No abstract

No abstract

Measurement and characterization of acoustic noise in water pipeline systems across the 10 Hz to 100 kHz acoustic frequency band are reported in this paper. The motivation for the noise characterization is to support the development of acoustic signal processing and communication systems for water pipeline systems. Measurements are reported from an urban water supply system in the city of Christchurch, New Zealand, using hydrophones inserted into the pipelines. These measurements are some of the first that have been reported for an operational water supply system. In addition, for comparison to a controlled baseline reference, measurements have also been performed in a 160 m pipeline laboratory testbed at The Hong Kong University of Science and Technology. A key finding is that the noise amplitude probability density function (PDF) is found to follow a heavy-tailed distribution and the α-stable distribution is proposed for its characterization. The resultant acoustic noise measurements have also been analyzed to determine their power spectral density (PSD) characteristics. The measured noise PSD is not flat and characteristic frequencies and power density levels are utilized to capture the features of the noise PSD. These findings could have a significant impact on the design of signal processing systems for water pipeline systems due to their deviation from the conventional added white Gaussian noise model. In addition, three different acoustic noise sources with individual power and distribution characteristics are also postulated. These postulated sources are used to propose a straightforward acoustic noise model for supporting the simulation of acoustic noise for signal processing system design in water pipelines.INDEX TERMS Acoustic noise, amplitude probability density function, noise power spectral density, water pipeline channels. I. INTRODUCTIONThe characterization and modeling of channel noise is an essential component in the development of signal processing and communication systems. Noise determines the lower power limit at which signals can be successfully received and also sets the criteria for the optimum design of signals. Key characteristics of the noise, such as its power spectrum density (PSD), amplitude distribution, stationarity and spatial correlation are required for the optimum design of signal processing and communication systems.The associate editor coordinating the review of this manuscript and approving it for publication was Jingchang Huang.