Based on real-time multi-domain communication signal analysis architecture, a high-efficiency blind carrier frequency estimation algorithm using the power spectrum symmetry of the measured modulated signal is presented. The proposed algorithm, which utilizes the moving averaged power spectrum achieved by the realtime spectrum analysis, iteratively identifies the carrier frequency in according to the power difference between the upper sideband and lower sideband, which is defined and revised by the estimated carrier frequency in each iteration. When the power difference of the two sidebands converges to the preset threshold, the carrier frequency can be obtained. For the modulation analysis, the measured signal can be coarsely compensated by the estimated result, and the residual carrier frequency error is eliminated by a following carrier synchronization loop. Compared with previous works, owing to the moving averaged power spectrum normalization and the smart iterative step variation mechanism for the two sidebands definition, the carrier frequency estimation accuracy and speed can be significantly improved without increasing the computational effort. Experimental results are included to demonstrate the outstanding performance of the proposed algorithm.