Low-latency and high-precision vehicle localization plays a significant role in enhancing traffic safety and improving traffic management for intelligent transportation. However, in complex road environments, the low latency and high precision requirements could not always be fulfilled due to the high complexity of localization computation. To tackle this issue, we propose a road-aware localization mechanism in heterogeneous networks (HetNet) of the mobile communication system, which enables real-time acquisition of vehicular position information, including the vehicular current road, segment within the road, and coordinates. By employing this multi-scale localization approach, the computational complexity can be greatly reduced while ensuring accurate positioning. Specifically, to reduce positioning search complexity and ensure positioning precision, roads are partitioned into low-dimensional segments with unequal lengths by the proposed singular point (SP) segmentation method. To reduce feature-matching complexity, distinctive salient features (SFs) are extracted sparsely representing roads and segments, which can eliminate redundant features while maximizing the feature information gain. The Cramér-Rao Lower Bound (CRLB) of vehicle positioning errors is derived to verify the positioning accuracy improvement brought from the segment partition and SF extraction. Additionally, through SF matching by integrating the inclusion and adjacency position relationships, a multi-scale vehicle localization (MSVL) algorithm is proposed to identify vehicular road signal patterns and determine the real-time segment and coordinates. Simulation results show that the proposed multi-scale localization mechanism can achieve lower latency and high precision compared to the benchmark schemes.

With the rapid expansion of autonomous vehicles and entertainment applications, video traffic in the Internet of Vehicles (IoV) faces exponential growth. This surge in video demand presents significant challenges for effective pre-caching strategies, particularly due to high vehicle mobility and heterogeneous dwell times at edge nodes caused by varying speeds. In this paper, we propose an efficient adaptive video segment pre-caching scheme (AVSC) for the IoV, addressing varying travel durations of vehicles on road. Specifically, we develop two video evaluation models to balance the popularity of cached video segments with the fidelity of their distribution across the entire video, ensuring temporal continuity. Then, a multi-objective optimization problem is formulated to jointly maximize highlight entropy and segment distribution fidelity. By leveraging the time-frequency characteristics of the wavelet transform, initial segment candidates are identified by detecting significant changes in the time series of chunk popularity (derived from analyzing frame-level popularity). This approach reduces the search space and computation time for subsequent segment selection. Based on the initial segment candidates, the highlight-direction optimal algorithm is proposed to iteratively identify highlight candidates by improving highlight entropy. For Pareto-optimal solutions, a caching-segment adjustment algorithm based on neighborhood search is proposed to determine the final cached video segments. Theoretical guarantees are provided for the identification process. Furthermore, the adjustment algorithm is proven to detect the maximal improvement direction of distribution fidelity, enhancing convergence speed. Simulations on real-world video datasets demonstrate the effectiveness of the proposed AVSC.

Unmanned aerial vehicles (UAVs) have received plenty of attention due to their high flexibility and enhanced communication ability, nonetheless, the limited onboard energy restricts UAVs’ application on persistent data collection missions in large areas. In this paper, we propose a rechargeable UAV-assisted periodic data collection scheme, where a UAV is dispatched to periodically collect data from sensor nodes (SNs) in the mission area and charged by a wireless charging platform. Specifically, the periodic data collection completion time is minimized by optimizing the UAV trajectory to reach the optimal balance among the collection time, flight time, and recharging time. The formulated problem is non-convex and difficult to solve directly. To tackle this problem, we divide the main problem into two sub-problems and address them by leveraging successive convex approximation (SCA), bisection search, and heuristic methods. Then, we propose a periodic trajectory optimization algorithm to iteratively solve the two sub-problems to minimize the completion time. Furthermore, to deal with the dynamics of SNs, we propose a low-complexity trajectory adjustment strategy, where the trajectory can be maintained or adjusted locally at the SNs change, which significantly mitigates the computation cost of re-optimization. The simulation results show the superiority and robustness of the proposed scheme and the completion time is on average 39% and 33% lower than the two benchmarks, respectively.

Vehicle localization is essential for intelligent trans-portation. However, achieving low-latency vehicle localization without sacrificing precision is challenging. In this paper, we propose a road-aware localization mechanism in heterogeneous networks (HetNet), where distinct features of HetNet signals are extracted for two-spatial-scale position mapping, enabling low latency with high precision. Specifically, we propose a sequence segmentation method to extract the low-dimensional positioning space on two scales. To represent roads and sub-segments according to HetNet signals, we propose a salient feature ex-traction method to eliminate redundant features and retain distinct features, thereby reducing feature-matching complexity and improving representation accuracy. Based on the extracted salient features, a two-spatial-scale localization algorithm is designed through salient feature matching, which can achieve low-latency road-aware localization. Furthermore, high-precision positioning is achieved by coordinate mapping based on curve fitting. Simulation results show that our mechanism can provide a low-latency and high-precision positioning service compared to the benchmark schemes.