چكيده به لاتين
Brain-computer interfaces (BCIs) seek to establish a direct connection from the brain to the computer, for use in applications such as motor prosthesis control, control of a cursor on the monitor, and so on. Hence, the accuracy of movement decoding from brain signals in BCIs is crucial. The Kalman filter is generally used in BCIs systems to decode neural activity and estimate the kinetic and kinematic parameters. To use the Kalman filter, the state transition matrix, the observation matrix and the covariance matrices of the process and measurement noises must be known in advance. However, in many applications, these matrices are not known. Typically, to estimate these parameters, the ordinary least squares (OLS) method and the sample covariance matrix (SCM) estimator are used. Our purpose is to enhance the decoding performance of the Kalman filter in BCI systems by improving the estimation of the mentioned parameters. Here, we propose the Regularized Kalman Filter (RKF) which implements two fundamental features: 1) Regulating the state equation to improve the estimation of the state transition matrix, and 2) Use of shrinkage method to improve the estimation of the unknown measurement noise covariance matrix. The performance of the proposed method has been investigated on two datasets of local field potential in the motor cortex of a monkey (Estimation of kinematic parameters during hand movement) and three rats (Estimation of the amount of force as a kinetic parameter applied by the hand). The results demonstrate that the proposed method outperforms the standard Kalman filter, the Kalman filter with feature selection, the partial least squares (PLS) and the Ridge regression.
