DOA Estimation Using Wideband OFDM Signals (CSM) - Algorithm and Implementation

Wideband OFDM signals are widely used in wireless communications, and their multi-carrier characteristics provide an excellent signal foundation for Direction of Arrival (DOA) estimation. In the CSM (Coherent Signal-Subspace Method) based DOA estimation framework, the first step involves implementing focusing processing for wideband signals to overcome the performance degradation of traditional narrowband algorithms in wideband scenarios. This typically requires frequency domain transformation and sub-band grouping operations in code implementation.

The core principle of the CSM algorithm involves constructing focusing matrices to align signal subspaces from different frequency sub-bands to a reference frequency, thereby creating a coherent signal subspace. This step effectively addresses the effects of frequency dispersion in wideband signals. In MATLAB implementation, this can be achieved using eigenvalue decomposition and matrix transformation operations. Subsequently, the classical MUSIC (Multiple Signal Classification) algorithm is applied to perform eigendecomposition on the focused covariance matrix. The algorithm then utilizes the orthogonality between noise subspace and signal subspace to construct a spatial spectrum through pseudo-spectrum calculation functions.

Finally, peak search algorithms are employed to locate extreme points on the spatial spectrum, where the corresponding angles represent the estimated directions of arrival. This method combines the advantages of wideband signal processing with the high-resolution capabilities of the MUSIC algorithm, demonstrating excellent robustness and estimation performance in complex electromagnetic environments. It is particularly suitable for multi-target direction finding scenarios in radar and communication systems, where implementation typically involves array signal processing toolboxes and spectrum peak detection functions.