This paper reports a method for interconnecting local 5G networks and existing networks. Since the two networks are often separated, there is a demand to interconnect the two networks. However, operation gaps such as addressing policy and security policy between the two networks will not allow one to interconnect. To fill the gaps, an interconnection method consists of inter-layer address translation and tunneling in existing networks is proposed and implemented. The evaluation result using an actual local 5G network and an actual corporate intranet revealed that the proposed method works properly.

This paper evaluates QoS of the next-generation in-vehicle Ethernet-based networks in combination with the following three controls defined in IEEE 802.1TSN by simulation. The authors target Strict Priority Queueing (SPQ), Frame Replication and Elimination for Reliability (FRER) and Frame Preemption (FP). The experimental results indicate the followings. First, traffic of the highest priority should be applied SPQ to guarantee its QoS. Second, it is necessary to adopt an appropriate combination of QoS control for traffics of other priority in order to prevent them from deteriorating too much.

In CP(Cyclic Prefix)-Free OFDM systems, an occurrence of IBI(Inter-Block-Interference) is inevitable due to the loss of CP. Since it deteriorates the system performance, a GSC(Generalized Sidelobe Canceller)-based equalizer for SIMO(MIMO)-OFDM systems have been proposed. In the equalizer, a set of basis of null space of channel matrix has to be found. In this letter, we propose a GSC equalizer using QR decomposition for CP-Free MIMO-OFDM systems, where the reduced complexity maximum likelihood detection by QR M-algorithm(QRM-MLD) can be applied after the equalization. By computer simulations, we show the BER performance can be improved as compared to the conventional GSC equalizer with the V-BLAST detection.

A base station scanner that can evaluate MIMO antennas for LTE downlink signaling has been proposed. In this system, all signal processing is done offline, so analysis is significantly delayed. Further, only FDD-LTE is supported. This report details a coherent base station scanner that offers real-time demodulation of not only FDD but also TDD. Experiments on LTE signals using both duplex modes confirm that the developed scanner offers excellent performance and reduced post analysis times.

In this paper, we discuss a visible light communication (VLC) system using a digital signage and an image sensor. We focus on the demodulation part and propose a method that demodulates transmitted data signals from received images of the image sensor using machine learning. The performance of the proposed machine learning-based demodulation method is evaluated for various datasets that simulate noise, blurring, and misalignment that may occur in received images.

In this letter, we propose clipping and clip-noise compensation techniques that improve the self-interference (SI) cancellation and bit error rate (BER) performance of an in-band full-duplex transceiver under power amplifier (PA) nonlinearities. These techniques can be used to improve performance without increasing the order of the parallel Hammerstein canceller. We introduce a clipping at the transmitter that suppresses nonlinear amplification of a PA to improve the SI cancellation performance. Furthermore, clip-noise compensation is presented at the receiver that prevents degradation of the BER performance caused by clipping.

This letter proposes a defective judgment method for an automotive wire harness by a deep learning technique based on a convolutional neural network (CNN). In the proposed method, the CNN is used to find the relationship between the cross-sectional shape of the wire harness and a common mode current. We consider several patterns of the inputs and outputs of the CNN and discuss which input-output pattern is adequate for the judgment.

When divers are doing rescue operations in the sea, they are in a hazardous environment. If it becomes possible to know their positions accurately, rescue operations can be carried out more safely and reliably. We propose a subsea position estimation system using received signal strength (RSS) of electromagnetic waves to assist rescue operations by locating divers in the sea. In this report, as a basic study for introducing machine learning into the subsea position estimation system, four supervised machine learning models were used for undersea position estimation and each model is compared. The best model is Multi-Layer Perceptron (MLP), and it is confirmed that there is no problem in real-time computation time for all models.

Elastic optical networks (EONs) have attracted attention as a means of accommodating the rapidly increasing traffic in core networks. In EONs, routing, modulation and spectrum assignment(RMSA) algorithm is implemented to flexibly allocate resources to optical paths. However, the application of EONs is limited because of the fragmentation of spectrum resources, which hinders further allocation of resources to new optical paths. Subsequently, the Min-RDDR method has been proposed to allocate resources for advance reservation requests. It defined an evaluation formula for calculating a metric to suppress fragmentation. Herein, we modified the Min-RDDR method and proposed a spectrum allocation method to further suppress fragmentation, wherein resources were considered not only within links but also between adjacent links. The simulation results confirmed that the proposed method causes a 47% reduction of request blocking probability, verifying its effectiveness toward fragmentation suppression.

In this letter, we have developed a novel and simple feature extraction that exploits the characteristics of I/Q signals received from ZigBee devices to represent RF circuit impairments such as the frequency offset and I/Q imbalance. Experiments have been carried out to evaluate the identification performance in several scenarios: line-of-sight (LOS) and non line-of-sight (NLOS) environments, varying distance between the transmitter and receiver. The results show that our method can identify twelve ZigBee devices with high accuracy. In addition, it is revealed that some features are sensitive to environmental change during the training and test phases.

HAPS (High-Altitude Platform Station) is expected to extend the communication coverage in the era of beyond 5G. HAPS moves around a circle at 20km altitude and tries to make beams fixed on the earth by beamforming. However, if HAPS uses polarization like satellite communication, there is a possibility that received SNR (Signal to Noise power Ratio) degrades because the fluctuation of HAPS aircraft could generate polarization mismatch between HAPS and earth stations. This paper reports experimental results using on-air polarization signal from a satellite with artificial polarization offset to evaluate feasibility of polarization utilization in HAPS. The results show SNR loss due to the polarization offset can be confined within 0.5dB and good tracking performance thanks to the polarization diversity by MMSE (Minimum Mean Square Error) receiver.

To increase symbol rate and security at the physical layer for uplink optical camera communication from a smartphone screen to indoor camera at a distance of 3.5 m, space division multiplexing by adaptive thresholding is proposed and experimentally verified. The adaptive thresholding sets the optimal threshold for a center cell with each outer 8-cell pattern in 3×3 cells. Although the outer 8-cell pattern is estimated from conventional fixed threshold, the pattern helps to determine the threshold adaptively. The adaptive thresholding achieved 216.75 and 168.75 kilo symbol per second using 85×85 and 75×75 cells at luminance of 255 and 95, respectively.

Dedicated hardware and cluster configuration tend to be applied to legacy servers built for infrastructure systems (ex. call control server) to reduce the failure rate. The transition of legacy servers to a virtual environment with common hardware like general information services is being considered, due to the advancement of virtualization and improved performance of common hardware. In conjunction with this, a suitable method is needed for updating applications of servers. This letter suggests methods for updating applications of legacy servers built in a virtual environment.

In order to transmit BS/CS left-handed IF signals used in 8K4K broadcasting, it is necessary to expand the maximum transmission frequency of the FM conversion system from 2.1GHz to 3.2GHz. Our solution is a new FM conversion optical transmitter using a wideband phase modulator and narrow linewidth LD without frequency divider. In addition, we conduct transmission experiments on a prototype transmitter, and clarify that this transmitter offers long-distance transmission over a relay section of 250 km and access section of 10 km, with some bands excluded.

In recent years, with the widespread use of the Internet, the number of malicious denial-of-service (DoS) attacks on network servers has increased. Malicious traffic reaches the server through multiple routers, which results in heavy load on many links; thus, load suppression of malicious traffic is important. Therefore, we propose a throughput suppression control scheme using multiple communication routes obtained by ant colony optimization (ACO). The ACO algorithm is used to distribute the routing path. The effectiveness of this proposed scheme is confirmed through computer simulations, which shows that the throughput can be adjusted by the selection probability of the longest route and the number of available routes.

We propose a new model that can accurately estimate the maximum output electrical power of an optical power supply system with low optical input power. The model includes dependence of open-end voltage on optical input power. The proposed model is compared with a conventional model that ignores the dependency. Our model shows that the maximum output power of an optical power supply significantly decreases as optical input power becomes weak.

Telephone scams called “special scams” have become a major problem, with people being defrauded of nearly 30 billion yen every year in Japan. In this study, we used emotion analysis techniques based on speech analysis to detect whether victims were being deceived by analyzing multiple emotional parameters of victims in a special scam. In the verification, we first clarified the characteristics of the emotions generated in victims by two scam methods. Next, we improved the accuracy of detecting the risk of a victim being deceived by limiting the analysis section.

Amazon EC2 provides Spot Instances that may change prices with time. To use them more efficiently, it is necessary to understand the Spot price fluctuations. This study evaluates the Spot price fluctuations with normalized values by On-Demand prices and Historical Volatilities (HVs). Then, using evaluation results, we classify the Instances into five groups. Each group has features of the Spot price fluctuations. We conclude that the groups that have a small fluctuation range are able to bid easily. Therefore, for the groups that have a large fluctuation range, we need more detailed evaluation standards.

Blood pressure (BP) is one of the four main vital signs providing valuable medical information about the cardiovascular activity. In this paper, a new method for estimating systolic and diastolic BP in contact-free for comfort is proposed using a millimeter-wave (MMW) sensor. The MMW sensor is sensitive to small cardiac motion of 1mm or less included in chest surface motion because of the higher frequency and wider bandwidth available, thereby continuous systolic and diastolic BP per heartbeat would be able to be estimated from the time-domain features of unique cardiac motion waveform. Measurement was conducted with the MMW sensor system and the estimated systolic and diastolic BP are also compared with an intermittent BP monitor. As a result, the sensor system is found to estimate systolic and diastolic BP continuously.

The quality of experience (QoE) is an index that evaluates the application-level quality of network services. In adaptive video streaming, stabilization of playback buffer level improves the QoE by avoiding video freezing. Although control-theoretic approaches for stabilizing the buffer level are effective to improve the QoE for a single user, they have not considered user preferences of the QoE or bandwidth allocation for multiple users. This study proposes a dynamic bandwidth allocation method that satisfies the QoE requirements of multiple users with different preferences. Simulation results show that the proposed method increases the number of satisfied users and QoE fairness for prioritized users at the expense of the average QoE for all users.

We have been considering applying discrete wavelet transform (DWT) to heart-rate estimation using 77GHz band FMCW radar. The detection accuracy and SNR were improved by using an elliptic High-pass filter (HPF) and a High-pass Chebyshev type II filter. The High-pass Chebyshev Type II filter has less ripple in the passband area, on the other the elliptic HPF has a steeper frequency response compared to the Chebyshev filter. Those filters were used in a pre-processing of DWT for the data acquired from the front and back to reduce respiratory signals and harmonics for comparison. In this letter, the detection accuracy and SNR estimated from the front and back were improved by using two types of filters in DWT pre-processing to decrease respiratory signals.

Radio over fiber (RoF) technology, especially intermediate frequency over fiber (IFoF), is attracting much attention as one way to realize high frequency band systems. Such systems require beamforming to attain the desired link budget. Furthermore, highly accurate beamforming permits the adoption of space division multiplexing (SDM), with its greater system capacity. Since beamforming should be controlled by the central station to achieve simple base stations, we proposed a novel remote beamforming scheme with fixed wavelength allocation. This paper introduces experiments conducted on modulated signals to evaluate the characteristics of the scheme’s beamforming gain and signal to interference ratio (SIR).

We propose a simple channel-allocation method based on tug-of-war (TOW) dynamics, combined with the time scheduling based on nonlinear oscillator synchronization to efficiently use of the space (channel) and time resources in wireless communications. This study demonstrates that synchronization groups, where each node selects a different channel, are non-uniformly distributed in phase space such that every distance between groups is larger than the area of influence. New type of self-organized spatiotemporal patterns can be formed for resource allocation according to channel rewards.

With the proliferation of automotive electronic devices, electronic control units are increasingly used to share information within in-vehicle networks. This induces a range of security risks, such as information disclosure, spoofing, and tampering. In this paper, we propose a symmetric-key cipher. The method generates pseudo-random numbers using a chaotic and random neural network, and encrypts and decrypts frame messages of in-vehicle networks based on the symmetric key. We also propose a lightweight ID-based key sharing protocol. We evaluated the key sharing, encryption, and decryption in Controller Area Network.