The authors propose a mode conversion technique in a cavity-backed circularly polarized magneto-electric (ME) antenna in order to achieve a wide axial ratio (AR) beamwidth covering a wide frequency range. The proposed technique allows both the ME antenna element and the cavity structural parameters to be independent of each other, which would otherwise always be associated with each other. The proposed antenna demonstrated impedance bandwidth (|S₁₁| ≦ -10 dB) of 38.73% (1.52-2.25 GHz), 3-dB axial ratio bandwidth of 33.05% (1.49-2.08 GHz), and 3-dB AR beamwidth exceeding 214° in a single plane (xz or yz-plane) across the operating band.

In use cases of IoT remote control, IoT applications such as image processing can be placed in the edge or cloud computing to reduce the cost of IoT devices. Additionally, end-to-end (E2E) communication quality including network quality and edge computing processing time is very important for IoT remote control. However, current measurement tools cannot measure the E2E quality directly when edge processing is involved. Therefore, the network quality and the processing time of edge applications need to be separately measured. Measuring the processing time of these applications requires not only modifications for each application in conventional methods but also complicated configuration work. Therefore, we propose a method to measure the processing time of edge applications by detecting packets embedded with unique identifiers, without requiring any configurations to applications. The implementation and evaluation results show that our method can accurately measure the processing time required on edge application.

This paper presents a trajectory tracking algorithm for high-mobility targets using an extended Kalman smoothing (EKS)-based factor graph (FG). Traditional tracking methods often face challenges in maintaining accuracy and computational efficiency when dealing with fast-moving objects. Leveraging the probabilistic framework of factor graphs and robust estimation of EKS, the algorithm enhances tracking precision for fast-moving objects. Extensive simulations across various motion models demonstrate improved accuracy and robustness. The results indicate that this method effectively addresses the limitations of conventional tracking algorithms, providing a promising solution for applications in aviation, autonomous vehicles, and other domains requiring high-mobility tracking.

Received signal strength (RSS) data is one of the useful data types for understanding wireless link quality. To perform WLAN monitoring over a wide area of the target environment, it is necessary to use multiple sensor nodes to observe RSS data at various positions. In this situation, achieving synchronization among sensor nodes is challenging. In this paper, we propose a synchronization method for different sets of RSS data. In our method, by analyzing In-phase and Quadrature (IQ) samples, the preamble, which is part of a wireless frame, is detected, and the reception start timing of each preamble is used as a synchronization timing point. Through experiments in an actual indoor environment, we confirmed that our method can achieve the synchronization accuracy of less than 1 μs between different sets of RSS data.

A single-mode labeling approach is proposed for a mode-multiplexed OPS system, and the corresponding detection performances are evaluated as a function of the received power. We introduce the mode dimension to increase the number of labels that can be generated and processed by a single multiport E/D, and we analyze configurations with reduced intra-mode crosstalk. We show that the label error rate is significantly influenced by inter-mode crosstalk, being TE10 labels more susceptible to crosstalk from adjacent ports, than TE00 labels.

Human motion capture has various use cases such as digital twinning. Most wearable sensing devices to capture data employ short-range wireless protocols such as serial communications over Bluetooth Low Energy (BLE) to satisfy the requirement of low power consumption. On the other hand, the short-range wireless communications limit the applications of human motion capture since digital twinning is often considered more useful in a wide area, e.g., capturing motions of sports players in a large stadium, etc. However, the problem is that replacing BLE with 5G for the existing sensors is not favorable for developers due to the additional cost of re-designing the system, not to mention that it is infeasible for end-users as the sensor protocol is proprietary. In order to overcome this challenge, we propose “black-magic 5G Sidelink” to extend the range of the existing human motion capture sensing system with serial communications via BLE, “transparently” tunneling the black-box sensor protocols over 5G. Our contributions are threefold. First, we propose a viable architecture of transparent extension of any existing BLE serial sensor systems to a wide area. Second, we actually implement the proposed system. Third, our evaluation reveals that the captured data from many people can be efficiently acquired with a low latency of about 40 msec over a wide area, and thus shows the feasibility of transparently extending short-range black-box wearable sensors to a wide area via 5G Sidelink by our proposal.