Advancing Multiband OFDM Channel Sounding: An Iterative Time Domain Estimation for Spectrally Constrained Systems

Abstract

The emerging wireless applications are facing new challenges in combating frequency congestion problems. As a result, the opportunistic utilization of available frequency spectrum and channel bonding is becoming increasingly common in new wireless standards. In these systems, the transmit waveforms are required to have nulls in specific frequency bands to avoid interference with primary users. However, these nulls can significantly affect the performance of channel estimation algorithms. This work proposes a novel Iterative Multiband (MB) Spectrally Constrained Time-Domain (SCTD) technique to developed to reduce the residual error of correlation due to spectrally constrained waveforms. The performance of the newly developed technique is evaluated through extensive numerical experiments, where the Mean Squared Error (MSE) and Bit Error Rate (BER) are computed for various scenarios. The accuracy of the proposed technique is compared with known channel state information and with conventional techniques. The simulation results show that the proposed Time-Domain Iterative Method, SCTD, performs better than conventional techniques for various Rayleigh channel conditions with Additive White Gaussian Noise (AWGN). It was found that after ten iterations, the proposed technique outperforms the conventional technique for both stationary and mobile frequency-selective channels. Furthermore, it was observed that the proposed SCTD technique requires fewer pilot signals to achieve a similar performance. The results show that the proposed SCTD method supersedes the conventional techniques for stationary and mobile frequency selective channel scenarios within ten iterations. Subsequently, it is observed that the proposed SCTD method requires 50% fewer pilots to provide similar performance compared to conventional methods. It is also observed that the proposed SCTD method provides an average of 6 dB mean squared error (MSE) advantage for low signal-to-noise ratio per bit (𝐸 𝑏 𝑁 0 ⁄ ) regime cases. It can be concluded that the proposed technique is highly suitable, particularly for low 𝐸 𝑏 𝑁 0 ⁄ regimes and can be used for various communication systems.