Accurate determination of the state of ionosphere has become a key point as radio-wave information systems become more spatially dependent. Data assimilation technique is an important method to obtain ionospheric parameters that is useful in ionospheric study. To provide specification of the ionosphere total electron content(TEC)for current conditions, a new ionospheric TEC reconstruction method is developed in this paper by utilizing data assimilation technique. With the output of NeQuick model as background field and Kalman filter as assimilation algorithm, regional three dimensional ionospheric electron density reconstruction can be implemented by data assimilation of 39 GPS observations from continuously operating reference stations(CORS) of NOAA(National Oceanic and Atmospheric Administration), while TEC map can be obtained by vertical integral of electron density al. For reducing the reduction in the computational requirements, the regional TEC maps are generated accordingly with the spatial and temporal resolution being 2.5° ×5° and 1 hour respectively. Data assimilation results have been validated through the comparison with 10 independent GPS observations, the NeQuick model values and the global ionosphere maps(GIM) from Center for Orbit Determination of Europe(CODE) over 31 days in December, 2014. Results show that data assimilation derived TEC maps are more effective in reducing the mean error and standard error of TEC than NeQuick model and GIM data. These results demonstrate the potential of data assimilation technique in providing accurate regional or even global specification of the ionosphere.