Enhancing indoor positioning system accuracy using a Feedback- Enhanced Federated Kalman Filter (FE-FKF)

In this paper, the performance of a multi-sensor fusion technique, namely Federated Kalman Filter (FKF) is studied in the context of indoor positioning problem. Kalman filters having centralized and decentralized structures are widely used in outdoor positioning and navigation applications. Global Positioning System (GPS) is the most commonly used system for outdoor positionin/navigation, which cannot be used indoors due to the signal loss. In this study, a decentralized structure for FKF is applied in indoor positioning problem by taking its outdoor navigation performance into consideration. Simulations are performed with distance measurements, which are assumed to be calculated by using Received Signal Strength (RSS). Results gathered via different simulations are evaluated as promising for future studies.

In this paper, federated Kalman filter (FKF) is applied for indoor positioning. Position information that is multi-laterated from the distance information obtained using the received signal strengths collected from several access points are processed in a FKF to estimate the position of the target. Two approaches are presented to adjust the information-sharing coefficients of FKF using online measurements. The data collected on a test bed composed of four access points are used to assess and compare the performances of the proposed algorithms. It is shown that the estimation error can be improved considerably by adjusting the information-sharing coefficients online.

Indoor Positioning via WLAN Channel State Information and Machine Learning Classification Approaches

Techniques for indoor positioning systems (IPSs) can be categorised as range-based or range-free. Range-based methods rely on geometric mappings to approximate a location given the calculated distances or angles from multiple reference points. In contrast, range-free strategies utilise fingerprinting, wherein an acquired fingerprint data is compared to a pre-collected dataset to identify the best position estimate. Among these, fingerprinting of channel state information (CSI) is preferred over other information such as received signal strength indicator as the former can exploit the effect of multipath propagation and is robust against non-line-of-sight channels. CSI has the potential to achieve cm-level positioning accuracy with a single reference point only. In this study, the authors survey CSI fingerprinting-based indoor positioning and mobility tracking systems. The process of fingerprinting that includes site surveying and signal preprocessing is discussed in detail. They determine the potential challenges of such systems and propose remedies to improve positioning accuracy. In general, spatial diversity, such as multiple-input multiple-output antennas and wireless sensor networks, or frequency diversity (e.g. high subcarrier count, frequency hopping mechanism) are exploited to achieve high positioning resolution. Such IPS can also be enhanced via additional sensors or spatial graphs for motion detection and tracking.

Improving measurement accuracy of indoor positioning system of a Mecanum wheeled mobile robot using Monte Carlo - Latin hypercube sampling based machine learning algorithm

The indoor positioning system (IPS) has been attracting great attention from researchers, thanks to the rapid adoption of smartphone technologies. Although there are many IPS proposed in the past decade that claimed to have good performance, all of them use their own method to evaluate and compare the accuracy of the proposed solution. During the evaluation phase, the method of gathering ground truth data (original position) is often not well described. As such, it is very difficult for other researchers to reproduce the work and improve on the existing methods. In this paper, we proposed a simple to implement framework to facilitate the process of evaluating IPS accuracy. Under this framework, the IPS position coordinates and ground truth are sent to the server using REST protocol when the phone reads an event triggered from tags scan placed on a fix position. We evaluated an existing well-known IPS technique, the Pedestrian Dead Reckoning (PDR) technique using our IPS evaluation framework. From our experiments, we showed that in addition to measuring the accuracy of IPS, the proposed solution can also measure the IPS accuracy deviation over time. Instead of relying on precision and recall, the framework also includes visualization tool for researchers to observe the overall accuracy of an IPS.

Indoor Positioning System (IPS) is a technology that provides precise and accurate location information in indoor environments. Since Global Positioning system (GPS) relies on line of sight communications with a satellite transceiver, it does not work within buildings. To overcome this challenge, IPS technology could use RF based communication and locate humans and objects indoors. The key advantage of RF based communication is its high coverage range and non-line of sight communications. Wi-Fi finger printing is an efficient RF based technique for IPS which utilizes signal strength of wireless LAN to identify different indoor locations. In this paper, we develop an experimental testbed to implement Wi-Fi fingerprinting based IPS. We analyse the performance of Wi-Fi fingerprinting based IPS for three different indoor scenarios. Results show that Wi-Fi finger printing based IPS accurately predicts user location when the multi-path fading is low and access points are closer to the user.

Accuracy and reliability are important performance indexes for integrated land navigation system. No-reset Federal Kalman Filter (FKF) developed by Carlson has optimal fault-tolerance performance, but error divergences as long time navigation, which is not suitable for practical land navigation system with Inertial Navigation System, GPS, and Odometer (INS/GPS/OD). In this paper, a new local feedback FKF algorithm is propose to improve the accuracy of land system. There are two local filters in the schematic of local feedback FKF, one is INS/GPS, and the other is INS/OD. Each local filter employs its own Inertial Navigation Update (INU), but shares the same Inertial Measurement Unit (IMU) output information. The output of local filters are no longer the linear, local-optimal estimations of common state vectors as in no-reset FKF, but the linear, local-optimal estimations of common navigation parameters. The master filter takes advantages of the parameter estimations and their error covariance to implement global optimal fusion after local feedback compensation. The feedback compensation can depress the nonlinear error accumulation of local filters, furthermore, the independence of local filters guarantees optimal reliability of the land system. Ground based navigation tests were carried out, the results of which verify the correctness and effectiveness of the improvement technique.

In order to solve the problems of heavy computational load and poor real time of the information fusion method based on the federated Kalman filter (FKF), a novel information fusion method based on the complementary filter is proposed for strapdown inertial navigation (SINS)/celestial navigation system (CNS)/global positioning system (GPS) integrated navigation system of an aerospace plane. The complementary filters are designed to achieve the estimations of attitude, velocity, and position in the SINS/CNS/GPS integrated navigation system, respectively. The simulation results show that the proposed information fusion method can effectively realize SINS/CNS/GPS information fusion. Compared with FKF, the method based on complementary filter (CF) has the advantages of simplicity, small calculation, good real-time performance, good stability, no need for initial alignment, fast convergence, etc. Furthermore, the computational efficiency of CF is increased by 94.81%. Finally, the superiority of the proposed CF-based method is verified by both the semi-physical simulation and real-time system experiment.

A primary challenge in indoor positioning is multipath signal propagation, which contributes to measurement error, yielding low positioning accuracy. As the demand for location-based services in indoor settings increases, developing accurate indoor positioning systems becomes crucial to accommodating user requirements. Numerous works have proposed methods to mitigate the effect of multipath error on indoor positioning accuracy. However, this issue remains a major topic in the researchers’ discourse. This paper proposes a novel Geometric Random Sample Consensus (Geometric-RANSAC) position estimation method to enhance positioning accuracy in single base station (BS) indoor positioning systems. The method considers the formation of a virtual multiple BS system model by treating each of the multipath components received at a mobile station (MS) position as a signal transmitted by a virtual BS. The position estimation is executed in two phases. In the first phase, the geometric information contained in each multipath component, including the time of arrival, angle of arrival, and elevation of arrival, is extracted to calculate a position estimate. The number of position estimates produced in this phase depends on the number of multipath components contained in the received signal. In the second phase, the RANSAC algorithm with a centroid model is employed to optimize the final MS position. The simulation results show that the proposed method outperforms several least squares method variants by achieving a median positioning error of less than 4.06 cm, and 95% of the MS positions are estimated with an error of less than 7.86 cm. This proposed method not only addresses the challenges posed by multipath propagation but also offers a simple solution for indoor positioning.

The complementary characteristics of the Strapdown Inertial Navigation System (SINS) and external non-inertial navigation aids like Global Positioning System (GPS) and Celestial Navigation System (CNS) make the integrated navigation system an appealing and cost effective solution for various applications. SINS exhibits position errors owing to errors in initialization of the inertial measurement unit (IMU) and the inherent accelerometer biases and gyroscope drifts. SINS also suffer from diverging azimuth errors and an exponentially increasing vertical channel error. Pitch and roll errors also exhibit unbounded growth with time. To mitigate this behavior of SINS, periodic corrections are opted for through measurements from external non-inertial navigation aids. These corrections can be in the form of position fixing, velocity fixing and attitude fixing from external aids like GPS, GLONASS (Russian Satellite Navigation System), BEIDU(Chinese Satellite Navigation System) and Celestial Navigation Systems (CNS) etc. In this research work GPS and CNS are used as external aids for SINS and the navigation solutions of all three systems (SINS, GPS and CNS) are fused using Federated Kalman Filter (FKF). The FKF differs from the conventional Central Kalman Filter (CKF) because each measurement is processed in Local Filters (LFs), and the results are combined in a Master Filter (MF). FKF is a partitioned estimation method that uses a two stage data processing scheme, in which the outputs of sensor related LFs are subsequently combined by a large MF. Each LF is dedicated to a separate sensor subsystem, and uses data from the common reference such as SINS. The SINS acts as a cardinal system in the combination, and its data is also available as measurement input for the master filter. In this research work, information from the GPS and the CNS are dedicated to the corresponding LFs. Each LF provides its solutions to the master filter all information is fused together forming a global solution. Simulation for the proposed architecture has validated the effectiveness of the scheme, by showing the substantial precision improvement in the solutions of position, velocity and attitude as compared to the pure SINS or any other standalone system.

Location-based services are among important applications in current telecommunication networks which causes an increasing demand in the advancements of indoor positioning systems (IPS). This paper presents a comprehensive review of the technologies and techniques employed in recent works related to IPS and discusses the challenges in IPS implementations. This study widely categorizes indoor positioning technologies into five types which are computer vision, short-range communication, acoustic-based, magnetic methods, and radio frequency (RF) technologies. The strengths and limitations of each technology is discussed based on its accuracy, coverage, infrastructure, implementation cost and signal characteristics. The literature study shows that range-based and fingerprinting are two main techniques employed in IPS. In addition, the study indicates that fingerprinting methods utilizing Wi-Fi and cellular networks are prevalent due to their widespread availability. However, these technologies face some challenges such as multipath fading, signal instability, device heterogeneity, infrastructure and cost implications, computational complexity, and privacy and security concerns. This paper emphasizes the need for innovative approaches to enhance positioning accuracy and reduce infrastructure costs, thereby fostering broader adoption of IPS across diverse applications.

People around the world have come to rely on digital maps–accessible with the tap of a smartphone screen–to guide them places, to locate areas with which they are unfamiliar, and to get directions to desired destinations. In a world that is overflowing with technological advancements, we expect to be shown, in real time, how to get from point A to point B. However, navigating indoors using assistive technology can prove more challenging than navigating outdoors, as indoor spaces have not traditionally been mapped the way outdoor spaces have by companies like Google and Apple. Additionally, most indoor mapping and navigation services are not very accurate or accessible to meet the needs of a variety of users, including individuals with disabilities. Founded by American Printing House for the Blind (APH) in 2019, the GoodMaps Indoor Navigation smartphone app is addressing these challenges by leveraging artificial intelligence (AI) and crowd-sourced data to scale its app and map coverage efficiently. GoodMaps uses AI–primarily through computer vision algorithms–to create highly accurate indoor maps and to enable precise navigation for users. The app allows them to navigate complex indoor spaces like airports, university campuses, and more by providing detailed turn-by-turn directions. GoodMaps’ computer vision technology interprets real-time data from a device's camera and sensors, effectively guiding users through the environment even without visual sightlines. This is particularly valuable for people with visual impairments or mobility challenges who need assistance navigating intricate indoor spaces. Currently, AI-driven translation capabilities enable the app to support four languages, with six more planned by early 2025, automating 90% of destination translations for non-English users. Looking ahead, GoodMaps aims to further harness AI to dynamically adapt to changes in indoor environments. AI-powered image recognition will enable automated detection of environmental updates via user device cameras, while crowd-sourced contributions from users will provide real-time feedback similar to Waze. This combination of AI and community-driven updates will streamline map maintenance, improve accessibility, and set a new standard for scalable indoor navigation solutions. GoodMaps is also partnering with Intel to deliver a high-quality indoor wayfinding solution for people who are blind or visually impaired. Safely and effectively navigating indoor spaces results in greater independence and confidence when traveling. Intel continues to investigate volumetric mapping algorithms and advances in artificial intelligence to improve the precision and accuracy of GoodMaps’ commercial indoor wayfinding service. This paper will explain GoodMaps’ user-centered design process, chronicling how user experience research has informed the development of AI-driven computer vision models to address user requirements.

The wireless indoor positioning system (IPS) has high demand due to wide applications such as asset and personal tracking, location analysis, and mapping. Global Positioning System (GPS) doesn't work within an indoor environment due to the non-line-of-sight (NLOS), to overcome this limitation we are using the most demanded technology Wi-Fi due to its high range, high data throughput with high availability and accessibility. The main problem is the positioning estimation accuracy in the Wi-Fi-based positioning system due to noise and interference in the environment. The received signal strength (RSSI) and signal quality (SQ) become low due to the multipath effects, reflections, signal loss, fluctuations, and other unknown factors. The objective of the proposed method is to reduce the fluctuations and the noise of the received signal and improve the signal strength and signal quality to increase the positioning estimation accuracy and distance estimation accuracy. RSSI smoothing technique of weighted moving average and feedback filter is used to make signal cleaner by removing the fluctuations and signal quality parameters SNR margin, frequency, and power noise to improve the signal quality. The trilateration technique is used to realize distance estimation and position estimation. To analyze the positioning estimation accuracy and distance estimation accuracy of the proposed method we implement the regression accuracy metrics based on mean absolute error (MAE), mean squared error (MSE), and root mean squared error (RMSE). By doing this we achieved an accuracy below 0.5 meter.

IoT (Internet of things) has penetrated every corner of personal life. IPS (Indoor positioning system) is such an exact example which needs the assistance of IoT technology. With the intricate architecture of IPS consisting of multiple components, the reliability of IPS appears essentially. However, besides the component limits the reliability of integral system reliability, there is one more restriction – TPT (Triangulation Positioning Technology). Therefore, in this paper, several experimental results are aimed at exploring the IPS reliability under different influence factors with the restriction of TPT. Experiments are primarily conducted with the form of contrast experiment of two frameworks, which includes the different number and failure rate antennas of framwork1, failure trending of the positioning system of framework1 and framwork2, overall reliability of those two frameworks. The experimental results show that IPS with more antennas and the less failure rate is more stable, framework1 is more reliable than framework2 generally. To conclude, TPT will make an inevitable consequence to the system reliability due to the number restriction of the positioning antenna; positioning processors could also affect the system reliability as a result of the connection method of the antenna. Finally, the last conclusion could be extracted from the connection method between the positioning antennas and positioning processors, experiment shows that separate connection (one antenna to one processor) is much more stable than multiple connection (several antennas to one processor), thus, framwork1 is of more stability than framework2.