Recent advancements in spaceborne GNSS missions have produced extensive global datasets, providing a robust basis for deep learning-based significant wave height (SWH) retrieval. While existing deep learning models predominantly utilize CYGNSS data with four-channel information, they often adopt single-channel inputs or simple channel concatenation without leveraging the benefits of cross-channel information interaction during training. To address this limitation, a novel spatial–channel attention-based network, namely SCAWaveNet, is proposed for SWH retrieval. Specifically, features from each channel of the DDMs are modeled as independent attention heads, enabling the fusion of spatial and channel-wise information. For auxiliary parameters, a lightweight attention mechanism is designed to assign weights along the spatial and channel dimensions. The final feature integrates both spatial and channel-level characteristics. Model performance is evaluated using four-channel CYGNSS data. Quantitative and qualitative experiments were conducted on CYGNSS-ERA5 test set, SCAWaveNet achieves an average RMSE of 0.438 m. Compared to state-of-the-art models, SCAWaveNet reduces RMSE by at least 3.52%. Furthermore, evaluations on WW3, Jason-3, and NDBC buoy data, as well as in wind speed, rainstorm, typhoon and noisy scenarios, further confirm the superiority of SCAWaveNet. The code is available at https://github.com/Clifx9908/SCAWaveNet.

Spaceborne Global Navigation Satellite System Reflectometry (GNSS-R) has been widely used in land scenarios, but there is limited research focused on Vegetation Optical Depth (VOD) retrieval. For this purpose, we propose a method for inverting VOD using a combination of spaceborne GNSS-R data and an ensemble machine learning algorithm. Integrating Bistatic Radar Cross Section (BRCS), Effective Scattering Area (ESA), CYGNSS variable parameters, and surface auxiliary parameters (i.e. soil moisture, surface temperature, and roughness coefficient) to construct an Extreme Randomized Tree (ET) ensemble machine learning model. The results show that the proposed ET model has a good correlation between VOD retrieval and SMAP VOD, with retrieval accuracy comparable to the Decision Tree (DT) model, but significantly better than the retrieval results of Adaptive Boosting (AdaBoost), Support Vector Machine (SVM), and Gradient Boosting Decision Trees (GBDT) models. The Root Mean Square Error (RMSE) has improved by 178.12%, 67.87%, and 85.26%, respectively, while the Mean Absolute Percentage Error (MAPE) has improved by 87.50%, 80.00%, and 92.31%, respectively. This indicates that the ET model is feasible for high-precision and high spatiotemporal resolution VOD retrieval on land.

This study utilizes L1B level data from reflected global navigation satellite system (GNSS) signals from the Cyclone GNSS (CYGNSS) mission to estimate sea surface significant wave height (SWH). The normalized bistatic radar cross Section (NBRCS), the leading edge slope (LES), the signal-to-noise ratio (SNR), and the delay-Doppler map average (DDMA) are used as the key variables for the SWH retrieval. Eight other parameters, including instrument gain and scatter area, are also utilized as auxiliary variables to enhance the SWH retrieval performance. A variety of multivariable regression models are investigated to clarify the relationship between the SWH and the variables by using the following five methods: stepwise linear regression, Gaussian support vector machine, artificial neural network, sparrow search algorithm–extreme learning machine, and bagging tree (BT). Results show that, among the five regression models developed, the BT model performs the best with the root mean square error (RMSE) of 0.48 m and the correlation coefficient (CC) of 0.82 when testing one million sets of data randomly selected, while the RMSE and CC of BT model are 0.44 m and 0.73 in the 4500 National Data Buoy Center (NDBC) buoy testing dataset. Meanwhile, the BT model also has the best generalization ability, which means that it performs well in practical applications. In addition, the impacts of different input variables, the size of the training dataset, and the sea surface wind speed are also investigated. These findings are anticipated to serve as helpful guides for creating future SWH retrieval algorithms that are more advanced.

Significant Wave Height (SWH) is a key parameter to characterize waves, which is typically used in sea state monitoring such as wave forecast to ensure ocean navigation safety. Satellite radar altimeter is probably the primary tool to obtain SWH information. However, it cannot be used for large-scale sea state monitoring unless many of theses satellites are deployed. In this article, we aim to study the potential of Global Navigation Satellite System (GNSS)-Reflectometry (GNSS-R) in SWH measurement based on spaceborne Delay-Doppler Maps (DDMs) data. First, 3 observables (i.e., Delay-Doppler Map Average (DDMA), leading edge slope (LES) of normalized integrated delay waveform (NIDW) (LES-NIDW), and trailing edge slope (TES) of NIDW (TES-NIDW) derived from the DDMs are introduced for SWH estimation. Then, an empirical SWH retrieval model is proposed based on three observables. Subsequently, ERA5 SWH is used as reference data to verify the performance of the proposed model. The experimental results show that the Root Mean Square Error (RMSE) and Correlation Coefficient (CC) estimated by SWH of the three observables are better than 0.54 m and 0.88 m, respectively. Among them, the estimation performance based on DDMA observable is the best, with RMSE and CC of 0.49 m and 0.89 m. This study shows the potential of spaceborne GNSS-R in SWH retrieval.

In this letter, we first propose two empirical models to retrieve significant wave height (SWH) using two GNSS Reflectometry (GNSS-R) observables derived from delay-Doppler map (DDM), namely DDM average (DDMA) and leading edge slope (LES). Then, we utilize minimum variance to establish a combined model to enhance the SWH estimation performance. Collocated ERA5 SWH data is utilized as the ground truth for the development and evaluation of the SWH models. The results show that the SWH estimates by the three models are highly consistent with ERA5 SWH data, with a root mean square error (RMSE) less than 0.502 m and a correlation coefficient (CC) higher than 0.88. In particular, the combined model has significantly smaller RMSE of 0.428 m and larger CC of 0.91; and compared with the combined model based on weighted average (WA) method and LES observable model based on integral delay waveform, the RMSE is improved by 20.15 % and 14.74 %, respectively. The performance of spaceborne GNSS-R SWH retrieval can be greatly enhanced by the construction of integrated model, as demonstrated by this letter.

A geophysical model function (GMF) for significant wave height (SWH) retrieval is developed based on the spaceborne Global Navigation Satellite System Reflectometry (GNSS-R) data measured by the Cyclone GNSS (CYGNSS) satellites. The spreading characteristics of delay-Doppler maps (DDMs) generated by receivers onboard satellites are affected by the surface roughness, which is closely related to the SWH. Four GNSS-R observables [i.e., leading edge slope (LES) of normalized integrated delay waveform (NIDW), LES of normalized central delay waveform (NCDW), trailing edge slope (TES) of NCDW, leading edge waveform summation (LEWS) of NCDW] derived from DDM are first used in this letter to retrieve SWH. Collocated ERA5 SWH data are used as the ground truth to develop and evaluate the SWH models based on the four GNSS-R observables. The results show that there is high consistency between the SWH estimates and the ground truth, with a correlation coefficient of 0.88 and a root mean square error (RMSE) of 0.503 m. This letter demonstrates the feasibility of the spaceborne GNSS-R in SWH retrieval.

This paper focuses on improving indoor Wi-Fi localization by mitigating the effect of fluctuation of received signal strength indication (RSSI). The RSSI data collected at each reference point is first transformed through translation and scaling. The BP (Back Propagation) neural network is then used to construct the ranging model using the transformed RSSI to determine the distances between the target point and each reference point. A genetic algorithm (GA) is developed to optimize the initial values of weights and biases of the BP neural network. For convenience, our proposed ranging model is denoted as GTBPD. A new localization algorithm is then proposed, which uses the GTBPD model and the sequential quadratic programming (SQP, an iterative nonlinear optimization algorithm), and the algorithm is denoted as GTBPD-LSQP for simplicity. Experiments were conducted in three areas of two different teaching buildings with complex environments. The performance of the proposed GTBPD-LSQP algorithm is evaluated and compared with four existing algorithms. The experimental results show that our proposed GTBPD-LSQP algorithm achieves significantly higher location accuracy than the four existing algorithms.

Rainfall detection (RD) and rainfall intensity (RI) retrieval are hot topics in the field of ocean remote sensing (RS). In the past, the sea surface RD and RI retrieval were usually based on X-band ocean radar image data. In this study, we aim to investigate the potential of global navigation satellite system-reflectometry (GNSS-R) for sea surface RD and RI retrieval based on delay Doppler maps (DDMs) data collected by the cyclone GNSS (CYGNSS) mission. First, the block-matching and 3-D filtering (BM3D) algorithm is proposed to improve the quality of DDM data. In addition, 12 GNSS-R observables derived from DDM are calculated, and an RD method based on the threshold of 12 GNSS-R observables is proposed based on the probability density function (PDF). When rainfall DDM data are detected, these data are used to develop and verify the sea surface RI retrieval model. The integrated multisatellite retrievals of global precipitation measurements (GPM-IMERG) data product are used as reference data to evaluate the performance of RD and RI retrieval model. The experimental results show that under very low wind speed (<5 m/s), the proposed trailing edge waveform summation of normalized integral delay waveform (TEWS-NIDW), TEWS of normalized center delay waveform (TEWS-NCDW), and TEWS of differential delay waveform (TEWS-DDW) observables are the best for RD, and the probability of detection of rainfall (PDr) is better than 75%. In the aspect of model retrieval performance, the root mean square error (RMSE) of the 12 observables is less than 4.66 mm/hr. Among them, the model based on TEWS-NCDW observables has the highest accuracy, better than 3.74 mm/hr.

This paper mainly studies the inversion of sea surface wind speed based on spaceborne Global Navigation Satellite System-Reflectometry (GNSS-R) data. Delayed Doppler Map (DDM) is one of the most important data acquired by GNSS-R receivers. We make use of four DDM-based observations (DDM average (DDMA), signal-to-noise ratio (SNR), leading edge slope (LES) of Integrated Delay Waveform (IDW) and Normalized Bistatic Radar Cross Section (NBRCS)) to retrieve sea surface wind speed. We first propose a data filtering method based on the LES of IDW for data quality control. This method selects high -quality DDM data by adjusting LES threshold. Then, four wind speed inversion models are developed by using the four observations calculated from the clean DDM data. On this basis, five more wind speed inversion models are developed by using five weighted combinations of the four individual models. The nine models are tested and the results show that the wind speed estimation accuracy of the combined DDMA+NBRCS+SNR+LES model is respectively improved by 23.53%, 13.90%, 25.06% and 14.70% compared with the individual models of DDMA, SNR, LES, and NBRCS in the wind speed range of 0–20 m/s. Also, compared with dual combinations (DDMA+SNR, DDMA+NBRCS, and SNR+NBRCS), the accuracy of triple combination of DDMA+NBRCS+SNR is improved by 3.45%, 11.38%, 9.14% and 1.46%, respectively.