The few-trial extraction of evoked potentials is very meaningful to the study of brain and many clinical applications.According to the characteristics of Electroencephalogram signal

this paper presents a novel algorithm for double-trial extracting evoked potentials based on joint sparse representation.Taking advantage of the quasi-periodic structure of evoked potentials and randomness of ongoing spontaneous Electroencephalogram

the observations of evoked potentials are considered as the superposition of the similar components and the different components.Evoked potential obtained by same stimulation of the nerves changes only in latency and scale parameters.Our method uses the average evoked potentials to model and construct the sparse dictionary

so the double-trial extraction of evoked potentials can be achieved with joint sparse representation.Experiment results show that the performance of the proposed method is better than that of other methods.