This paper explores the use of few-shot learning in Wi-Fi-based indoor positioning, utilizing convolutional neural networks (CNNs) combined with meta-learning techniques to enhance the accuracy and efficiency of positioning systems. The focus is on addressing the challenge of limited labeled data, a prevalent issue in extensive indoor environments. The study explores various scenarios, comparing the performance of the base CNN and meta-learning models. The meta-learning approach involves few-shot learning tasks, such as three-way N-shot, five-way N-shot, etc., to enhance the model’s ability to generalize from limited data. The experiments were conducted across various scenarios, evaluating the performance of the models with different numbers of samples per class (K) after filtering by cosine similarity (FCS) during both the stages of data preprocessing and meta-learning. The scenarios included both base classes and novel classes, with and without meta-learning. The results indicated that the base CNN model achieved varying accuracy levels depending on the scenario and the number of samples per class retained after FCS. Meta-learning performed acceptably in scenarios with fewer samples, which are the distinct datasets pertaining to novel classes. With 20 samples per class, the base CNN achieved an accuracy of 0.80 during the pre-training stage, while meta-learning (three-way one-shot) achieved an accuracy of 0.78 on a new small dataset with novel classes.

This study compares the effectiveness of three methods—Fréchet Distance, Dynamic Time Warping (DTW), and Convolutional Neural Networks (CNNs)—in detecting similarities and pattern recognition in time series. It proposes a Pattern-Driven Case Generator (PDCG) framework to automate the creation of labeled time series data for training CNN models, addressing the challenge of manual dataset curation. By injecting controlled noise and interpolating diverse shapes (e.g., W/M/nAn/vVv), a PDCG synthesizes realistic training data that enhances model robustness. Experimental results demonstrate that the CNN model, trained with 10,000 PDCG-generated cases, achieves 86–98% accuracy in pattern recognition, outperforming traditional methods (Fréchet and DTW) for complex, misaligned, and variable-length sequences. The PDCG-enhanced CNN’s scalability and adaptability improve with larger datasets, validating the PDCG’s efficacy in bridging simulation and real-world applications.

The future ITS-based road network can provide a seamless and integrated transportation. The network makes it more complicated for the cars to decide for an optimal choice or K-shortest paths under the autonomous driving situation. The paper presents the description and optimization procedure of a K-shortest path algorithm for autonomous driving in a road network system. Computation analysis was conducted to determine the effectiveness and efficiency of the algorithms and to validate the results. The implementation of such a routing algorithm could enhance the utilization of the autonomous driving system in future.

Indoor positioning services (IPS) in large commercial complexes and shopping centers have broad application prospects. This study uses raw WiFi signal data and supervised learning through Convolutional Neural Networks (CNN) for modeling and predicting indoor positioning. By collecting real-location indoor WiFi signal information, and performing data transformation, normalization, and noise reduction, the CNN algorithm is used for model training and evaluated using a validation set. The results show the positioning model using the CNN algorithm achieves significant improvements in positioning accuracy, with an accuracy rate of 73% and an F1-Score of 0.72 on the validation set. The CNN model effectively improves signal stability, demonstrating superior performance compared to traditional WiFi fingerprinting algorithms, which typically have an average positioning error of 2-3 meters. These findings suggest that CNN models have great potential in enhancing the accuracy and efficiency of indoor positioning systems.

Indoor Positioning Services (IPS) allow mobile devices or bionic robots to locate themselves quickly and accurately in large commercial complexes, shopping malls, supermarkets, exhibition venues, parking garages, airports, or train hubs, and access surrounding information. Wi-Fi-based indoor positioning technology can use existing WLAN networks, and has promising prospects for broad market applications. This paper presents a method using the Multinomial Logit Model (MNL) to generate Wi-Fi signal fingerprints for positioning in real time. In an experiment, 31 locations were randomly selected and tested to validate the model, showing mobile devices could determine their locations with an accuracy of around 3 m (2.53 m median).