<p>The randomized extended Kaczmarz method proposed by Zouzias and Freris (2013) is highly effective for solving large-scale inconsistent linear systems. However, when the coefficient matrix is column-rank deficient or contains many linearly dependent columns, the method tends to select such linearly dependent columns, slowing down the convergence rate. In this paper, we derive an efficient block iteration scheme for column iteration steps of the randomized extended Kaczmarz method and construct a column-block enhanced randomized extended Kaczmarz method. We further establish the convergence theory of the column-block enhanced randomized extended Kaczmarz method, which exhibits a faster convergence rate than the original randomized extended Kaczmarz method. Finally, numerical experiments are conducted to validate the effectiveness of our approach in terms of both iteration counts and computing times.</p>