Poor stability of resonant frequency estimation during wireless passive surface acoustic wave （SAW）sensor measurements can limit the performance of SAW reader measurement systems. To improve the stability of frequency estimation, in this study, a vector reader architecture is proposed based on traditional power spectrum frequency estimation methods. Furthermore, phase detection technology is introduced to obtain vector spectra. Additionally, an optimization method combining phase measurement and filtering processing is proposed. This method uses inverse fast Fourier transform （IFFT） and finite impulse response （FIR） filtering methods to improve the signal-to-noise ratio of the echo signal, and it combines multiple model fitting methods to optimize the estimation of resonance frequencies for different peak characteristics. To verify the effectiveness of this method, a wireless passive testing platform for SAW sensors was constructed, and wireless frequency testing experiments were conducted to acquire multiple sets of SAW echo signal data for processing. The experimental results show that after filtering, the signal-to-noise ratio of the echo signal is improved by approximately 17 dB. After further optimization with fitting methods, the standard deviation of resonant frequency estimation is reduced to below 0.5 kHz, indicating that this method effectively improves the frequency estimation sensitivity and stability of the surface acoustic wave reader system.