Dynamic road traffic density estimation employing noise mapping with the use of grid supercomputing

Abstract

A noise prediction model of a large city agglomeration was elaborated in order to allow for a dynamic road traffic density estimation in vehicular networks. The implemented application adopts the model fed with traffic noise data based on frequently refreshed LDEN levels. Calculations were made with the use of the numerical model developed for his purpose and then implemented on the PL-Grid supercomputing infrastructure. Data obtained through supercomputing and through the use of a standard noise map computing software were collated with measured levels acquired from the acoustic city monitoring system and then analyzed. The comparison performed afterwards shows a relatively good accuracy of the developed model. The numerical model of traffic noise and its main sources are briefly characterized. A full day dynamic noise map can be browsed as a set of 24 noise maps, one for each hour of the day which in turn allows for vehicular traffic density estimation based exclusively on acoustical data. [Research was subsidized by the Polish National Centre for Research and Development within the Grant No. OT4-4B/AGH-PG-WSTKT.]

A novel quantum-inspired solution for high-performance energy-efficient data acquisition from IoT networks

“…It is important to mention that only 4-Wheeler like buses, cars, and trucks were monitored for acquiring data sets. The reference range of noise was considered from 60 to 90 decibels and for vibrations, a range of 1.2 mm/seconds to 1.7 mm/seconds was referred (Czyzewski et al (2016)). Figure 8 depicts the hardware devices used for experimental implementation.…”