Deterministic pilot design for MIMO OFDM system based on compressed sensing

We consider the problem of pilot design for sparse multipath channel estimation in multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) system in this paper. In order to reconstruct MIMO sparse channel, the prevalent compressed sensing (CS) theory is applied. Contrary to random pilot construction, a deterministic pilot design method is exploited to form the measurement matrix by deterministically extracting some rows of Fourier matrix that satisfies Restricted Isometry Property (RIP). Two families of continuous functions, namely f₁ -SL0 and f₂ -SL0, are investigated in this paper to approximate ℓ₀ norm for MIMO sparse channel reconstruction. Simulation results suggest that the deterministic pilot design scheme performs as well as the random one at both low and high signal noise ratio (SNR). Moreover, f₁ - SL0 and f₂ -SL0 outperform ℓ₁ -dantzig algorithm with faster running time and better anti-noisy performance.