Ultrawideband Channel Estimation: A Bayesian Compressive Sensing Strategy Based on Statistical Sparsity

Compressive Sensing Techniques for Next-Generation Wireless Communications

“…2) Larger transmission bandwidth may be invoked relying on both CR [7], [8] and UWB [9], [10] techniques, both of which can coexist with licenced services under the umbrella of spectrum sharing, where the employment of sub-Nyquist sampling is of salient importance. As another promising candidate, mmWave communications is capable of facilitating high data rates with the aid of its wider bandwidth [1], [11], [12].…”

“…However, the intrinsic time-domain sparsity of the received line-of-sight (LOS) or non-line-of-sight (NLOS) UWB signals inspires the employment of an efficient sampling approach under the framework of CS, where the sparse UWB signals can be recovered by using sub-Nyquist sampling rates. Moreover, the UWB signals received over multipath channels can also be approximately considered as a linear combination of several signal bases, as in the standard CS Model (1) of Table I, where these signal bases are closely related to the UWB waveform, such as the Gaussian pulse or it derivatives [9], [10]. Compared to those users, who only exploit the time-domain sparsity of UWB signals, the latter approach can lead to a higher energy-concentration and to the further improvement of the sparse representation of the received UWB signals, hence enhancing the reconstruction performance of the UWB signals received by using fewer measurements.…”

Efficient Bayesian compressed sensing-based channel estimation techniques for massive MIMO-OFDM systems

“…Consequently, the channels at different time instants/locations are different but share the same common statistical property. As a result, to estimate the current channel, we can exploit the previous compressive vectors in addition to the current compressive vector [15].…”

“…. J where J is the number of the task, A n c * ,j , g n c * ,c * ,i,j and z c * ,i,j represents the jth measurement matrices,channel vector and the noise vector, respectively [15].…”

“…Furthermore, by exploiting the common statistical sparsity inherent in different multipath signals, we extend the BCS algorithms to a multi-task version for simultaneously reconstructing multiple signals, thus leading to MT-BCS [15,16].…”