It is critically meaningful to accurately predict the ionospheric F2 layer critical frequency (foF2), which greatly limits efficiency of communications, radar, and navigation systems. This paper introduced entropy weight method develop combination prediction model (CPM) for long-term foF2 at Darwin (12.4° S, 131.5° E) in Australia. The coefficient each individual CPM determined by using after completing simulation calibration period. We analyzed two sets data validate used this study: One...

To further improve the short-term forecasting ability of critical frequency ionosphere F2 layer (foF2), a sample entropy optimized deep learning long-short-term memory (LSTM) model based on improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) is proposed. The ICEEMDAN-LSTM uses foF2 hour-level time series data Dourbes station from 2009 to 2019 for training and verification realizes single-step high-precision forecast. Through statistical analysis observation...

The latest research indicates that there are time-lag effects between the normalized difference vegetation index (NDVI) and precipitation variation. It is well known time-lags different from region to region, for NDVI itself correlated precipitation. In arid semi-arid grasslands, annual has proved not only be highly dependent on of concurrent year previous years, but also years. This paper proposes a method using recurrent neural network (RNN) capture both with respect itself, To...

Abstract The highly nonlinear variation of the ionospheric F 2 layer critical frequency ( f o ) greatly limits efficiency communications, radar, and navigation systems that employ high‐frequency radio waves. This paper proposes an effective method to predict using extreme learning machine (ELM). Compared with previous neural network model based on feedforward algorithm, ELM offers advantages faster training speed less manual intervention. is trained daily hourly values at Darwin (12.4°S,...

Normalized difference vegetation index (NDVI) has been used to conduct important research on plant growth and productivity. In this paper, a new approach predict NDVI based precipitation in the grass-growing season for arid semi-arid grassland is proposed using time-delay neural network (TDNN). To intuitively know ability of TDNN learn relationship between NDVI, performance model compared with back propagation (BPNN) trained same data. The results indicate that works well precipitation....

A method for predicting the dynamic spatio-temporal variations of normalized difference vegetation index (NDVI) based on precipitation is proposed using combined nonlinear autoregressive with exogenous input (NARX) networks and artificial neural (ANNs). The validated by applying to predict NDVI Hulunbuir grassland located in Inner Mongolia, China. results show good predictive ability mean absolute percentage error 11.59%, 7.11 × 10−2 root square 8.06 10−2, respectively. approach presented...

Abstract A model based on modified temporal‐spatial reconstruction is proposed to improve the accuracy of predicting monthly median ionospheric critical frequency F 2 layer. This has three new characteristics. (1) The solar activity parameters 10.7‐cm radio flux and sunspot number are together introduced into temporal reconstruction. (2) Both geomagnetic dip its value chosen as features geographical spatial variation for (3) Harmonic functions used represent layer, which reflects seasonal...

To improve the accuracy of predictions and simplify difficulty with algorithm, a simplified empirical model is proposed in developing long-term predictive approach determining ionosphere’s F2-layer critical frequency (foF2). The main distinctive features introduced this are: (1) Its vertical incidence sounding data, which were obtained from 18 ionosonde stations east Asia between 1949 2017, used reconstructing verification; (2) use second-order polynomial triangle harmonic functions, instead...

Abstract The 16 years of normalized difference vegetation index (NDVI) and precipitation data are used to analyze the time‐lag effects growing‐season NDVI response at regional scales. This study focuses on arid semi‐arid Hulunbuir grassland dominated by perennials in northeast China. multi‐month examined using simple statistical approaches, which can detect two distinct time‐lags for four subregions with major land‐cover types. A “positive” effect is observed 1‐month (May current year)...

In order to predict the livestock carrying capacity in meadow steppe, a method using back propagation neural network is proposed based on meteorological data and remote-sensing of Normalised Difference Vegetation Index. method, was first utilised build behavioural model forecast precipitation during grass-growing season (June–July–August) from 1961 2015. Second, relationship between Index 2000 2015 modelled with help network. The prediction results demonstrate that network-based effective...

The characteristics of microwave power devices working in large signal state can be accurately characterized by X-parameters. However, the extraction X-parameter is inefficient. To improve this problem, an modeling method based on long short-term memory (LSTM) neural network proposed. It proved experiments that mean square error (MSE) LSTM model 0.1291. Therefore, X-parameters transistor represented established model.

This paper presents a monolithic resonant room-temperature CMOS terahertz (THz) thermal detector composed of strong metamaterial absorber and an optimized PTAT sensor. The overall design considerations, as well relatively better characterization results are demonstrated at 4.3THz for the natural atmospheric window. At present, 4.2THz quantum cascade laser (QCL) is used to obtain performances due accessible THz source. proposed achieves maximum responsivity 20.9V/W minimum noise equivalent...

Abstract Restoration of natural vegetation in arid and semi‐arid grasslands is facing severe challenges. The easy to lose their vitality, resulting the loss cover under high grazing pressure. To address this situation, paper proposes a novel method for accurately predicting pressure using nonlinear autoregressive with exogenous input (NARX) network based on remote sensing data normalized difference index (NDVI) precipitation. proposed uses NARX networks predict temporal variations NDVI...