This paper presents a novel framework for constructing channel knowledge maps with high accuracy and computational efficiency. The proposed radio propagation graph representation learning (RPGRL) framework integrates physicsbased modeling and data-driven learning through end-to-end optimization using only measured received power data. In RPGRL, both propagation probability and received power estimations are jointly optimized without requiring explicit path supervision, enabling adaptive graph-based representations that account for uncertainties in reflections caused by surface scattering and angular errors introduced by three-dimensional building map quantization. Experimental evaluations using real-world signal measurement data collected in an urban environment demonstrate that RPGRL significantly improves received power estimation accuracy while reducing computation time by more than 70% compared to conventional ray tracing. Furthermore, it outperforms existing deep learning-based methods, such as RadioUNet, by leveraging path-aware estimation grounded in radio propagation physics. These results highlight RPGRL as a promising and practical approach for environment-aware communication and intelligent wireless network design toward 6G.

We consider the design of low complexity and highperforming mean square error (MSE) minimization combiners for over-the-air-computing (AirComp) applications operating over the uplink of a system with one multiple-antenna access point (AP) and multiple single-antenna edge devices (EDs). Within that paradigm, we offer two contributions, namely, a simple initial combiner based on a Rayleigh quotient (RQ) design, and a low-complexity refinement stage based on a convex concave procedure (CCP). The new refinement stage algorithm is further enriched with an efficient (offline) hyper-parameter tuning mechanism via Bayesian optimization (BO) and acceleration method based on a half-space constrained least square problem reformulation solved via the adaptive moment estimation (Adam) algorithm. The low complexity and good performance of the proposed method help address typical limitations of edge devices. Numerical results demonstrate that the proposed design can achieve MSE performances equivalent to those of the best stateof-the-art (SotA) alternatives currently known, at about 200-times less complexity than the highest-performing SotA, and about 4-times less complexity than its low-complexity counterpart.

This paper proposes a novel model for radio propagation graph representation learning in multi-hop path representation. A concept of radio propagation graph representation learning has been proposed to express the radio propagation process using graph data structure, thereby improving the prediction accuracy with small computation time. In the existing work, a direct path representation model based on feedforward neural network (FFNN) has been proposed; however, it achieves poor prediction accuracy at points far from a transmitter due to the lack of consideration of reflection, i.e., multi-hop path. Therefore, this paper proposes a novel learning method to apply the existing concept to multi-hop path cases. Through the performance evaluation using Long Term Evolution signal data in a 2120 MHz band measured in an urban area, we demonstrate that the proposed model outperforms the existing representation model in terms of prediction accuracy and achieves shorter computation time compared with the ray tracing application.

In this paper, we analyze performance of an intelligent reflecting surface (IRS)-assisted simultaneous wireless information and power transfer (SWIPT) system with the optimal phase shift. Specifically, we consider a transmitter sends power and information signals with the assistance of an IRS and spatially correlated fading channels. In practice, the channel between the transmitter and the IRS and between IRS and the receiver are spatially correlated, which constitutes a challenge for accurate performance analysis. In the system, we derive an optimal phase shift, in which the main lobe of the reflected signal at the IRS is directed to the receiver. Then, we develop a closed-form expression to evaluate the average harvested energy and information outage probability. We validate that the proposed model via Monte Carlo simulation.

This study presents a fast and accurate blockchain-based decentralized federated learning (BC-DFL) based on an adaptive client/miner selection algorithm. BC-DFL is a learning method that manages the machine learning models on a blockchain. Although the blockchain can improve the security of model sharing and realize reward management, its mining process extensively increases computation and communication load. The training accuracy could also be degraded if each client’s data distribution follows non-independent and identically distributed (Non-IID) conditions. In the proposed method, a client selection allows parallel processing of local training and mining on the network, reducing round time. In addition, using a client selection algorithm based on the estimation of the label distribution, the accuracy degradation caused by Non-IID is suppressed. Numerical results demonstrate that the proposed method can improve training time and accuracy performances compared to related frameworks.

This paper proposes a novel graph-learning-based radio propagation model, referred to as propagation graph representation learning. Recent advancements in deep learning have succeeded in developing site-specific path loss prediction; however, predicting shadowing without on-site measurement data is still a critical challenge. Propagation graph representation learning aims to express the site-specific propagation process. In the envisioned graph, nodes represent a transmitter, receivers, and obstructions, while edges represent propagation conditions between nodes. The graph structure enables us to recognize the reflection, diffraction, and shielding for accurate shadowing estimation. We also propose a feedforward neural network (FFNN) based representation model in a direct path scenario. Through the simulation using actual datasets in urban areas, we demonstrate that the proposal achieves twenty times faster computation than ray tracing while predicting shadowing well.

Federated Learning (FL) [1] , a new form of distributed machine learning, has attracted much attention as a technique for privacy-preserving big data analysis. Because FL does not disclose raw data outside the device, it can improve data privacy compared to the method aggregating raw data in the cloud.

With the increase in Internet of Things (IoT) devices, machine learning and big data analysis have been rapidly developing. The current big data analysis based on deep learning assumes that the raw data owned by countless users is aggregated into the cloud. However, the number of applications requiring data containing personal information, such as healthcare, increases every year. There are many concerns about operating a system that assumes data aggregation to the cloud due to risks such as data leakage.

It is necessary to develop an efficient channel selection method with low power consumption to achieve high communication quality for distributed massive IoT system. To this end, Ma et al. [1] proposed an autonomous distributed channel selection method based on the Tug-of-War (ToW) dynamics. The ToW-based method can achieve equivalent performance to UCB1-tuned [2], [3] with low computational complexity and power consumption, which is recognized as a best practice technique for solving multi-armed bandit (MAB) problems. However, Ref. [1] only considered fixed IoT devices with simplex communication.

This paper proposes a secure routing protocol based on an ad-hoc on-demand distance vector (AODV) that successfully achieves both security and communication efficiency. Although many studies have previously discussed such a secure protocol, the conventional methods significantly degrade the communication efficiency because of large packets and complex communication procedures. The proposed method allows the intermediate node to generate a route reply (RREP). This is impossible in conventional methods due to the restriction of digital signatures in route requests (RREQ). In the proposed method, each intermediate node holds a packet received from a specific node in the past, and appends the held packet to the RREQ of another node and generates its own signed RREP. This procedure guarantees that the third party holds the route to the destination. It is shown that the proposed method outperforms conventional secure protocols in terms of network load.

This paper proposes a communication strategy for decentralized learning on wireless systems. Our discussion is based on the decentralized parallel stochastic gradient descent (D-PSGD), which is one of the state-of-the-art algorithms for decentralized learning. The main contribution of this paper is to raise a novel open question for decentralized learning on wireless systems: there is a possibility that the density of a network topology significantly influences the runtime performance of DPSGD. In general, it is difficult to guarantee delay-free communications without any communication deterioration in real wireless network systems because of path loss and multi-path fading. These factors significantly degrade the runtime performance of D-PSGD. To alleviate such problems, we first analyze the runtime performance of D-PSGD by considering real wireless systems. This analysis yields the key insights that dense network topology (1) does not significantly gain the training accuracy of D-PSGD compared to sparse one, and (2) strongly degrades the runtime performance because this setting generally requires to utilize a low-rate transmission. Based on these findings, we propose a novel communication strategy, in which each node estimates optimal transmission rates such that communication time during the D-PSGD optimization is minimized under the constraint of network density, which is characterized by radio propagation property. The proposed strategy enables to improve the runtime performance of D-PSGD in wireless systems. Numerical simulations reveal that the proposed strategy is capable of enhancing the runtime performance of D-PSGD.

To reduce the registered data size and maintain the estimation accuracy of a radio map, we have proposed a shadowing classifier-based radio map. On the other hand, the radio map has been widely utilized in various systems, such as a spatial spectrum sharing, spectrum sensing, and localization. Thus, it is necessary to clarify the appropriate classification method of the shadowing components in various environments. In this paper, we evaluate the accuracy of the shadowing classification using two kinds of datasets and two comparison methods. The emulation results show that it is necessary to appropriately choose the classification method according to the presence or absence of outliers of the average received signal power.

We developed Sports Support System that augments the perception of beginner players and supports situation awareness to motivate beginners in multiplayer sports through visual and haptic feedback. Our system provides positional relationship of the opponents using visual feedback. In addition, the position of the opponent beyond the field of view was provided using haptic feedback. An experiment of pass interception as used to compare the visual and haptic feedback. The results confirmed the effectiveness and characteristics of these feedback processes in a multiplayer ball game.

In this paper, we propose a measurement-based spectrum database using model classifier. In the radio propagation, path loss is the fundamental factor to recognize the coverage area. However, it is difficult to accurately estimate the radio environment only path loss estimation because of the shadowing deviation. Therefore, we estimate the propagation model including shadowing, and by determining the usage range of the estimated model, we reduce the registered data size while accurately estimating the radio environment. The database firstly accumulates the received signal strength indicator (RSSI) related to the locations of receivers and we construct the model classifier. Then, the database assigns the propagation model in each mesh so that Root Mean Squared Error (RMSE) between datasets and the models is minimized. We used measurement datasets of a 3GPP cellular band in the real environment to construct the model classifier. Our results show that the proposed method can accurately estimate the radio propagation while the registered data size is significantly reduced. Additionally, we discuss a method of power control based on the proposed method for improving the communication efficiency.

In this paper, we propose a spectrum database for storing received signal power distribution using clustering for reducing the types of registered distribution. In order to estimate radio environment, a measurement-based spectrum database (MSD) is considered and the distribution of the received signal is important for keeping the outage probability of the system performance. However, if we consider the database stores all of the distributions corresponding to each location, the amount of stored data becomes very large. In order to construct a practical database, the amount of stored data should be reduced while the highly estimation accuracy is maintained. Therefore, in this paper, the database firstly estimates the received signal power distribution in each location. Then, the similar distributions are integrated by clustering so the amount of stored data can be reduced. We evaluated the accuracy of the clustering and showed that similar distributions can be accurately clustered. In addition. we performed the transmit power control (TPC) by using the clustered distributions. The results showed that the permissible outage probability can be guaranteed while the amount of stored data can be drastically reduced.

In a received signal strength indicator (RSSI)-based path loss estimation where mobile terminals report the decoding results from the received packets, the reported results are strongly affected by the receiver noise. Because low power instantaneous signals cannot be obtained due to the decoding failure, a path loss estimation considering no effects of the receiver noise underestimates the path loss models. In this paper, we propose a low complexity compensation method for improving the bias of the path loss estimation. The proposed method first estimates the true mean received signal power in each measurement area via a simple filtering for the reported signals. We show that the proposed method can mitigate the effect of the noise.

In order to realize reliable Vehicular-to-Vehicular (V2V) communication systems, the recognition of radio propagation becomes an important basic technology. However, in the current wireless distributed network systems, it is difficult to accurately estimate the radio propagation characteristics because of the locality of the radio propagation due to surrounding buildings geographical features. In this paper, we construct a measurement-based radio environment database for V2V communication systems. The database first accumulates the received signal strength indicator (RSSI) related to the transmission/reception locations from V2V systems. By using the datasets, the average received power maps linked with transmitter and receiver locations are generated. We have performed measurement campaigns of V2V communications in the real environment to observe Received Signal Strength Indicator (RSSI) for the database construction. Our results show that the proposed method has higher accuracy of the radio propagation estimation than the conventional path loss model based estimation.

The concern over wireless distributed networks (WDNs) such as Machine-to-Machine (M2M) systems and Internet of Things (IoT) has risen. In massive WDNs, it is important to strictly manage the interference power among wireless links: radio propagation prediction between distributed terminals is a critical issue for highly efficient spectrum use. In this paper, we propose a novel radio propagation estimation using spectrum database which collects communication results from mobile terminals. The proposed method focuses on the spatial-correlation of radio propagation characteristics between different wireless links. Using maximum likelihood-based pathloss estimation and Kriging-based shadowing estimation, the radio propagation of the wireless link that has arbitrary location relationship can be predicted. From numerical results, it is shown that the proposed method achieves higher estimation accuracy than the conventional pathloss-based estimation methods. Furthermore, we show that the proposed technique can predict the probability density function (PDF) of the estimation error.

In order to fill the gap between various levels of ball game players, we propose the Sports Support System that can reduce the difference among the skills of various players by augmenting the traditional ball game. We implemented a system that visualizes the trajectory and velocity of the ball for soccer games, as one example of ball games. This visualization and the enhancement of the player positions in the field can help beginner players recognize what they should do next and improve their playing skills. This paper introduces the proposed system and discusses, through a user study, its effectiveness for improving reaction speed and enhancing the pleasure of playing sports. Based on the experiment results, the proposed system contributes to reaction time improvements in passing and receiving the ball, i.e., one of the most important skills in playing soccer.