Localization Error Research Articles

Robust fusion filter for networked uncertain descriptor systems with colored noise and cyber-attacks

The robust fusion filtering problem of multi-sensor networked uncertain descriptor systems (NUDSs) with colored noise, uncertain noise variances and cyber-attacks is investigated. During data transmission in unreliable communication networks, the data can be maliciously attacked by attackers. In other words, the local filters (LFs) may receive false data or may not receive data because of the cyber-attacks. By adopting the singular value decomposition (SVD) method, the original NUDSs can be converted into two reduced-order subsystems with uncertain correlated fictitious white noises, and the cyber-attacks are transformed into the fictitious noises. Cross-covariance matrices between local filtering errors are derived. The robust LFs are obtained according to the minimax robust estimation principle. Under the linear unbiased minimum variance criterion, three weighted fusion algorithms are applied to fuse the LFs. For all allowable uncertainties of noise variances and cyber-attacks, the minimal upper bounds of covariance matrices of the local and distributed fusion filters are guaranteed. The proof of their robustness is established through the minimax estimation principle and Lyapunov equation method. Finally, the correctness and effectiveness of the proposed algorithms are verified by a circuit system example.

Eyes and hand are both reliable at localizing somatosensory targets.

Body representations (BR) for action are critical to perform accurate movements. Yet, behavioral measures suggest that BR are distorted even in healthy people. However, the upper limb has mostly been used as a probe so far, making difficult to decide whether BR are truly distorted or whether this depends on the effector used as a readout. Here, we aimed to assess in healthy humans the accuracy of the eye and hand effectors in localizing somatosensory targets, to determine whether they may probe BR similarly. Twenty-six participants completed two localization tasks in which they had to localize an unseen target (proprioceptive or tactile) with either their eyes or hand. Linear mixed model revealed in both tasks a larger horizontal (but not vertical) localization error for the ocular than for the manual localization performance. However, despite better hand mean accuracy, manual and ocular localization performance positively correlated to each other in both tasks. Moreover, target position also affected localization performance for both eye and hand responses: accuracy was higher for the more flexed position of the elbow in the proprioceptive task and for the thumb than for the index finger in the tactile task, thus confirming previous results of better performance for the thumb. These findings indicate that the hand seems to beat the eyes along the horizontal axis when localizing somatosensory targets, but the localization patterns revealed by the two effectors seemed to be related and characterized by the same target effect, opening the way to assess BR with the eyes when upper limb motor control is disturbed.

Multi-Source Information Fusion-Based Localization in Wireless Sensor Networks

In wireless sensor networks, accurate node localization is essential for ensuring the precision of data collection. The DV-Hop algorithm, a popular range-free localization method, estimates distances between nodes by multiplying hop counts with average hop distances obtained through distance vector routing. However, this algorithm often experiences localization errors in randomly distributed network environments due to considerable inaccuracies in average hop distance calculations and the approximation of actual paths by straight-line paths. This paper introduces an enhanced DV-Hop localization algorithm, which constructs a mathematical model to optimize the mean square error of the average hop distance for any anchor node. This optimization corrects the average hop distance across the network, bringing it closer to the actual value, thus reducing errors and enhancing accuracy. Simulation results indicate that with 150 nodes, a 30% beacon node ratio and a 100-m communication range, the localization error of the improved FuncDV-Hop model decreased from 0.3916 to 0.1705, and the Root Mean Square Error (RMSE) decreased from 24.78[Formula: see text]m to 14.39[Formula: see text]m, thereby improving localization accuracy by 56.46%.

Enhancing Indoor Localization Accuracy in Wireless Sensor Networks: A Hybrid Approach Integrating Hierarchical Structure Poly Particle Swarm Optimization and Teaching–Learning-Based Optimization (HSPPSO-TLBO)

Background and Objective: Enhancing localization accuracy while minimizing development costs poses significant challenges in deploying and managing wireless sensor networks (WSNs). This paper presents an advanced algorithm for node localization in indoor environments, integrating sophisticated optimization techniques. The hybrid algorithm, HSPPSO-TLBO, combines Hierarchical Structure Poly Particle Swarm Optimization (HSPPSO) with Teaching– Learning-Based Optimization (TLBO). Methods: The proposed algorithm HSPPSO-TLBO aims to minimize the mean squared range error (MSRE) resulted by calculating internal distances between nodes using Received Signal Strength Indicator (RSSI). TLBO, with its robust global search capabilities, complements HSPPSO’s local search, preventing convergence to inappropriate local optima. HSPPSO-TLBO offers easy implementation and leverages the cost-free feature of RSSI, making it an attractive choice for enhancing localization precision. Results: Simulation results demonstrate the superior performance of HSPPSO-TLBO compared to other algorithms using different meta-heuristic optimization techniques. The outstanding performance of HSPPSO-TLBO is evident across various evaluation metrics, including localization error, localization rate, and simulation runtime. Conclusion: The proposed algorithm utilizing HSPPSO and TLBO is exceptionally effective in improving localization precision in indoor WSNs due to several key characteristics. These include the seamless integration and easy implementation of both HSPPSO and TLBO, along with the costfree advantage of using the RSSI technique. This combination makes the algorithm a highly functional solution for improving localization accuracy. other: N/A

Assessment of the State of Pharmacovigilance in Secondary Hospitals in the Federal Capital Territory of Nigeria, Using WHO Pharmacovigilance Indicators.

The importance of pharmacovigilance (PV) in ensuring drug safety, especially in the detection and prevention of adverse drug reactions (ADRs) is critical. However, PV activities in Nigeria still face many challenges, such as very low spontaneous reporting rates, and inadequate training and funding. This study assessed the state of pharmacovigilance in the federal capital territory of Nigeria (FCT), using WHO pharmacovigilance indicators. A cross-sectional questionnaire-based survey was carried out among secondary healthcare facilities in the FCT. The WHO Pharmacovigilance Indicators Questionnaire, which consists of the structural, process and outcome measures, was used to collect data from the focal person for pharmacovigilance at all the consenting facilities. Descriptive statistics were used to summarize all variables. Ethical approval was obtained from the Ethics Review Committee of the FCT development authority. Of the 14 secondary healthcare facilities in the FCT, 11 agreed to the study (response rate = 84.6%). Among the respondents, 4 (36.4%) were females, and 2 (18.2%) had 9 years of experience in pharmacovigilance. For the core structural indicators, 7 (63.6%) of the facilities had a pharmacovigilance center while only 4 (36.4%) had a copy of the Nigerian pharmacovigilance policy. Regarding financial provisions, 10 (90.9%) hospitals reported that there was no regular financial provision for the center while 10 (90.9%) centers had a standard adverse drug reaction reporting form. For the core process indicators, the mean ± SD of the nine core process indicators ranged from 0.9 ± 3.0 to 75.6 ± 38.6 and the total number of reports in the local database, therapeutic ineffectiveness, and medication error were limited. The assessment of pharmacovigilance activities in the Federal Capital Territory of Nigeria revealed significant gaps in infrastructure, financial support, and process implementation. Despite the presence of pharmacovigilance centers in the majority of facilities, the lack of consistent financial support and limited adherence to core process indicators highlight the need for enhanced training, resources, and policy enforcement to improve ADR reporting and overall drug safety monitoring. Strengthening these areas is crucial for advancing pharmacovigilance practices and ensuring patient safety in Nigeria.

How optimal is the minimum-time manoeuvre of an artificial race driver?

Minimum-lap-time optimal control problems (MLT-OCPs) are a popular tool to assess the best lap time of a vehicle on a racetrack. However, MLT-OCPs with high-fidelity dynamic vehicle models are computationally expensive, which limits them to offline use. When using autonomous agents in place of an MLT-OCP for online trajectory planning and control, the question arises of how far the resulting manoeuvre is from the maximum performance. In this paper, we improve a recently proposed artificial race driver (ARD) for online trajectory planning and control, and we compare it with a benchmark MLT-OCP. The novel challenge of our study is that ARD controls the same high-fidelity vehicle model used by the benchmark MLT-OCP, which enables a direct comparison of ARD and MLT-OCP. Leveraging its physics-driven structure and a new formulation of the g-g-v performance constraint, ARD achieves lap times comparable to the offline MLT-OCP (few milliseconds difference). We analyse the different trajectories resulting from the ARD and MLT-OCP solutions, to understand how ARD minimises the effect of local execution errors in search of the minimum-lap-time. Finally, we show that ARD consistently maintains its performance when tested on unseen circuits, even with unmodelled changes in the vehicle's mass.

Optimizing Indoor Localization Precision: Advancements in RFID Strategies for Enhanced Performance

Abstract RFID technology is being used more and more for complex interior localization, as a result of the quick spread of mobile devices and the smooth incorporation of artificial intelligence (AI). But conventional approaches have severe difficulties with antenna placement and constrained signal feature sizes, which limits their performance. To improve indoor localization and get over these drawbacks, this study presents a novel method that combines the Unscented Kalman Filter (UKF) with Wavelet Neural Networks (WNN).The most significant development is the incorporation of phase data into input vectors along with Received Signal Strength Indication (RSSI), which significantly increases accuracy. Empirical results show notable improvements of 88.9%, 99.2%, 95.2%, and 37.5% in critical metrics like Mean Localization Error (MLE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Localization Error Offset (LEO) in particular. This work highlights the critical role of sophisticated techniques like WNN and UKF, highlighting the importance of phase as a critical component, to improve the performance of RFID-based indoor localization. The results indicate a promising trajectory for future advancements in this field as well.

Research on Signal Noise Reduction and Leakage Localization in Urban Water Supply Pipelines Based on Northern Goshawk Optimization.

In order to enhance the accuracy and adaptability of urban water supply pipeline leak localization, based on the Northern Goshawk Optimization, a novel joint denoising method is proposed in this paper to reduce noise in negative pressure wave signals caused by leaks. Firstly, the Northern Goshawk Optimization optimizes the decomposition levels and penalty factors of Variational Mode Decomposition, and obtains their optimal combination. Subsequently, the optimized parameters are used to decompose the pressure signals into modal components, and the effective components and noise components are distinguished according to the correlation coefficients. Then, an optimized wavelet thresholding method is applied to the selected effective components for secondary denoising. Finally, the signal components that have been denoised twice are reconstructed with the effective signal components, and the denoised negative pressure wave signals are obtained. Simulation experiments demonstrate that compared to wavelet transforms and Empirical Mode Decomposition, our method achieves the highest signal-to-noise ratio improvement of 12.23 dB and normalized cross correlation of 0.991. It effectively preserves useful leak information in the signal while suppressing noise, laying a solid foundation for improving leak localization accuracy. After several leak simulation tests on the leakage simulation test platform, the test results verify the effectiveness of the proposed method. The minimum relative error of the leakage localization is 0.29%, and an average relative error is 1.64%, achieving accurate leakage localization.

Localisation for underwater acoustic networks with a piece-wise linear sound speed profile

ABSTRACT Underwater agent localisation is a precondition for plenty of ocean applications. The variation of the underwater sound speed (USS) is one of the main factors that constrain the performance of underwater localisation. In this paper, a scheme is designed to reduce the agent localisation error arising from the USS variation in deep water and an agent localisation algorithm named ‘localisation with a piecewise linear USS' (LPL) is proposed. According to the characteristics of a real USS in deep water, a piecewise linear USS model is introduced to approximate the real USS. The agent above under the ‘sound fixing and ranging’ (SOFAR) channel are both considered. The LPL algorithm consists of three parts: initial value estimation, horizontal distance estimation and localisation. Considering the piecewise linear USS profile, the proposed LPL algorithm improves the performance for localisation with 60% more precision than the state-of-the-art methods.

Deep Learning-Based Fine-Tuning Approach of Coarse Registration for Ear-Nose-Throat (ENT) Surgical Navigation Systems.

Accurate registration between medical images and patient anatomy is crucial for surgical navigation systems in minimally invasive surgeries. This study introduces a novel deep learning-based refinement step to enhance the accuracy of surface registration without disrupting established workflows. The proposed method integrates a machine learning model between conventional coarse registration and ICP fine registration. A deep-learning model was trained using simulated anatomical landmarks with introduced localization errors. The model architecture features global feature-based learning, an iterative prediction structure, and independent processing of rotational and translational components. Validation with silicon-masked head phantoms and CT imaging compared the proposed method to both conventional registration and a recent deep-learning approach. The results demonstrated significant improvements in target registration error (TRE) across different facial regions and depths. The average TRE for the proposed method (1.58 ± 0.52 mm) was significantly lower than that of the conventional (2.37 ± 1.14 mm) and previous deep-learning (2.29 ± 0.95 mm) approaches (p < 0.01). The method showed a consistent performance across various facial regions and enhanced registration accuracy for deeper areas. This advancement could significantly enhance precision and safety in minimally invasive surgical procedures.

Improved vision-only localization method for mobile robots in indoor environments

To solve the problem of mobile robots needing to adjust their pose for accurate operation after reaching the target point in the indoor environment, a localization method based on scene modeling and recognition has been designed. Firstly, the offline scene model is created by both handcrafted feature and semantic feature. Then, the scene recognition and location calculation are performed online based on the offline scene model. To improve the accuracy of recognition and location calculation, this paper proposes a method that integrates both semantic features matching and handcrafted features matching. Based on the results of scene recognition, the accurate location is obtained through metric calculation with 3D information. The experimental results show that the accuracy of scene recognition is over 90%, and the average localization error is less than 1 meter. Experimental results demonstrate that the localization has a better performance after using the proposed improved method.

On the adaptive solution of phase‐field problems with A‐stable explicit last‐stage diagonally implicit Runge–Kutta (ELDIRK) methods

AbstractRunge–Kutta algorithms with adaptive step size control provide reliable tools for the solution of initial value problems with diagonally implicit Runge‐Kutta (DIRK) methods as the most common approach. In this contribution, the new low‐order explicit last‐stage diagonally implicit Runge–Kutta (ELDIRK) methods are investigated, combining implicit schemes with an additional explicit evaluation as an explicit last stage. This results in Butcher tableaus with two solutions of different convergence orders suitable for embedded methods, where the higher‐order solution is achieved by additional explicit evaluations. Thus, the iterative solution of non‐linear systems is omitted for the additional stage, presenting a major reduction in computational cost for the determination of a local error estimate. The key contribution is the application of the novel Butcher tableaux to phase‐field problems, solved with the finite‐element method, leading to substantial numerical investigations with an efficient approach for DIRK schemes. The most important aspects are the investigations of stability properties which lead to the novel class of A‐stable ELDIRK methods. In accordance, the study of the convergence orders is presented. A local error estimator is presented, such that adaptive step size control for the new low‐order embedded schemes based on an empirical approach for error estimation is achieved. A suitable parallel algorithm is presented with conclusive two‐dimensional phase‐field simulations based on a Kobayashi‐Warren‐Carter model. The higher‐order convergence suggested by the novel schemes is confirmed, and their effective results are demonstrated, resulting in a valuable semi‐explicit addition to the family of Runge–Kutta time integration schemes.

An efficient three-dimensional node localization using recurrent neural networks in unmanned aerial vehicle-assisted wireless networks: an optimization perspective

PurposeUnlike many existing methods that are primarily focused on two-dimensional localization, this research paper extended the scope to three-dimensional localization. This enhancement is particularly significant for unmanned aerial vehicle (UAV) applications that demand precise altitude information, such as infrastructure inspection and aerial surveillance, thereby broadening the applicability of UAV-assisted wireless networks.Design/methodology/approachThe paper introduced a novel method that employs recurrent neural networks (RNNs) for node localization in three-dimensional space within UAV-assisted wireless networks. It presented an optimization perspective to the node localization problem, aiming to balance localization accuracy with computational efficiency. By formulating the localization task as an optimization challenge, the study proposed strategies to minimize errors while ensuring manageable computational overhead, which are crucial for real-time deployment in dynamic UAV environments.FindingsSimulation results demonstrated significant improvements, including a channel capacity of 99.95%, energy savings of 89.42%, reduced latency by 99.88% and notable data rates for UAV-based communication with an average localization error of 0.8462. Hence, the proposed model can be used to enhance the capacity of UAVs to work effectively in diverse environmental conditions, offering a reliable solution for maintaining connectivity during critical scenarios such as terrestrial environmental crises when traditional infrastructure is unavailable.Originality/valueConventional localization methods in wireless sensor networks (WSNs), such as received signal strength (RSS), often entail manual configuration and are beset by limitations in terms of capacity, scalability and efficiency. It is not considered for 3-D localization. In this paper, machine learning such as multi-layer perceptrons (MLP) and RNN are employed to facilitate the capture of intricate spatial relationships and patterns (3-D), resulting in enhanced localization precision and also improved in channel capacity, energy savings and reduced latency of UAVs for wireless communication.

PyGRF: An Improved Python Geographical Random Forest Model and Case Studies in Public Health and Natural Disasters

ABSTRACTGeographical random forest (GRF) is a recently developed and spatially explicit machine learning model. With the ability to provide more accurate predictions and local interpretations, GRF has already been used in many studies. The current GRF model, however, has limitations in its determination of the local model weight and bandwidth hyperparameters, potentially insufficient numbers of local training samples, and sometimes high local prediction errors. Also, implemented as an R package, GRF currently does not have a Python version which limits its adoption among machine learning practitioners who prefer Python. This work addresses these limitations by introducing theory‐informed hyperparameter determination, local training sample expansion, and spatially weighted local prediction. We also develop a Python‐based GRF model and package, PyGRF, to facilitate the use of the model. We evaluate the performance of PyGRF on an example dataset and further demonstrate its use in two case studies in public health and natural disasters.

4D flow MRI velocity uncertainty quantification.

An automatic method is presented for estimating 4D flow MRI velocity measurement uncertainty in each voxel. The velocity distance (VD) metric, a statistical distance between the measured velocity and local error distribution, is introduced as a novel measure of 4D flow MRI velocity measurement quality. The method uses mass conservation to assess the local velocity error variance and the standardized difference of means (SDM) velocity to estimate the velocity error correlations. VD is evaluated as the Mahalanobis distance between the local velocity measurement and the local error distribution. The uncertainty model is validated synthetically and tested in vitro under different flow resolutions and noise levels. The VD's application is demonstrated on two in vivo thoracic vasculature 4D flow datasets. Synthetic results show the proposed uncertainty quantification method is sensitive to aliased regions across various velocity-to-noise ratios and assesses velocity error correlations in four- and six-point acquisitions with correlation errors at or under 3.2%. In vitro results demonstrate the method's sensitivity to spatial resolution, venc settings, partial volume effects, and phase wrapping error sources. Applying VD to assess in vivo 4D flow MRI in the aorta demonstrates the expected increase in measured velocity quality with contrast administration and systolic flow. The proposed 4D flow MRI uncertainty quantification method assesses velocity measurement error owing to sources including noise, intravoxel phase dispersion, and velocity aliasing. This method enables rigorous comparison of 4D flow MRI datasets obtained in longitudinal studies, across patient populations, and with different MRI systems.

A nonsmooth Levenberg–Marquardt method based on KKT conditions for real-time pricing in smart grid

The real-time pricing of smart grid, which is an important mean of demand side management, is an ideal method to adjust the power balance between the supply and demand in smart grid system. By transforming the social welfare maximization model into a nonsmooth system of equations based on KKT conditions and investigating the related nonsmooth-function’s generalized Jacobi, a nonsmooth Levenberg–Marquardt method with global convergence is proposed in this paper. Also, the local convergence rate is obtained under the local error bound, which is weaker than the nonsingular condition. Finally, numerical simulations show that the smaller the approximate parameter in smoothing Newton’s method, the closer the electricity price is to the value obtained by the proposed algorithm. And the price of the multi price based on the method is lower than that of the single price algorithm that does not consider user types. Meanwhile, compared with the distributed algorithm with different scales, it is found that the price is lower and the social welfare is higher, which shows that the algorithm is robust and effective.

Multi-level surfel map construction for accurate localization in intelligent vehicles

Abstract The role of mapping in achieving accurate localization in intelligent vehicles is crucial. Accurate maps enable intelligent vehicles to attain accuracy in localization and path planning, ultimately enhancing the safety and overall driving experience. However, the interference of dynamic objects poses a challenge to map updates and subsequent localization. To address this issue, this study proposes a novel algorithm for constructing multi-level semantic surfel maps, along with a corresponding hierarchical localization method. The proposed method divides the map into three levels: semantic level, surfel level, and trajectory level. During the map construction phase, we utilize semantic segmentation technology to accurately distinguish between static and dynamic environments, filter out dynamic point clouds, and focus solely on modeling static scenes. This effectively eliminates the interference of dynamic objects on map updates, ensuring the construction of a precise and reliable static environment model. In the subsequent hierarchical localization phase, a node-level localization method is introduced, which achieves coarse localization by matching perception information with the map.The results of experiments conducted in various scenarios indicate an average absolute trajectory error of 4cm between the transformation matrices of each node in the map. Our algorithm achieves a 92% success rate in node-level localization and an average pose-level localization error of 8.2cm, demonstrating the feasibility and robustness of the proposed algorithm.

A Remote Two-Point Magnetic Localization Method Based on SQUID Magnetometers and Magnetic Gradient Tensor Invariants.

In practical application, existing two-point magnetic gradient tensor (MGT) localization methods have a maximum detection distance of only 2.5 m, and the magnetic moment vectors of measured targets are all unknown. In order to realize remote, real-time localization, a new two-point magnetic localization method based on self-developed, ultra-sensitive superconducting quantum interference device (SQUID) magnetometers and MGT invariants is proposed. Both the magnetic moment vector and the relative position vector can be directly calculated based on the linear positioning model, and a quasi-Newton optimization algorithm is adopted to further improve the interference suppression capability. The simulation results show that the detection distance of the proposed method can reach 500 m when the superconducting MGT measurement system is used. Compared with Nara's single-point tensor (NSPT) method and Xu's two-point tensor (XTPT) method, the proposed method produces the smallest relative localization error (i.e., significantly less than 1% in the non-positioning blind area) without sacrificing real-time characteristics. The causes of and solutions to the positioning blind area are also analyzed. The equivalent experiments, which were conducted with a detection distance of 10 m, validate the effectiveness of the localization method, yielding a minimum relative localization error of 4.5229%.

Improved Binaural Hearing Effects and Sound Source Localization in Listeners with Simulated Unilateral Conductive Hearing Loss: Effect of a Bone Conduction Device.

Objectives: It has been proven that patients with unilateral conductive hearing loss (UCHL) may encounter typical problems associated with asymmetric hearing, especially in challenging listening environments. In this study, we aimed to determine how UCHL affects speech recognition under multisource competing environments and the ability of sound source localization, as well as whether assistance with a bone conduction device (BCD) can confer hearing benefits in such listening tasks. Design: Acquired UCHL was simulated using an earplug combined with an earmuff in 10 listeners (mean age: 29.9 ± 4.77 years) with bilateral normal hearing (NH), and a within-subject repeated-measures design was used to compare hearing results among three listening conditions: NH (both ears open, C1), unilateral plug (UP) (simulated UCHL, C2), and UP + BCD (simulated UCHL aided with a BCD, C3). The speech reception threshold (SRT) of summation, squelch, and head shadow (HS) effects and the mean absolute error of sound source localization were used as markers for binaural hearing abilities. Results: BCD assistance (C3) improved the summation and HS effects in all listeners with simulated UCHL, resulting in a lower (i.e., better) SRT than that observed in C2; however, no significant differences in squelch effects were observed between C2 and C3. Notably, most listeners exhibited more accurate sound source localization in C3 than in C2. Further, BCD assistance mainly improved localization accuracy when the noise stimuli were presented at low intensities and on the hearing-impaired (plugged) side, suggesting that the benefits of BCD for sound localization are not based on the reacquisition of binaural processing. Conclusions: The current results have clinical implications for the promotion of BCDs in patients with UCHL, especially those with acquired UCHL who are unable to undergo surgery.

An experimental system for detection and localization of hemorrhage using ultra-wideband microwaves with deep learning

Stroke is a leading cause of mortality and disability. Emergent diagnosis and intervention are critical, and predicated upon initial brain imaging; however, existing clinical imaging modalities are generally costly, immobile, and demand highly specialized operation and interpretation. Low-energy microwaves have been explored as a low-cost, small form factor, fast, and safe probe for tissue dielectric properties measurements, with both imaging and diagnostic potential. Nevertheless, challenges inherent to microwave reconstruction have impeded progress, hence conduction of microwave imaging remains an elusive scientific aim. Herein, we introduce a dedicated experimental framework comprising a robotic navigation system to translate blood-mimicking phantoms within a human head model. An 8-element ultra-wideband array of modified antipodal Vivaldi antennas was developed and driven by a two-port vector network analyzer spanning 0.6–9.0 GHz at an operating power of 1 mW. Complex scattering parameters were measured, and dielectric signatures of hemorrhage were learned using a dedicated deep neural network for prediction of hemorrhage classes and localization. An overall sensitivity and specificity for detection >0.99 was observed, with Rayleigh mean localization error of 1.65 mm. The study establishes the feasibility of a robust experimental model and deep learning solution for ultra-wideband microwave stroke detection.