Abstract: Objectives To address the problems of structural complexity, large size, and high cost in traditional direction-finding platforms and to achieve miniaturization and portability of direction-finding equipment, this paper proposes a multi-snapshot atomic norm minimization direction-of-arrival (DOA) estimation algorithm based on channel switching. Methods A switching matrix is constructed whose configuration randomly varies with each sampling instance, generating the switched observation data. Under a multi-snapshot noisy model, the atomic norm minimization problem is formulated and transformed into a semidefinite programming (SDP) problem. The optimal solution of the SDP is used to construct a Toeplitz matrix, from which the DOA information of the signals is extracted via Vandermonde decomposition. Results Simulation results show that when the number of array elements is 24, the number of receiving channels is 12, the signal-to-noise ratio is 5 dB, and the number of snapshots is 100, the proposed algorithm achieves a root-mean-square error below 0.1°. Moreover, when the angular separation between two signals is 0.6°, the probability of successful resolution reaches 100%. Conclusions The proposed method effectively reduces system complexity and hardware cost while maintaining high direction-finding accuracy and angular resolution, providing a feasible solution for compact and high-precision DOA measurement systems.