A Blind Signal Localization and SNR Estimation Method
Johanna Vartiainen, Harri Saarnisaari, Janne J. Lehtomäki, Markku J. Juntti · 2006
Congested radio frequencies call for efficient and flexible spectrum use and, hence, spectrum sensing. As cognitive radios need channel information for achieving better use of radio frequencies, military systems require information about unknown signals. In this paper, a novel method for computing signal-to-noise ratios (SNR) of several unknown narrowband signals without a priori information is proposed. The presented frequency domain method is an extension of the localization algorithm based on double-thresholding (LAD). The main point is to produce fast and cost-efficient estimators with adequate accuracy. The proposed method is verified via computer simulations and tested also with real-life radio channel measurement data. The results verify that the proposed method gives sufficient approximations of SNR values with low overall computational complexity