A Low-Complexity High-Accuracy AR Based Channel Prediction Method for Interference Alignment

Ozawa, Masayoshi; Ohtsuki, Tomoaki; Panahi, Fereidoun H.; Jiang, Wenjie; Takatori, Yasushi; Nagatsuma, Tadao

doi:10.1109/glocom.2018.8647292

Cited by 3 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1) The Differencing Sub-Module: Differencing CSI between adjacent time steps is considered to perform channel prediction in [11], [28]. It can be explained from time series analysis that the non-stationary process can be gradually transformed to stationarity through differencing, improving the predictability of time series [46].…”

Section: B the Differencing-attention Modulementioning

confidence: 99%

“…We compare the proposed STNN with four benchmarks, the outdated, linear regression, S times channel difference based forward AR (SDFAR) [11] and forward AR (FAR) [11]. The outdated method uses the last CSI for any time in the future.…”

Section: A Experimental Settingmentioning

confidence: 99%

“…The conventional channel prediction methods can be mainly divided into two categories, the parametric radio channel (PRC) model based methods [8], [9] and the autoregressive (AR) model based methods [10], [11]. However, both of them are based on the theoretical channel models and strict assumptions, which are quasi-static and wide-sense stationary (WSS), respectively.…”

Section: Introductionmentioning

confidence: 99%

“…However, both of them are based on the theoretical channel models and strict assumptions, which are quasi-static and wide-sense stationary (WSS), respectively. In [11], inspired by employing the difference of the past CSI between adjacent times instead of the past CSI, two AR based channel prediction methods were proposed to suppress interference in MIMO systems. In [12], based on the first-order Taylor expansion (FIT), a FIT prediction method was designed to track CSI in time-varying massive MIMO systems.…”

Section: Introductionmentioning

confidence: 99%

“…In [12], based on the first-order Taylor expansion (FIT), a FIT prediction method was designed to track CSI in time-varying massive MIMO systems. Both [11] and [12] assumed different transceiver antennas are mutually uncorrelated. Similarly, every sub-carrier in massive MIMO/MIMO-OFDM systems was treated as a SISO flat-fading channel in [13] and [14].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

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

Liu

Wang

et al. 2022

IEEE Trans. Commun.

View full text Add to dashboard Cite

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.

show abstract

Section: B the Differencing-attention Modulementioning

confidence: 99%

Section: A Experimental Settingmentioning

confidence: 99%