Frequency-phase brain-computer interface decoding method and device based on fft spectrum correction

Abstract

The invention discloses a frequency phase brain-computer interface decoding method based on FFT spectrum correction. The method comprises the steps that digital sampling is carried out on collected SSVEP signals to obtain N discrete samples, windowing is carried out on the discrete samples for FFT analysis, and the spectrum of the SSVEP signals is obtained; a peak value spectrum is sought out, the phase value of the peak value spectrum is recorded, the secondary high spectrum position needs to be determined, and a ratio is obtained; a frequency offset estimated value is obtained through the ratio, and corresponding phase estimated values are obtained based on the frequency offset estimated value; a stimulation object is recognized by solving measurement phase errors of two corrected stimulation frequencies. According to the decoding device, the collected signals are sampled by an analog-digital converter to obtain a sample sequence, the collected signals enter into a DSP device in a parallel digital input mode, internal processing is carried out, and parameter estimation of the signals is obtained; then a command sent by a tested person is displayed through an output drive and a display module of the output drive, and finally external equipment responds to the corresponding command.

Description

technical field [0001] The invention relates to the field of digital signal processing, in particular to a frequency-phase brain-computer interface decoding method and device based on FFT spectrum correction. , and the command decoding is realized by extracting the phase information through the FFT spectrum correction method. Background technique [0002] brain-computer interface [1] (Brain-Computer Interface, abbreviated as BCI) is a direct information exchange and control channel established between the human brain and a computer or other electronic devices. It does not depend on the normal output pathways of the brain (peripheral nervous system and muscle tissue). ), is a new communication and control method [2] . It aims to establish a direct communication channel between the human brain and the outside world, identify brain instructions by extracting the characteristics of EEG signals, and finally complete the direct control of external devices by the brain. [0003...

AI Technical Summary

Problems solved by technology

[0007] Obviously, because the decoding process in literature [14] is carried out in the time domain, it is highly sensitive to noise; the decoding in literature [12] is directly realized in the FFT frequency domain, but the parameters of this literature can be found that, The excitation frequencies are all selected as integer multiples of the FFT frequency resolution (just corresponding to the frequency offset of each excitation frequency is 0, so that there is no error in the direct FFT phase measurement), so the number of excitation targets is still limited (the same limitation also exists in the literature [ 6,10,12,15] in)

The reason why the excitation frequency without frequency offset is selected is to avoid the inherent defect of inaccurate phase measurement caused by FFT spectrum leakage in the case of frequency offset

Because of spectral leakage, even for a single-frequency signal, there will be a large error in the phase value on its peak spectral line; when the signal contains multiple frequency components, the interspectral interference caused by the spectral leakage of each component will further increase. large phase measurement error [16-18]

Therefore, in order to improve the frequency utilization rate within the limited bandwidth, it is urgent to solve the FFT phase decoding problem of the steady-state visual evoked potential in the presence of frequency offset

Images