This study is motivated by the demand for an efficient deep learning-based model that helps us predict the future link quality for intelligent decision-making systems. In this letter, we propose a transfer learning-based approach to predict millimeter-wave future received power in an indoor environment. The model is pre-trained using formulation-aid generated data and fine-tuned using measured data. The proposed framework reduces more than 30% in root-mean-square error and 6.5% in accuracy with high training speed compared to the baseline training from scratch.
Studies are being conducted on analysis methods for multilayer networks. If each layer network of the original multilayer network can be reconstructed from the single-layer network obtained by aggregating the multilayer network, a more detailed analysis of the multilayer network will become possible by using conventional network analysis methods. In this study, we investigate the extent to which the original multilayer network can be reconstructed from the single-layer network with the aggregated multilayer network. Our findings include that semisupervised learning of graph neural networks can reconstruct more than 50% of an original multilayer network from the single-layer network with the aggregated multilayer network.
In this paper, we describe a novel 3-D Luneburg lens using radial holes. A Luneburg lens can suppress the side lobe of the antenna and increase the gain. The prototype Luneburg lens was fabricated using a 30 mm radius dielectric sphere with uniform permittivity by 2.1 and many holes by 1 mm radius, called radial holes drilled vertically on the sphere surface. The relative permittivity distribution of the Luneburg lens is realized by changing the number and depth of five types of holes. To confirm the effectiveness of the fabricated Luneburg lens, the far-field radiation pattern at 25 GHz was measured using a flanged rectangular waveguide as a wave source. It achieved a high gain of 22.4 dBi. The measured radiation pattern generally agreed well with the theoretical radiation pattern of the Luneburg lens, and the validity of the Luneburg lens with radial holes at the quasi-millimeter wave was confirmed. It can be used for beam steering in 5G/6G base stations.
Regarding the waking up of terrestrial Internet of Things devices via a low Earth orbit satellite downlink signal, it is difficult to demodulate the wake-up signal received by such devices that are directly below a satellite due to Doppler variation caused by the satellite's high-speed movement. This letter proposes a device wake-up signal transmission method for reducing the effect of Doppler shift variation on a device’s receive signal by adding frequency variation to the wake-up signals that cancels out the Doppler variation and transmitting the signal in downlink. Computer simulations showed that the proposed method enables the waking up of devices directly below a satellite.
The non-backtracking random walk is known to have nicer properties than simple random walks. However, the effect of graph structure on the properties of the non-backtracking random walk has not been fully clarified. In this paper, we elucidate the mean first passage time of the non-backtracking random walk on two different types of graphs with different characteristics as well as the mean cover time on random regular graphs. Furthermore, we analyze the effects of the graph structure on the properties of the non-backtracking random walk using several numerical examples. Our findings include that the number of nodes in random regular graphs has a great impact on the mean first passage time of the non-backtracking random walk.
Studies have recently been conducted to model mobile agents on unknown graphs, such as random walks (RWs) on graphs, and to understand their mathematical properties. In this study, we investigate the extent to which the properties of RWs can be improved when mobile agents have access to very limited information. We propose Q-weighted random walk (QW-RW), in which an agent decides a next node by using Q-values learned by Q-learning, and examine its effectiveness. We find that in small scale-free graphs, QW-RW is 1.25 times faster than self-avoiding RW to cover 80% of the entire graph.
For flexible and cost-effective deployment of millimeter-wave band wireless systems, we have proposed a novel remote beamforming scheme with fixed wavelength allocation for Radio-over-Fiber systems, which can remotely control the beam direction of base station from central station. This paper applies the scheme to two-dimensional array so that beamsteering in both horizontal and vertical directions can be carried out, which offers more flexible design of wireless coverage area. Simulations are conducted to reveal the performance of the scheme for two-dimensional array. The results confirm its feasibility and indicate that the total beam performance mainly depends on one of the axes of the array.
Handover is very crucial in cellular communication since it may interrupt data transmission. The traditional handover algorithm works well in ideal conditions, but it may fail in certain conditions such as coverage holes. In this letter, we present our simulation to improve handover performance using machine learning. We implement and perform novel modifications on Near-RT RIC, a new network element by O-RAN, to solve the target cell decision problem using a machine learning algorithm in a network with a coverage hole. We show that this solution is proven superior to the traditional handover algorithm.
The Internet of Things (IoT) have been widely used to monitor control systems of critical infrastructure. IoT device’s self-localization functionality is important to realize a connected IoT ecosystem. This study proposes a probabilistic approach to evaluate self-localization quantitatively by utilizing network gateway allocation while considering probabilistic loss of WiFi signals from gateways. We use probabilistic model checking to evaluate how probabilistic signal loss affects self-localization accuracy. The investigation of such uncertainty serves as the foundation for real-world IoT self-localization.
UAV route constructing methods have been researched based on an ad-hoc networks protocol, OLSR, composed by smart meter radio devices. We established the method which construct the most optimal route for UAV travel in terms of travel distance and safety avoiding densely populated area. In this research, in order to eliminate the risk of UAVs collision due to overlapping routes of multiple UAVs navigating simultaneously, we propose a method of locking links so that the link used for a UAV constructing route cannot be used for other UAVs. We also propose link hierarchization in the air to solve the link shortage caused by locking links. We evaluate effectiveness and characteristics of the proposed methods by computer simulation.
This study significantly reduces the height of a shared antenna for overhead line voltage detection and wireless VHF band communication installed on the roof of the Shinkansen vehicle. We propose an antenna configuration employing an inverted-F antenna utilizing a wide-shaped branching element with a notched tip that can maintain antenna performance, even if the height of the antenna is significantly lower. It was proven that the notch at the tip improves average gain in the horizontal plane by 2.3 dB and 3.9 dB at two operating frequencies, respectively. In addition, the deterioration of the overhead line voltage detection sensitivity was only 30% even though the antenna height was lowered by approximately 72%. To put this antenna to practical use, it is necessary to verify the usage environment and vehicle mounting conditions.