Received Signal Strength Indicator Data Research Articles

A Combined Filtering Method for ZigBee Indoor Distance Measurement.

Indoor distance measurement technology utilizing Zigbee's Received Signal Strength Indication (RSSI) offers cost-effective and energy-efficient advantages, making it widely adopted for indoor distance measurement applications. However, challenges such as multipath effects, signal attenuation, and signal blockage often degrade the accuracy of distance measurements. Addressing these issues, this study proposes a combined filtering approach integrating Kalman filtering, Dixon's Q-test, Gaussian filtering, and mean filtering. Initially, the method evaluates Zigbee's transmission power, channel, and other parameters, analyzing their impact on RSSI values. Subsequently, it fits a signal propagation loss model based on actual measured data to understand the filtering algorithm's effect on distance measurement error. Experimental results demonstrate that the proposed method effectively improves the conversion relationship between RSSI and distance. The average distance measurement error, approximately 0.46 m, substantially outperforms errors derived from raw RSSI data. Consequently, this method offers enhanced distance measurement accuracy, making it particularly suitable for indoor positioning applications.

Improving ventilation classification in under-actuated zones: a k-nearest neighbor and data preprocessing approach

This study investigates the use of k-nearest neighbors (k-NN) for classifying occupant positions in under-actuated zones, aiming to enhance ventilation control. The focus is on evaluating different data preprocessing techniques, particularly cumulative moving average (CMA), Kalman filtering (KF), and their combination, to boost the k-NN model's reliability and accuracy. The research uses received signal strength indicator (RSSI) data in a controlled setting. The methodology involves dividing the dataset into training and testing subsets and using root mean squared error (RMSE) to determine the best k value for model validation. The study performs a comparative analysis of the k-NN model's performance with both original and preprocessed RSSI data, focusing on metrics such as accuracy, precision, recall, F1-score, and RMSE. The findings emphasize the significant impact of the combined CMA-KF preprocessing technique in improving the model's accuracy and reliability. Specifically, this approach achieved an accuracy of 98.58%. The RMSE values are particularly noteworthy, exhibiting a perfect fit (RMSE of 0) for training data and a remarkably low RMSE of 0.119 for testing data, confirming the model's high accuracy and predictive capability.

Improved-RSSI-based indoor localization by using pseudo-linear solution with machine learning algorithms

With the rapid advancement of the Internet of Things and the popularization of mobile Internet-based applications, the location-based service (LBS) has attracted much attention from commercial developers and researchers. Received signal strength indicator (RSSI)-based indoor localization technology has irreplaceable advantages for many LBS applications. However, due to multipath fading, noise, and the limited dynamic range of the RSSI measurements, precise localization based on a path-loss model and multiliterate becomes highly challenging. Therefore, this study proposes a machine learning (ML)-based improved RSSI-based indoor localization approach in which RSSI data is first augmented and then classified using ML algorithms. In addition, we implement an experimental testbed to collect the RSSI value based on Wi-Fi using various reference and target nodes. The received RSSI measurements undergo pre-processing using pseudo-linear solution techniques for closed-form solutions, approximating the original system of nonlinear RSSI measurement equations with a system of linear equations. Finally, the RSSI measurement are trained using ML models such as linear regression, polynomial regression, support vector regression, random forest regression, and decision tree regression. Consequently, the experimental results express in terms of root mean square error and coefficient of determinant compared with various machine learning models with hyper-parameter tuning.

Multi-Dimensional Wi-Fi Received Signal Strength Indicator Data Augmentation Based on Multi-Output Gaussian Process for Large-Scale Indoor Localization

Location fingerprinting using Received Signal Strength Indicators (RSSIs) has become a popular technique for indoor localization due to its use of existing Wi-Fi infrastructure and Wi-Fi-enabled devices. Artificial intelligence/machine learning techniques such as Deep Neural Networks (DNNs) have been adopted to make location fingerprinting more accurate and reliable for large-scale indoor localization applications. However, the success of DNNs for indoor localization depends on the availability of a large amount of pre-processed and labeled data for training, the collection of which could be time-consuming in large-scale indoor environments and even challenging during a pandemic situation like COVID-19. To address these issues in data collection, we investigate multi-dimensional RSSI data augmentation based on the Multi-Output Gaussian Process (MOGP), which, unlike the Single-Output Gaussian Process (SOGP), can exploit the correlation among the RSSIs from multiple access points in a single floor, neighboring floors, or a single building by collectively processing them. The feasibility of MOGP-based multi-dimensional RSSI data augmentation is demonstrated through experiments using the hierarchical indoor localization model based on a Recurrent Neural Network (RNN)—i.e., one of the state-of-the-art multi-building and multi-floor localization models—and the publicly available UJIIndoorLoc multi-building and multi-floor indoor localization database. The RNN model trained with the UJIIndoorLoc database augmented with the augmentation mode of “by a single building”, where an MOGP model is fitted based on the entire RSSI data of a building, outperforms the other two augmentation modes and results in the three-dimensional localization error of 8.42 m.

Research Progress of Wireless Positioning Methods Based on RSSI

Location-based services are now playing an integral role in the development of emerging industries, such as the Internet of Things, artificial intelligence and smart cities. Although GPS, Beidou and other satellite positioning technologies are becoming more and more mature, they still have certain limitations. In order to meet the needs of high-precision positioning, wireless positioning is proposed as a supplementary technology to satellite positioning, in which the Received Signal Strength Indication (RSSI) is one of the most popular positioning methods. In this paper, the application scenarios, evaluation methods and related localization methods of wireless positioning based on RSSI are studied. Secondly, the relevant optimization methods are analyzed and compared from different angles, and the methods of RSSI data acquisition are described. Finally, the existing problems and future development trends in RSSI positioning methods are expounded, which has certain reference significance for further research on RSSI localization.

RSSI 데이터 기반 머신러닝 실내 측위 연구

With the recent development of science and technology, localization technology has made it possible to provide customized mobile services based on real-time location information to users through smartphones and Internet of Things devices. In particular, GPS-based real-time localization is commonly used. However, unlike outdoors, not only does the GPS reception rate decrease significantly indoors, but it is also difficult to accurately measure the user's indoor location using only GPS latitude and longitude information. In this study, we propose to measure the user's indoor location using RSSI (received signal strength indicator). For this purpose, an indoor localization study was conducted by applying machine learning classifiers, SVM, decision tree, ExtraTrees, random forest, and KNN to RSSI data. Among the machine learning classifiers, random forest exhibited the best performance. Therefore, we applied random forest-based RFE to extract features from RSSI data. We confirmed that even with a smaller amount of data, it was possible to achieve more accurate indoor positioning. In addition, it was confirmed through regression analysis that latitude and longitude can be estimated from RSSI data, so that indoor as well as outdoor locations can be estimated.

Indoor Localization System Based on RSSI-APIT Algorithm.

An indoor localization system based on the RSSI-APIT algorithm is designed in this study. Integrated RSSI (received signal strength indication) and non-ranging APIT (approximate perfect point-in-triangulation test) localization methods are fused with machine learning in order to improve the accuracy of the indoor localization system. The system focuses on the improvement of preprocessing and localization algorithms. The primary objective of the system is to enhance the preprocessing of the acquired RSSI data and optimize the localization algorithm in order to enhance the precision of the coordinates in the indoor localization system. In order to mitigate the issue of significant fluctuations in RSSI, a technique including the integration of Gaussian filtering and an artificial neural network (ANN) is employed. This approach aims to preprocess the acquired RSSI data, thus reducing the impact of multipath effects. In order to address the issue of low localization accuracy encountered by the conventional APIT localization algorithm during wide-area localization, the RSSI ranging function is incorporated into the APIT localization algorithm. This addition serves to further narrow down the localization area. Consequently, the resulting localization algorithm is referred to as the RSSI-APIT positioning algorithm. Experimental results have demonstrated the successful reduction of inherent localization errors within the system by employing the RSSI-APIT positioning algorithm. The present study aims to investigate the impact of the localization scene and the number of anchors on the RSSI-APIT localization algorithm, with the objective of enhancing the performance of the indoor localization system. The conducted experiments demonstrated that the enhanced system exhibits several advantages. Firstly, it successfully decreased the frequency of anchor calls, resulting in a reduction in the overall operating cost of the system. Additionally, it effectively enhanced the accuracy and stability of the system's localization capabilities. In a complex environment of 100 m2 in size, compared with the traditional trilateral localization method and the APIT localization algorithm, the RSSI-APIT localization algorithm reduced the localization error by about 2.9 m and 1.8 m, respectively, and the overall error was controlled within 1.55 m.

Intelligent Reflecting Surface Enabled Fingerprinting-Based Localization With Deep Reinforcement Learning

Intelligent reflecting surface (IRS) is considered a promising solution to manipulate the radio frequency transmission environment in the sixth-generation (6 G) wireless systems. However, little attention was received by IRS-aided localization techniques. Among range-free wireless localization strategies, received signal strength indicator (RSSI) fingerprinting-based technique is preferred since it can be easily accessed. Inspired by these and the tremendous success of deep reinforcement learning (DRL), we propose an IRS-enabled fingerprinting-based localization methodology with the aid of DRL. Specifically, we firstly propose an IRS-enabled fingerprinting-based localization system. In this system, RSSI lists are created by periodic IRS configurations and pre-collected as database. When a request of localization from a receiver is sent to the server, the database is compared with the online-measured RSSI data to identify the best receiver position estimate using the nearest neighbor algorithm. In addition, we develop a DRL-based IRS configuration selector to identify the most qualified IRS configurations so as to minimize the localization error. We also propose a communication protocol for the operation of the proposed localization methodology. Extensive simulation under different circumstances have been conducted and the results indicate that the localization accuracy scales with the number of IRS configurations. With the aid of DRL, the localization accuracy is further boosted by more than 40% as compared with previous work.

Improved RSSI Indoor Localization in IoT Systems with Machine Learning Algorithms

Recent developments in machine learning algorithms are playing a significant role in wireless communication and Internet of Things (IoT) systems. Location-based Internet of Things services (LBIoTS) are considered one of the primary applications among those IoT applications. The key information involved in LBIoTS is finding an object’s geographical location. The Global Positioning System (GPS) technique does not perform better in indoor environments due to multipath. Numerous methods have been investigated for indoor localization scenarios. However, the precise location estimation of a moving object in such an application is challenging due to the high signal fluctuations. Therefore, this paper presents machine learning algorithms to estimate the object’s location based on the Received Signal Strength Indicator (RSSI) values collected through Bluetooth low-energy (BLE)-based nodes. In this experiment, we utilize a publicly available RSSI dataset. The RSSI data are collected from different BLE ibeacon nodes installed in a complex indoor environment with labels. Then, the RSSI data are linearized using the weighted least-squares method and filtered using moving average filters. Moreover, machine learning algorithms are used for training and testing the dataset to estimate the precise location of the objects. All the proposed algorithms were tested and evaluated under their different hyperparameters. The tested models provided approximately 85% accuracy for KNN, 84% for SVM and 76% accuracy in FFNN.

RSSI based Device-free Human Identification

Researchers have explored many methods and techniques for human detection and identification in diverse contexts. One such approach is studying variations of Radio Frequency (RF) signals (e.g., Received Signal Strength Indicator (RSSI)). RSSI based techniques have been widely used in human detection, but not for unique human identification due to its noisy nature. This research focused on investigating the possibilities and limitations of device-free human identification using Wi-Fi RSSI data with machine learning-based classification techniques. To inspect the characteristics of Wi-Fi RSSI data, the authors have conducted statistical analysis techniques. A Kalman filter was applied to minimize the noise in Wi-Fi RSSI data, followed by a feature extraction process. Furthermore, the authors have conducted multiple research experiments in different configurations of receivers and participant numbers. The experimental results show that the human identification accuracy level increases with the number of receivers used for the data collection. Moreover, the authors have identified that human identification accuracy can be further improved by leveraging proper noise reduction methods and feature extraction processes. For a Kalman filter applied and feature-extracted Wi-Fi RSSI dataset of 20 people, the Support Vector Machine (SVM) - Radial Kernel model recorded the highest average accuracy of 99.58%.

Novel integrated matching algorithm using a deep learning algorithm for Wi-Fi fingerprint-positioning technique in the indoors-IoT era.

The Internet-of-Things (IoT) has been used with greater frequency to track peoples' daily activities, particularly those conducted indoors. Wi-Fi technology has been also been used as an alternative to global navigation satellite system (GNSS) technologies to track indoor activities. The received signal strength indicator (RSSI) is widely used to assist in the positioning of Wi-Fi signals. However, the RSSI-based technique suffers from multipath, non-line-of-sight (NLOS) problems and the fluctuation of RSSI measurements via Wi-Fi chipsets. One of the most well-known RSSI-based approaches is to apply the fingerprinting method to do the positioning. However, the fingerprinting-based form has an additional problem due to the lack of RSSI data samples, specifically in harsh area with a huge number of classes or reference points (RPs) and an unstable matching process algorithm. To mitigate the problems of the RSSI-based fingerprinting approach, this research proposes a novel matching process algorithm called Norm_MSATE_LSTM. We first performed the augmentation process to increase the RSSI data records via the Mean Stander deviation Augmentation TEchnique (MSATE). The RSSI records were normalized (norm), and the long short-term memory (LSTM) technique was applied to estimate the correct positions. Finally, the proposed matching algorithm was compared with the stand-alone matching algorithms, including the weighted k-nearest neighbors (WkNN) and LSTM. The results obtained from the experiments and the simulated experiments using OMNeT++ show that the proposed matching algorithm may improve positioning accuracy by 33.1% and 57.5% when only augmentation and augmentation with normalization are applied, respectively.

Radio map generation approaches for an RSSI-based indoor positioning system

The use of Radio map fingerprinting, which relies on a received signal strength indicator (RSSI), is a popular indoor positioning method that offers high accuracy and cost-effective deployment. However, the generation of an RSSI radio map requires significant time and effort. This paper presents three methods for generating radio maps, with the aim of reducing the time required. The first method involves a dedicated mobile application that collects experimental RSSI data, while the second method uses biharmonic spline interpolation (BSI) to expand a prerecorded experimental radio map. The third method, the Wireless InSite simulator-based method, generates a fully simulated radio map. All three methods are studied in detail to evaluate their effectiveness in reducing the time required for radio map generation. Location estimation is then carried out based on the prerecorded radio maps. The combined method presented in this study increases the efficiency of building indoor positioning mobile applications. Experiments using combined real and simulated datasets collected at An Najah National University and University of Dubai campuses demonstrate that the model outperforms similar methods, improving the localization accuracy to approximately 0.45 meters. This level of accuracy is suitable for a variety of location-based applications, including critical ones such as evacuating people from buildings during emergencies.

BLE beacon-based floor detection for mobile robots in a multi-floor automation laboratory

Abstract As an important task of multi-floor localization, floor detection has elicited great attention. Wireless infrastructures like Wi-Fi and Bluetooth Low Energy (BLE) play important roles in floor detection. However, most floor detection research tends to focus on data modelling but pays little attention to the data collection system, which is the basis of wireless infrastructure-based floor detection. In fact, the floor detection task can be greatly simplified with proper data collection system design. In this paper, a floor detection solution is developed in a multi-floor life science automation lab. A data collection system consisting of BLE beacons, a receiver node and an Internet of Things (IoT) cloud is provided. The features of the BLE beacon under different settings are evaluated in detail. A mean filter is designed to deal with the fluctuation of the received signal strength indicator data. A simple floor detection method without a training process was implemented and evaluated in more than 100 floor detection tests. The time delay and floor detection accuracy under different settings are discussed. Finally, floor detection is evaluated on the H20 multi-floor transportation robot. Two sensor nodes are installed on the robot at different heights. The floor detection performance with different installation heights is discussed. The experimental results indicate that the proposed floor detection method provides floor detection accuracy of 0.9877 to 1 with a time delay of 5 s.

Multi-Boundary Empirical Path Loss Model for 433 MHz WSN in Agriculture Areas Using Fuzzy Linear Regression.

Path loss models are essential tools for estimating expected large-scale signal fading in a specific propagation environment during wireless sensor network (WSN) design and optimization. However, variations in the environment may result in prediction errors due to uncertainty caused by vegetation growth, random obstruction or climate change. This study explores the capability of multi-boundary fuzzy linear regression (MBFLR) to establish uncertainty relationships between related variables for path loss predictions of WSN in agricultural farming. Measurement campaigns along various routes in an agricultural area are conducted to obtain terrain profile data and path losses of radio signals transmitted at 433 MHz. Proposed models are fitted using measured data with "initial membership level" (μAI). The boundaries are extended to cover the uncertainty of the received signal strength indicator (RSSI) and distance relationship. The uncertainty not captured in normal measurement datasets between transmitter and receiving nodes (e.g., tall grass, weed, and moving humans and/or animals) may cause low-quality signal or disconnectivity. The results show the possibility of RSSI data in MBFLR supported at an μAI of 0.4 with root mean square error (RMSE) of 0.8, 1.2, and 2.6 for short grass, tall grass, and people motion, respectively. Breakpoint optimization helps provide prediction accuracy when uncertainty occurs. The proposed model determines the suitable coverage for acceptable signal quality in all environmental situations.

Hybrid indoor positioning for smart homes using WiFi and Bluetooth low energy technologies

In indoor positioning problems, GPS technology used in outdoor positioning needs to be improved due to the characteristic features of wireless signals. There currently needs to be a generally accepted standard method for indoor positioning. In this study, an ecosystem consisting of Beacon devices, Bluetooth intelligent devices, and Wi-Fi access points has been created to propose an effective indoor location determination method by using Wi-Fi and BLE technologies in a hybrid way. First, RSSI (Received Signal Strength Indicator) data were collected using the fingerprint method. Then, Kalman Filter and Savitzky Golay Filter are used in a hybrid manner to reduce the noise on the obtained signal data and make it more stable. In the first part, using the collected data from Wi-Fi and Beacon devices, the Non-linear least squares method (NLLS), including Levenberg-Marquardt (LM), is used for indoor tracking. In the second part, a fingerprinting-based approach is tested. K Nearest Neighbor (KNN) and Support Vector Machine (SVM) algorithms estimate the area where the client is located. Each algorithm’s accuracy rate are calculated on different training and test data and presented.

Synthesis of a Small Fingerprint Database through a Deep Generative Model for Indoor Localisation

In deep learning (DL), the deep generative model is helpful for data augmentation objectives to tackle the lack of datasets that have a significant impact on learning performance. Data augmentation or synthesis is expected to solve the issue in a small/sparse database. The problem of databasing also exists in the fingerprint-based indoor localisation system. The dense offline fingerprint database must be constructed with the accuracy requirement. However, this will affect the high cost, massive laborious work, and increase the complexity of the system. Therefore, this paper proposes to address these issues by generating synthetic data via a deep generative model. The generative adversarial network (GAN) is selected to generate the synthetic fingerprint database for indoor localisation. Our database consideration consists of power-based parameters, i.e., the received signal strength indicator (RSSI) from Wi-Fi devices obtained from the actual measurement campaign. Some of the literature mainly discusses how GAN works in a vast and complex dataset. Here, we consider applying GAN in a relatively small dataset and for a simple setup. Our results show that by only using the 20 % fraction of actual RSSI data combined with the synthetic RSSI, the accuracy validation performance is slightly higher than when using all actual data usage. Moreover, in only 60 % of actual data usage and in combination with 625 samples of synthetic data, the accuracy performance is improved to 0.73 (1.37 times higher than the use of all actual data, 0.53). Thus, this result proves that the challenges of offline fingerprint databases can be alleviated by data synthesis through GAN by using only a small dataset.

An Optimization Method Combining RSSI and PDR Data to Estimate Distance between Smart Devices for COVID-19 Contact Tracing

Distance estimation methods arise in many applications, such as indoor positioning and COVID-19 contact tracing. The received signal strength indicator (RSSI) is favored in distance estimation. However, the accuracy is not satisfactory due to the signal fluctuation. Besides, the RSSI-only method has a large-ranging error because it uses fixed parameters of the path loss model. Here, we propose an optimization method combining RSSI and pedestrian dead reckoning (PDR) data to estimate the distance between smart devices. The PDR may provide high accuracy of walking distance and direction. Moreover, the parameters of the path loss model are optimized to dynamically fit the complex electromagnetic environment. The proposed method is evaluated in outdoor and indoor environments and compared with the RSSI-only method. The results show that the mean absolute error is reduced up to 0.51 m and 1.02 m, with an improvement of 10.60% and 64.55% for outdoor and indoor environments, respectively, compared with the RSSI-only method. Consequently, the proposed optimization method has better accuracy of distance estimation than the RSSI-only method, and its feasibility is demonstrated through real-world evaluations.

Indoor 2D Positioning Method for Mobile Robots Based on the Fusion of RSSI and Magnetometer Fingerprints.

Received signal strength indicator (RSSI)-based fingerprinting is a widely used technique for indoor localization, but these methods suffer from high error rates due to various reflections, interferences, and noises. The use of disturbances in the magnetic field in indoor localization methods has gained increasing attention in recent years, since this technology provides stable measurements with low random fluctuations. In this paper, a novel fingerprinting-based indoor 2D positioning method, which utilizes the fusion of RSSI and magnetometer measurements, is proposed for mobile robots. The method applies multilayer perceptron (MLP) feedforward neural networks to determine the 2D position, based on both the magnetometer data and the RSSI values measured between the mobile unit and anchor nodes. The magnetic field strength is measured on the mobile node, and it provides information about the disturbance levels in the given position. The proposed method is validated using data collected in two realistic indoor scenarios with multiple static objects. The magnetic field measurements are examined in three different combinations, i.e., the measurements of the three sensor axes are tested together, the magnetic field magnitude is used alone, and the Z-axis-based measurements are used together with the magnitude in the X-Y plane. The obtained results show that significant improvement can be achieved by fusing the two data types in scenarios where the magnetic field has high variance. The achieved results show that the improvement can be above 35% compared to results obtained by utilizing only RSSI or magnetic sensor data.

Non-invasive Frozen Meat Monitoring System Using UHF RFID Tag Antenna-Based Sensing and RSSI

The conditions of frozen meat products must be closely monitored in cold chain logistics (CCL) to maintain their quality and safety. Sensing and monitoring meat products are currently invasive, costly, and lacking tracing capabilities. Therefore, developing a wireless, passive, and cost-effective sensing system capable of tracking and monitoring remains challenging. This work investigates the UHF RFID system performing antenna-based sensing for monitoring frozen meat using the received signal strength indicator (RSSI) data. A commercial off-the-shelf (COTS) UHF RFID reader is programmed through a single-board computer to acquire the RSSI data throughout the RFID 902-926 MHz band. In the experiments, RSSI data from an RFID inlay tag affixed to a defrosted frozen meat sample is acquired for approximately 20 minutes. Then, the RSSI data is recorded periodically during the changes in the sample condition. The experimental results signify that the RSSI data have monotonic relationships with the temperature and hardness of the meat sample. The three-degree polynomial regression models are constructed to show the non-linear relationships between the RSSI and the frozen meat condition. During defrosting, the RSSI lowers as the meat temperature rises and the hardness reduces. Therefore, antenna-based sensing employing the RFID RSSI data can detect changes in frozen meat temperature and hardness, allowing conditional fluctuations in the CCL to be monitored. This work paves the way for low-cost IoT-based sensing systems for improving food safety in cold chain applications.

Research on site selection of agricultural internet of things nodes based on rapid terrain sampling

With the development of Internet of Things (IoT) technology, modern agriculture is moving in the direction of 4.0. The Agricultural IoT is inseparable from wireless communication. However, in traditional agricultural IoT router and gateway site selection, the influence of the actual terrain on transmission loss is not considered, which results in node power wastage and increased maintenance costs. Based on a multi-sensor fusion algorithm, a fast terrain sampler is designed in this study to collect point-cloud data of the experimental site terrain. A reasonable objective function is then designed under the premise of consideration of the electromagnetic wave free-space and diffraction losses, and the locations of the routers and gateway are optimized based on k-means and particle swarm optimization (PSO) algorithm. Simulations show that the running time of the PSO algorithm is very sensitive to the changes in the execution parameters, and the improved PSO algorithm converges faster than the genetic algorithm (GA) in all three initialization methods. After collecting field terrain data, five interpolation methods were compared, and the nearest-neighbor algorithm is used to obtain the terrain model. On-site collection of received signal-strength indication (RSSI) shows that the communication quality of the optimal point selected via this algorithm is significantly higher than those of nearby points. At the same time, it is proved that the RSSI data has serious discontinuities, so the traditional gradient descent method is not suitable for solving the objective function in this work. Therefore, the algorithm used herein is of great significance for the site selection of agricultural Internet of Things nodes. However, since there are many ways to calculate the diffraction loss, the objective function of this work still needs more correction studies. The tool proposed in this work can obtain a 3D model of farmland faster than traditional surveying and mapping methods. With future research, the 3D model may be applied to soil moisture analysis, rainfall and moisture direction inversion, and plant light exposure prediction.