Channel state information (CSI) estimation is part of the most fundamental problems in 5G wireless communication systems. In mobile scenarios, outdated CSI will have a serious negative impact on various adaptive transmission systems, resulting in system performance degradation. To obtain accurate CSI, it is crucial to predict CSI at future moments. In this paper, we propose an efficient channel prediction method in multipleinput multiple-output (MIMO) systems, which combines genetic programming (GP) with higher-order differential equation (HODE) modeling for prediction, named GPODE. In the first place, the variation of one-dimensional data is depicted by using higher-order differential, and the higher-order differential data is modeled by GP to obtain an explicit model. Then, a definite order condition is given for the modeling of HODE, and an effective prediction interval is given. In order to accommodate to the rapidly changing channel, the proposed method is improved by taking the rough prediction results of Autoregression (AR) model as a priori, i.e., Im-GPODE channel prediction method. Given the effective interval, an online framework is proposed for the prediction. To verify the validity of the proposed methods, We use the data generated by the Cluster Delay Line (CDL) channel model for validation. The results show that the proposed methods has higher accuracy than other traditional prediction methods.
Carbon dioxide capture, conversion, and utilization with
carbonic
anhydrase (CA) as a green and sustainable method still faces the challenges
of complex and costly purification, relatively low yield, and poor
catalytic performances and reusability. Herein, we proposed an all-in-one
strategy to solve these problems by self-assembling nanosized CA oligomeric
particles (nCAOPs) into micrometer-sized CA supraparticles (mCASPs)
with well-designed tags. The preparation of the purified mCASPs was
simple and effective by an easy scalable low-speed centrifugation
method with 78% activity recovery. The obtained mCASPs with a yield
of 870 mg/L was the reported highest yield of CAs. Interestingly,
mCASPs could serve as carrier-free immobilized CAs and retain over
90% original activity after 15 reuse cycles. More encouragingly, mCASPs
could redissolve and form nCAOPs, which had excellent catalytic performances.
The hydrated activity of nCAOPs was 1.05 times that of free CAs. Also,
the k
cat/K
m value was 1.74 times that of free CAs, and the half-life at 40 °C
was 3.83 times that of free CAs. Due to the simplicity of purification
and immobilization, high yield, and excellent enzymatic properties,
mCASPs and nCAOPs are considered novel bioactive materials, which
offer the feasibility of CAs for sustainable CO2 capture
to achieve the target of carbon neutrality.
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.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.