چكيده به لاتين
Abstract:
One of the most important ongoing issues of these days is designing Brain Computer Interfaces to investigate brain functionality and extract kinetic and kinematic parameters which are used to simulate movement types. Hence this would be possible to control a robot using brain signals, which has various applications.
In this Thesis, the purpose is to decode the force parameter from Rat intracortical(LFP) signals. The signals are recorded using a 16-channel micro wire array from forelimb-related area of the motor cortex. force signals also are recorded by a force sensor simultaneously.
After signal preprocessing methods such as DC cancellation, power noise cancellation, CAR filtering and etc., decoding procedure is applied using PLS method as a linear regression and Kalman filtering method.
After all, the performance of the two methods are compared.
According to the results, this is possible to estimate force amplitude from Rat motor cortex interacortical( LFP) signals with Correlation Coefficient of 0.72 and Nrmse of 0.19. Using this decoded parameter and due to the stability of LFP signals compared to Spike trains, under consideration of the low channel array we can expect performance of the BCIs to get improved.
Keywords: BCI, Rat, Force, Decode, PLS, Kalman
