Channel Covariance Matrix Construction for DOA Estimation with Limited Communication Symbols

The covariance matrix serves an important role in direction-of-arrival (DOA) estimation for joint sensing and communication systems. However, under the condition of limited communication symbols, the conventional sample data covariance matrix derived from received symbols deviates from its ideal version, leading to a deteriorated DOA estimation performance. To address this issue, we formulate a sample channel covariance matrix from the estimated sensing channels in the frequency domain. It is proved that, by imposing sufficient subcarriers, the approximation error between the sample channel covariance matrix and the ideal one can be minimized. As such, the constructed sample channel covariance matrix facilitates an enhanced DOA estimation performance even with a small number of communication symbols. The effectiveness of the proposed method is validated via simulations in terms of both estimation accuracy and angular resolution.