Spatio-Temporal Neural Network for Channel Prediction in Massive MIMO-OFDM Systems

Abstract

In massive multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems, a challenging problem is how to predict channel state information (CSI) (i.e., channel prediction) accurately in mobility scenarios. However, a practical obstacle is caused by CSI non-stationary and nonlinear dynamics in temporal domain. In this paper, we propose a spatio-temporal neural network (STNN) to achieve better performance by carefully taking into account the spatiotemporal characteristics of CSI. Specifically, STNN uses its encoder and decoder modules to capture the spatial correlation and temporal dependence of CSI. Further, the differencingattention module is designed to deal with the non-stationary and nonlinear temporal dynamics and realize adaptive feature refinement for more accurate multi-step prediction. Additionally, an advanced training scheme is adopted to reduce the discrepancy between STNN training and testing. Evaluated on a realistic channel model with enhanced mobility and spherical waves, experimental results show that STNN can effectively improve the accuracy of prediction and perform well with respect to different signal to noise ratios (SNRs). Visualization and testing for unit root illustrate STNN is able to learn CSI time-varying patterns by alleviating series non-stationarity.

Risk-Aware Antenna Selection for Multiuser Massive MIMO Under Incomplete CSI

“…This approach works mainly because the real propagation environment is often time-varying and temporally correlated due to Doppler effects [20]- [22]. Based on this, researchers have developed different channel prediction algorithms in recent years [23]- [25]. In [23], the authors proposed a spatio-temporal autoregressive method for the prediction of the high mobility channel, where the prediction was performed utilizing the temporal correlation in the angle-delay domain.…”

“…The proposed method employed the convolutional neural network (CNN) and the long-short-term memory network (LSTM), which allows a multistep prediction of the channels. Similarly in [25], the authors predicted channel states by utilizing the spatio-temporal characteristics of CSI and a combination of CNN and convolutional LSTM. Although showing improved AS performance, these channel prediction methods are based on the historic complete CSI measurements [23]- [25].…”

“…Similarly in [25], the authors predicted channel states by utilizing the spatio-temporal characteristics of CSI and a combination of CNN and convolutional LSTM. Although showing improved AS performance, these channel prediction methods are based on the historic complete CSI measurements [23]- [25]. It is noted that obtaining complete CSI is spectral-inefficient in the considered massive MIMO systems, especially when the number of antennas significantly exceeds the number of RF chains.…”

Scenario-Aware Learning Approaches to Adaptive Channel Estimation

“…However, the algorithms in these systems must be manually adjusted for each scenario, with the channel estimation module of the receiver being a critical component. The accuracy of CSI estimation is essential for the performance of other modules in the system such as signal detection [5], [6], CSI feedback [7], [8], CSI prediction [9], [10] and beam-forming [11]- [13]. This paper focuses on this scenario-adaptive channel estimation problem, where CSI estimation is tailored to the different scenarios, to address the challenge of varying characteristics of CSIs in different scenarios.…”

Dynamic channel pruning via activation gates

“…Therefore, a few physics-informed deep learningbased [21,22] works have considered the unique characteristics of CSI, referred to as hybrid methods [23,24] . For instance, a 3-Dimensional (3D) complex CNN-based predictor [23] is utilized to capture temporal and spatial correlations based on angle-delay domain representation.…”