Exploiting 2-D Representations for Enhanced Indoor Localization: A Transfer Learning Approach

IEEE Sensors Journal, Journal Year: 2024, Volume and Issue: 24(12), P. 19745 - 19755

Published: May 2, 2024

Indoor localization systems predominantly depend on one-dimensional signal measurements, such as the Received Signal Strength Indication (RSSI) from Bluetooth or WiFi access points (AP). Such methods, however, grapple with issues like interference other APs and environmental challenges. To address these, our paper introduces an innovative indoor technique employing a classification system bolstered by transfer learning. Instead of relying solely signals, we transform them into images using techniques spectrograms, scalograms, Gramian Angular Fields. These transformed feed learning approach. We tested method two public datasets, achieving remarkable accuracy rates 99% Google-Net model 98% Squeeze-Net architecture. figures underscore efficacy for localization, marking notable advancement over existing strategies.

Language: Английский

---

Applications of knowledge distillation in remote sensing: A survey

Information Fusion, Journal Year: 2024, Volume and Issue: unknown, P. 102742 - 102742

Published: Oct. 1, 2024

Language: Английский

---

Detection of high-risk diseases in poultry feces through transfer learning

Engineering Science and Technology an International Journal, Journal Year: 2025, Volume and Issue: 64, P. 102002 - 102002

Published: Feb. 21, 2025

Language: Английский

---

High-Accuracy and Real-Time Fingerprint-Based Continual Learning Localization System in Dynamic Environment

Sensors, Journal Year: 2025, Volume and Issue: 25(5), P. 1289 - 1289

Published: Feb. 20, 2025

In dynamic environments, localization accuracy often deteriorates due to an outdated or invalid database. Traditional approaches typically use Transfer Learning (TL) address this issue, but TL suffers from the problem of catastrophic forgetting. This paper proposes a fingerprint-based Continual (CL) system designed retain old data while enhancing for new data. The works by rehearsing parameters in lower network layers and reducing training rate upper layers. Simulations conducted with fused show that proposed approach improves 16% 29% compared smaller rooms. larger rooms, it achieves 14% improvement 44% over TL. These results demonstrate CL not only enhances also effectively mitigates issue

Language: Английский

---

A survey and future outlook on indoor location fingerprinting privacy preservation

Computer Networks, Journal Year: 2025, Volume and Issue: unknown, P. 111199 - 111199

Published: March 1, 2025

Language: Английский

---

Optimizing Indoor Localization and Tracking: An Energy‐Efficient Approach Using Received Signal Strength and Mixstyle Neural Networks With Implicit Unscented Particle Filtering

International Journal of Communication Systems, Journal Year: 2025, Volume and Issue: 38(7)

Published: March 24, 2025

ABSTRACT In indoor environments, the unpredictable noise in received signal strength indicator (RSSI) measurements causes very high estimation errors for target localization. Nowadays, RSSI‐based localization systems are widely used to deal with higher levels RSSI and assure more accuracy this paper, Optimizing Indoor Localization Tracking: An Energy‐Efficient Approach Using Received Signal Strength Mixstyle Neural Networks Implicit Unscented Particle Filtering (OILT‐MNN‐IUPF) is proposed. The proposed method consists of two range‐free schemes wireless sensor networks (WSN) an setup: (i) mixstyle neural network (MNN) regression tasks (ii) fusion MNN implicit unscented particle filter (IUPF). fusion‐based model named + IUPF approach. There no need compute distances using field solutions, here three trace mobile target. Also, paper discusses energy consumption related typical trilateration MNN‐based With schemes, linear, sigmoid, RBF, polynomial four kernel functions estimated on OILT‐MNN‐IUPF achieves 25.05%, 20.17%, 23.19% lower average error 23.11%, 20.11%, 24.09% less root mean square compared existing models.

Language: Английский

---

Localization technologies for smart agriculture and precision farming: A review

Computers and Electronics in Agriculture, Journal Year: 2025, Volume and Issue: 236, P. 110464 - 110464

Published: May 6, 2025

Language: Английский

---

Time matters: Empirical insights into the limits and challenges of temporal generalization in CSI-based Wi-Fi sensing

Internet of Things, Journal Year: 2025, Volume and Issue: unknown, P. 101634 - 101634

Published: May 1, 2025

Language: Английский

---

Smart tech meets forensics: Enhancing crime scene investigation with digital simulations

Forensic Science International, Journal Year: 2024, Volume and Issue: 365, P. 112296 - 112296

Published: Nov. 12, 2024

Language: Английский

---

Deep Learning-Empowered RF Sensing in Outdoor Environments: Recent Advances, Challenges, and Future Directions

Electronics, Journal Year: 2024, Volume and Issue: 14(1), P. 125 - 125

Published: Dec. 31, 2024

Recently, with advancements in Deep Learning (DL) technology, Radio Frequency (RF) sensing has seen substantial improvements, particularly outdoor applications. Motivated by these developments, this survey presents a comprehensive review of state-of-the-art RF techniques challenging scenarios practical issues such as fading, interference, and environmental dynamics. We first investigate the characteristics environments explore potential wireless technologies. Then, we study current trends applying DL to RF-based systems highlight its advantages dealing large-scale dynamic environments. Furthermore, paper provides detailed comparison between discriminative generative models support sensing, offering insights into both theoretical underpinnings applications Finally, discuss research challenges present future directions leveraging sensing.

Language: Английский

---