چكيده به لاتين
Using compressed sensing, sparse or compressive signals can be reconstructed from the measurement samples with a number much less than the Nyquist rate. Sparsity of impulse response of wireless communication channels enables us to utilize compressed sensing for sparse channel estimation. Moreover, the theory of sparse recovery with side information can be employed to simultaneously reconstruct signals with multiple structures.
In this thesis, the theory of sparse recovery with side information is applied to estimate dynamic sparse channels in the time domain for OFDM systems. To this end, firstly, the time correlation property of the dynamic sparse channel model is considered. Then, in order to estimate and track the time variations in the framework of sparse recovery theory, a cost function including l_1-l_1 norm minimization problem is proposed. Finally, through imposing some conditions, the unique response of the proposed problem is formulated as a closed form solution.
Simulation results show that the accuracy of the proposed estimation and tracking method of dynamic sparse channels significantly outperforms that of BP, DOMP and SMP algorithms, in the sense of normalized mean square error.
Key words: Dynamic sparse channel, sparse recovery with side information, mixed norms, OFDM systems.
