unit=the start time Wh _(n) +ΔT of the last time window, the end time Wt _{( n+1)} = end time of the last time window Wt _(n) +ΔT, and the length of the time window L (that is, the length L of the EEG signal segment) > the time interval period ΔT.

The standard deviation calculation module is used to: calculate the average value of the current EEG signal segment, and use the average value to calculate the standard deviation of the EEG signal segment in the current time window.

Specifically, the standard deviation calculation module includes an average value operation unit and a standard deviation operation unit, wherein:

The average value calculation unit is used to calculate the average value of the current EEG signal segment according to the value of each signal data piece contained in the current EEG signal segment through the average value calculation formula In this embodiment, the formula for calculating the average value is:

In the formula, E ₁ , E ₂ and E _L are the first signal data slice, the second signal data slice and the Lth signal data slice of the EEG signal segment respectively, and L is the length of the EEG signal segment (that is, time window length L). It can be understood that the number of signal data pieces in the EEG signal segment is equal to the length of the EEG signal segment.

The standard deviation operation unit is used to: calculate the EEG signal in the current time window according to the average value calculated by the average value operation unit and the value of each signal data piece included in the EEG signal segment, through the standard deviation calculation formula Standard deviation SD of fragments. In this embodiment, the formula for calculating the standard deviation is:

In the formula, E _i is the i-th signal data slice of the EEG signal segment, and i is a positive integer.

The signal quality determination module is used to: obtain the volatility determination result of the EEG signal segment in the current time window through the calculated standard deviation; use the volatility determination result of the current EEG signal segment and the previous EEG signal segment The results of the volatility judgment jointly determine the quality of the current EEG signal.

Specifically, the signal quality determination module includes an EEG signal fluctuation determination unit, a signal fluctuation determination history result queue storage unit, and a signal quality correlation determination unit, wherein:

The EEG signal fluctuation judging unit is used for: comparing the calculated standard deviation with the upper limit of the threshold value and the lower limit of the threshold value, so as to obtain the determination result of the fluctuation of the EEG signal segment in the current time window. In this embodiment, when the EEG signal fluctuation determination unit obtains the volatility determination result of the EEG signal segment in the current time window, it follows the following derivation process:

If SD≥Lower and SD≤Upper, then MD is 1;

If SD<Lower or SD>Upper, then MD is 0;

Among them, Lower is the lower limit of the threshold, Upper is the upper limit of the middle threshold, and MD is the determination result of the volatility of the EEG signal segment.

It can be understood that the standard deviation SD of the EEG signal segment is used to reflect the degree of dispersion of an EEG signal segment. The standard deviation of a good-quality EEG signal is generally less than or equal to the upper threshold upper limit of the EEG signal fluctuation determination unit, and is also greater than or equal to the lower threshold lower limit of the EEG signal fluctuation determination unit.

The signal fluctuation determination history result queue storage unit is used to: store the volatility determination result MD _(n) of the EEG signal segment in the current time window; and provide the EEG signal segment of the previous time window to the signal quality correlation determination unit The volatility determination result MD _(n-1) of .

The signal quality correlation determination unit is used to: use the volatility determination result MD _(n) of the EEG signal segment in the current time window and the volatility determination result MD _(n-1) of the EEG signal segment in the previous time window , jointly determine the quality of the current EEG signal.

In this embodiment, the specific process of determining the quality of the current EEG signal by the signal quality correlation determination unit is: combining the volatility determination result of the EEG signal segment in the current time window with the EEG signal segment in the previous time window Perform logical AND operations on the volatility judgment results to obtain the quality of the current EEG signal, that is, MD _(n) ∧MD _(n-1) = Q _(n) , where Q _(n) is the quality judgment of the current EEG signal result.

It can be understood that, in order to ensure the real-time performance of the calculation, the data length in the current time window used for each determination cannot be too long. Due to some bad signals caused by the electrode polarization of the EEG acquisition equipment, it is difficult to use the standard deviation to directly judge the volatility or goodness of the EEG signal with a short data length, so it is necessary to use the historical results MD _{( n-1)} to jointly determine the quality of the current EEG signal.

Referring to Figure 2, an embodiment of the present invention provides a method for real-time determination of EEG signal quality, comprising the following steps:

S1: Receive and store the EEG signal data from the EEG acquisition device in real time; generate a time window for EEG analysis, and go to S2;

S2: Using the generated time window, take out the EEG signal segment corresponding to the time window from the stored EEG signal data, and transfer to S3;

S3: Calculate the average value of the current EEG signal segment according to the value of each signal data piece contained in the EEG signal segment in the current time window, and turn to S4;

S4: Calculate the standard deviation of the EEG signal segment in the current time window according to the calculated average value and the value of each signal data piece included in the EEG signal segment, and go to S5;

S5: Comparing the calculated standard deviation with the upper limit of the threshold value and the lower limit of the threshold value to obtain the determination result of the volatility of the EEG signal segment in the current time window, and go to S6;

S6: Save the volatility determination result of the EEG signal segment in the current time window; and obtain the volatility determination result of the EEG signal segment in the previous time window, and turn to S7;

S7: jointly determine the quality of the current EEG signal by using the fluctuation determination result of the EEG signal segment in the current time window and the fluctuation determination result of the EEG signal segment in the previous time window.

In actual operation, when generating the time window for EEG analysis described in S1, the time window for EEG analysis is generated according to the specified time interval cycle ΔT and the length of the time window L; and the generated time window satisfies the following Require:

ΔT=T _(n+1) -T _(n) ;

Wh _(n+1) = Wh _(n) +ΔT;

Wt _(n+1) = Wt _(n) +ΔT;

L>ΔT;

Wherein, T _(n) is the analysis time of the last EEG signal segment, T _(n+1) is the analysis time of the current EEG signal segment, Wh _(n+1) is the start time of the generated time window, Wh _(n) is the start time of the last time window, Wt _(n+1) is the end time of the generated time window, and Wt _(n) is the end time of the last time window.

Further, in actual operation, S3 specifically includes the following steps: according to the value of each signal data piece contained in the current EEG signal segment, calculate the average value of the current EEG signal segment through the average value calculation formula The formula for calculating the average value is:

In the formula, E ₁ , E ₂ and E _L are respectively the first signal data piece, the second signal data piece and the Lth signal data piece of the EEG signal segment, and L is the length of the EEG signal segment.

Further, in actual operation, S4 specifically includes the following steps: according to the average value calculated by the average value calculation unit and the values of each signal data slice included in the EEG signal segment, and through the standard deviation calculation formula, calculate The standard deviation SD of the EEG signal segment; the standard deviation calculation formula is:

In the formula, E _i is the i-th signal data slice of the EEG signal segment, and i is a positive integer.

Further, when the determination result of the volatility of the EEG signal segment in the current time window is obtained as described in S5, the following derivation process is followed:

If SD≥Lower and SD≤Upper, then MD is 1;

If SD<Lower or SD>Upper, then MD is 0;

Among them, Lower is the lower limit of the threshold, Upper is the upper limit of the middle threshold, and MD is the determination result of the volatility of the EEG signal segment.

Furthermore, in actual operation, S7 specifically includes the following steps: perform logical AND operation on the volatility determination result of the EEG signal segment in the current time window and the volatility determination result of the EEG signal segment in the previous time window, to obtain The quality of the current EEG signal, that is, MD _(n) ∧MD _(n-1) = Q _(n) , where Q _(n) is the quality judgment result of the current EEG signal.

In order to better understand the present invention, the main design principles and basis of the present invention are analyzed in detail below:

In the follow-up EEG analysis application, bad signals need to be filtered in time to avoid a small amount of bad signals entering the EEG analysis process and causing serious errors in the analysis results; moreover, the judgment of signal quality needs to take EEG signal data for a period of time, which is objective Compared with the original data, the above analysis results will have a certain delay. Therefore, in order to provide more real-time EEG signal quality determination results for subsequent EEG applications, the design of the analysis window is very important.

Furthermore, the standard deviation algorithm with low calculation cost is usually used to judge the degree of signal fluctuation. Standard Deviation is often used as a measurement algorithm for statistical dispersion in probability statistics. It reflects the degree of dispersion among individuals in a set of data. But in special cases, as shown in Figure 3, in the middle of the waveform of the bad signal, the signal rises in a curve, and the standard deviation is not large at this time, but the smooth signal does not reflect the fluctuation of the EEG below 50Hz. Therefore, there are inherent deficiencies in directly using the standard deviation to measure the signal quality.

In view of this, the present invention optimizes on the basis of the standard deviation algorithm, and performs logical AND operations on the threshold judgment result (fluctuation judgment result) of the standard deviation this time and the threshold judgment result of the previous standard deviation, and finally obtains The current quality judgment results of EEG signals reduce the shortage of directly using the standard deviation to measure the signal quality, thereby improving the overall judgment accuracy rate of EEG signal quality judgments, and can reduce the occurrence of misjudgments in most cases.

The present invention is not limited to the above-mentioned embodiments. For those of ordinary skill in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications are also considered protection of the present invention. within range. The content not described in detail in this specification belongs to the prior art known to those skilled in the art.

Claims (8)

1. a kind of EEG signals quality realtime analysis system, it is characterised in that：The system includes the analysis data life being sequentially connected At module, standard deviation computing module and signal quality determination module；

The analysis data generation module is used for：Using the EEG signals data of caching and the time window of generation, obtain and when Between the corresponding EEG signals segment of window；

The standard deviation computing module is used for：The average value for calculating the EEG signals segment in actual time window, utilizes this Mean value calculation goes out the standard deviation of the EEG signals segment in actual time window；

The signal quality determination module is used for：The EEG signals piece in actual time window is obtained by calculated standard deviation The fluctuation sex determination result of section；Utilize the fluctuation sex determination result of current EEG signals segment and EEG signals segment before Fluctuation sex determination as a result, determining the quality of current EEG signals jointly；

The signal quality determination module includes EEG signals fluctuation judging unit, signal fluctuation judgement historical results queue storage Unit is associated with judging unit with signal quality；

The EEG signals fluctuation judging unit is used for：Calculated standard deviation is compared with the threshold upper limit, threshold lower limit, Obtain the fluctuation sex determination result of the EEG signals segment in actual time window；

Signal fluctuation judgement historical results queue storage location is used for：Store the EEG signals segment in actual time window Fluctuation sex determination result；And the fluctuation of the EEG signals segment of a upper time window is provided to signal quality association judging unit Sex determination result；

The signal quality association judging unit is used for：Utilize the fluctuation sex determination of the EEG signals segment in actual time window As a result with the fluctuation sex determination of the EEG signals segment of a upper time window as a result, determining the matter of current EEG signals jointly Amount.

2. EEG signals quality realtime analysis system as described in claim 1, it is characterised in that：The analysis data generate mould Block includes EEG signals data receiver buffer unit and brain electricity analytical time window generation unit；The EEG signals data receiver Buffer unit is used for：Real-time reception simultaneously stores the EEG signals data transmitted by brain wave acquisition equipment；The brain electricity analytical time Window generation unit is used for：The time window for brain electricity analytical is generated according to specified time interval period and time window length Mouthful；From EEG signals data receiver buffer unit, EEG signals corresponding with the time window length of time window are taken out Segment.

3. EEG signals quality realtime analysis system as claimed in claim 2, it is characterised in that：The time interval period note For Δ T, the time window length is denoted as L, and the time window generated meets claimed below：

Δ T=T_(n+1)–T_(n)；

Wh_(n+1)=Wh_(n)+ΔT；

Wt_(n+1)=Wt_(n)+ΔT；

L ＞ Δs T；

Wherein, T_(n)For the analysis time of last EEG signals segment, T_(n+1)For the analysis time of current EEG signals segment, Wh_(n+1)Initial time for the time window generated, Wh_(n)For the initial time of last time time window, Wt_(n+1)To be generated Time window the termination time, Wt_(n)For the termination time of last time time window.

4. EEG signals quality realtime analysis system as described in claim 1, it is characterised in that：The standard deviation computing module Including average value arithmetic element and standard deviation arithmetic element：The average value arithmetic element is used for：According to current EEG signals piece The value for each signal data piece that section is included calculates the average value of current EEG signals segment by mean value calculation formula； The standard deviation arithmetic element is used for：Included according to the calculated average value of average value arithmetic element and EEG signals segment The value of each signal data piece the mark of the EEG signals segment in actual time window is calculated by standard deviation calculation formula It is accurate poor.

5. EEG signals quality realtime analysis system as claimed in claim 4, it is characterised in that：The mean value calculation formula For：

In formula,For the average value of current EEG signals segment, E₁、E₂And E_LRespectively the 1st signal of EEG signals segment Data slice, the 2nd signal data piece and l-th signal data piece, L are the length of EEG signals segment, i.e. time window is long Degree.

6. EEG signals quality realtime analysis system as claimed in claim 4, it is characterised in that：The standard deviation calculation formula For：

In formula, SD is the standard deviation of the EEG signals segment in actual time window,For being averaged for current EEG signals segment Value, E_iFor i-th of signal data piece of EEG signals segment, i is positive integer, and L is the length of EEG signals segment, i.e. time window Mouth length.

7. EEG signals quality realtime analysis system as described in claim 1, it is characterised in that：The signal quality association is sentenced The detailed process that order member determines current EEG signals quality is：By the fluctuation of the EEG signals segment in actual time window Sex determination result and the fluctuation sex determination result of the EEG signals segment of a upper time window carry out logic and operation, obtain current The quality of EEG signals.

8. EEG signals quality realtime analysis system as described in claim 1, it is characterised in that：The EEG signals fluctuation is sentenced When order member obtains the fluctuation sex determination result of the EEG signals segment in actual time window, it then follows following derivation：

If SD >=Lower and SD≤Upper, MD 1；

If SD ＜ Lower or SD ＞ Upper, MD 0；

Wherein, SD is the standard deviation of the EEG signals segment in actual time window, and Lower is bottom threshold, and Upper is threshold value The upper limit, MD are the fluctuation sex determination result of EEG signals segment.