The purpose of this study is to make software of equipment that detects sound of ambulance and to program it on a microcontroller in the equipment. By using two stages FFT (Fast Fourier Transform) on detecting, the characteristics of the sound could be converted into numerical values. The proposed method could detect the sound in which S/N ratio is approximately 0[dB]. The sound was detected properly even under the Doppler effect.

抄録全体を表示

This paper reports a spatial extrapolation of the sound field with a physics-informed neural network. We investigate the spatial extrapolation of the room impulse responses with physics-informed

architecture. Furthermore, we proposed a noise-robust extrapolation method by introducing a tolerance term to the loss function.

抄録全体を表示

In 2017, we proposed a new

system of the ambulance for early perception of its existence or approach. The system has four speakers, two on the front and one on each body side. When passing through an intersection or changing a lane, the ambulance crews produce a rapid increase in sound pressure by turning on the side speakers. Since the blows from four speakers simultaneously, the sound interference occurs, that is, the sound pressure differs according to location. To produce a larger increase of sound pressure in wide area, we optimized the installation location of the side speakers under some legal constraints via the computer simulation. The simulation was conducted for the sound frequency of 770 Hz and 960 Hz with the FDTD method. As a result of the simple line search, the optimal installation location was 0.5 meters back from the front. By comparing the sound pressure level for four-speakers sounding with that for two-speakers sounding, we confirmed that the sound pressure effectively increases around the body side.

抄録全体を表示

In order to operate an ambulance efficiently, we developed a Smart Ambulance Approach Alarm System using smartphone, by notifying the approach of an ambulance to other vehicles on public roads. The position information of ambulances has not been opened in view of development costs and privacy protection. Therefore, our study opens the position information inexpensively by loading commodity smartphones, not special devices, into ambulances. The position information is made to be open as minimum necessary information by our developed cloud server application, considering dynamic state of other vehicles on public roads and privacy of ambulance service users. We tested the functional efficiency of this system by the demonstration experiment on public roads.

抄録全体を表示

An automobile is a convenient means of transportation for hearing-impaired persons to participate in society because it is less burdensome than public transportation. When driving a vehicle, a hearing-unimpaired person can obtain information from both hearing and vision, however, it is difficult for a hearing-impaired person to obtain auditory information without using a hearing aid. There are many demands, especially to be able to know the approach of an emergency vehicle, while auditory information is not sufficiently obtained. Therefore, the purpose of this study is to investigate the actual situation of the of an emergency vehicle when driving a vehicle for hearing-impaired persons, and to understand the current situation and problems from the viewpoint of consciousness. As a result of a questionnaire survey, more than half of the people have not noticed the of an emergency vehicle. However, about 70 % of the hearing-impaired persons do not think of it as a hearing problem. Also, about half of them feel that it is more difficult to recognize visual information during nighttime driving than during daytime. Hearing-impaired persons may have difficulty in obtaining visual information in the dark, and if the amount of auditory information decreases, it may be more difficult for them to collect information around them, and driving may be difficult.

抄録全体を表示

Communication is a primary challenge in response to natural and man-made disasters. The purpose of this study is to determine the forms of communications that played a critical role in disaster preparedness during the 2011 Great East Japan earthquake and tsunami (Tohoku earthquake). The study used a case study research method and included six informants following the purposive sampling technique. Data was collected through unstructured in-depth interviews. The findings suggest that communication in all forms played a vital role in disaster preparedness. In the survivor’s view, however, the impact of broadcast media in establishing situational awareness was limited. Local disaster management efforts, such as, “on call,” or emergency measures, such as, personnel and trucks with communication systems, mobile phone alarms, and sirens were viewed by survivors as useful in transmitting unambiguous, specific messages in a demanding tsunami context. In particular, the study demonstrates that residents’ trust in risk/disaster related messages and mediums play vital roles in subsequent behaviour/response to natural disasters.

抄録全体を表示

The acoustics of combustion chambers and the interaction of acoustics and combustion is investigated in a model combustor operated with LOX/H₂ and LOX/CH₄. Acoustic excitations are induced by a during hot fire tests and the response of atomization and combustion is recorded with dynamic pressure sensors and high speed OH-imaging is applied. By analyzing the temporal and spatial distribution of the flame response the question is addressed, whether in the experiments the coupling of acoustics to combustion is via pressure or velocity sensitive processes. In the experiments it appeared that the acoustic eigenmodes of the combustor are characteristically deviating from cylinder modes. The results obtained can be explained by a modal analyses of the combustor geometry. These investigations have been extended to study the influence of absorbers and absorber rings on the acoustics of combustion chambers and resonance frequencies predicted by modal analysis and experimental results are presented.

抄録全体を表示

At the Great East Japan Earthquake May 11th 2011 , many peoples died by the TSUNAMI. We think that one of cause of it is on inadvisability of evacuation guidance. We are feeling the necessity for an effective evacuation guidance sound which all peoples feel necessity to escape immediately if hear it. Therefore we try to evaluate sounds to feel necessity to escape. Sin wave, Triangle wave, Saw-tooth wave, Square wave and wave were evaluated as test sound. all test sound sweep 2 octave or 4 octave. The n・equency and band width were evaluated too. Scores of 'feeling of escape' of triangle wave and square wave were higher the other wave forms. Wide frequency band made to feel necessity of escape strong.

抄録全体を表示

A questionnaire survey was administered by Tohoku University hospital to investigate the current status of gynecological laparoscopic operations in teaching hospitals in which residents were trained as obstetricians and gynecologists. Thirty-eight senior doctors in 18 hospitals were sent a postal questionnaire. The response rates of facilities and doctors were 94.4% and 76.3%, respectively. Diagnostic and operative laparoscopy was performed in all of the hospitals surveyed, however only a few hospitals performed more than 100 laparoscopies per year, and many surgeons did not perform advanced laparoscopic operations including hysterectomy, myomectomy and lymphadenectomy. Half of the doctors performed minilaparotomy, which is a so-called “open” technique. The standard surgical style involved the use of a carbon dioxide (CO₂) pneumoperitoneum and an endoscope 10 mm in diameter. Unipolar endocoagulating instruments were generally used for hemostasis and incision. Most of the doctors surveyed thought that operative laparoscopy would become a necessary procedure in gynecological field, and half of them hoped to receive training in the technique. Instruction of expert endoscopic surgeons is necessary, especially in teaching hospitals, for the spread of safe and advanced laparoscopic surgery. An educational and training system for established surgeons as well as for young residents is urgently needed in Japan.

抄録全体を表示

Because of increasingly stringent regulations on jet engines pollutant emissions, new ways of reducing them are needed. A promising way is the use of Lean Premixed Prevaporized (LPP) combustion where the liquid fuel is vaporized and mixed with air in excess before burning. In the present work, a model gas turbine combustor fed with liquid dodecane is studied experimentally. It is equipped with two fuel injection stages: a pressurized nozzle called the pilot and a multipoint device. When the split of the fuel injection between the stages is changed, several phenomena can be observed. The flame shapes can change drastically and present different behaviors, some of them showing a strong acoustic activity. In particular, two different flame shapes can be obtained for the exact same operating conditions depending on the burner history. The first flame, named sV, is stabilized thanks to an internal reaction zone that modifies the air flow compared to non-reacting conditions. The second state, associated with a Lifted flame, exhibits a strong thermo-acoustic instability linked to the quarter-wave mode of the chamber. A switch from the sV to the Lifted state can be triggered by air flow rate modulations while the opposite change can occur naturally as the fuel split is changed. These complex phenomena originate from multiple interactions between the fuel spray, the gaseous flow and the flame itself.

抄録全体を表示

この研究では、救急車と一般の車両の事故を防止する救急車のサイレン音検出装置を製作することを目的とする。特に、救急車のサイレン音を検出する装置に、検出アルゴリズムを実装することを目的とする。検出方法はまず、マイクからシリアル通信を用いて音声を取得する。そのデータを、2段のFFTを用いて周波数解析を行い、音程周波数と断続周波数を解析する。解析結果からサイレン音の特徴の有無を判断し、サイレン音があればドライバーに通知する。実験により、S/N比0dBでも検出が可能であること、ドップラー効果が加わっても検出が可能であることを示した。

抄録全体を表示

Conciderable skill is needed to detect traffic accident sound which was recorded in the intersection. In this paper, we proposed detection method of accident sound data using time-frequency analysis and correlation method in a short time.

抄録全体を表示

There is a growing need to introduce warning dissemination systems in disaster-prone regions to improve the coverage of information distribution. In this study, a warning dissemination system was designed in which disaster information transmitted by a global navigation satellite system (GNSS) is received by terrestrial infrastructure, such as sirens and public transportation, converted into audio messages, and delivered automatically. The originality of the designed system lies in its appropriate integration of existing satellite systems and terrestrial infrastructure, making the system potentially applicable in many regions. First, we evaluated the effectiveness of the designed system in distributing audio messages using public buses in Brisbane, Australia, where large floods occur frequently. Real-time location information for public buses was acquired in the format of General Transit Feed Specification (GTFS), which is currently used in many countries. Time-series changes in the coverage rate relative to both the flood inundation zone and population were calculated using a geographic information system (GIS). The simulation results showed that the system could reach 60% of the flood inundation zone and 70% of the population on a holiday, indicating that the designed system could be effectively adapted to the target area. The coverage rate was found to peak during 15:00–16:00, with minimum rates observed late at night and early in the morning. These results will allow the development of an effective disaster management plan. In the future, this system will be evaluated in other regions using the same calculation process.

抄録全体を表示