Research On Compressive Channel Sensing Methods Under Power Leakage

Due to its high spectral efficiency and data transmission rate, the orthogonal frequency division multiplexing (OFDM) technology is introduced. In order to ensure high-speed transmission and high-quality performance, the receiver uses coherent reception generally. That is, the receiver must estimate the channel state information between the transmitter and receiver to complete the signal reception. The accuracy of channel estimation (CE) is an important index to measure the performance of the OFDM system. However, channel power leakage will have a great impact on the CE performance. Based on this background, this paper focuses on the channel power leakage scenario in OFDM system, and propose the channel estimation methods to suppress power leakage to further improve the system performance, including channel sparse representation and channel estimation algorithm. This paper focuses on the high-speed movement and Non-sample-Spaced Multipath power leakage scenarios.The high-speed movement between the Base Station (BS) and Mobile Station would cause large Doppler shifts and Inter-Carrier Interference (ICI),which will degrade the channel estimation severely due to the channel power dispersion. Moreover, the channel structure may vary fast such that the channel structure-based estimation methods require a large number of pilots.Therefore, one key challenge is to design an efficient and reliable estimation method to adapt to the fast time-varying channels in high-speed movement scenarios. In this paper, a basis expansion model (BEM) based compressive channel sensing algorithm is proposed to estimate the channel state information (CSI) for OFDM systems under fast fading channels. Firstly, in order to mitigate the channel power leakage, Discrete Prolate Spheroidal Sequences (DPSS) are employed as the basis of the channel model, the DFT-DPSS basis is formed to represent the CSI sparsity of time delay -Doppler domain. Then, by fully exploiting the sparse characteristics of the channel, an Orthogonal Matching Pursuit (OMP) algorithm with optimal thresholding is developed, which can flexibly support the varying channel structure without channel side information. At last, Monte Carlo simulations are presented to evaluate the performance of the proposed method, and the results show that it greatly enhances the CE performance in the fast time-varying channels.In broadband OFDM systems, the channel features sparse structure. By fully exploiting the sparse structure of the Channel Impulse Response (CIR),transform domain (TD) channel estimation (CE) methods such as discrete Fourier transform (DFT) method and compressed sensing (CS)-based methods can greatly upgrade the performance of channel estimation. When the paths of the channel are not located in the sampling grids, which is a normal case due to the randomly distributed path delay, power leakage effect happens. This Non-sample-Spaced problem will severely degrade TD methods, then we develop a CS-based method combines first-order Taylor approximation (CBP-T) to estimate the channel impulse information. The improvement of the first-order Taylor approximation method to resist power leakage is analyzed, and a Continuous Basis Pursuit (CBP) method to reconstruct channel is also analyzed. Based on the rotation invariant characteristic of the pilot subcarrier sets which arrange as the uniform pilot pattern, the ESPRIT algorithm is applied to estimate. At last, Monte Carlo simulations are presented to evaluate the performance of the proposed two methods, and the results show that the proposed methods solve the problem of channel power leakage effectively.