This paper proposes a new mono-pulse radar beam sharpening method with multichannel L1 regularization. First, a multi-channel L1 regularization model is derived for mono-pulse radar beam sharpening based on the maximum a posteriori probability criterion. Then, an extended iterative shrinkage threshold algorithm is proposed to solve the multi-channel L1 regularization problem. Theoretical analysis and simulation results show that this new method can ensure beam sharpening performance and improve the ability to suppress noise, and it can efficiently solve the noise leaks problem caused by channels pattern, which does not meet the strong prime conditions in mono-pulse radar. The performance of the proposed algrithem is significantly better than the existing mono-pulse beam sharpening methods.