Browse Prior Art Database

An ad-hoc, mobile early-warning system using inexpensive networked sensors using sound monitoring and triangulation

IP.com Disclosure Number: IPCOM000241292D
Publication Date: 2015-Apr-14
Document File: 5 page(s) / 88K

Publishing Venue

The IP.com Prior Art Database

Abstract

Today’s applications need to detect any kind of warning and error when anything could go wrong, which could cause a problem related to noise or vibration. Those applications domains include, but is not limited for example to cars, trains or bridges. Instead of using build-in expensive and not flexible sensors the concept proposed is to use networked mobile sensors for sound monitoring. First the sensors will learn the normal behavior and then are ready to detect abnormal (warning, error) conditions. The networked sensors combined with sound triangulation techniques will provide a location of the source of the noise irregularities.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 35% of the total text.

Page 01 of 5

An ad- -hoc

hoc, ,

mobile early

mobile early -

-warning system using inexpensive networked sensors using

                warning system using inexpensive networked sensors using sound monitoring and triangulation

The issue addressed: Many application scenarios require warnings and errors to be issued when things go wrong. Very often such warnings are not available, depend on system components to be interconnected and able to detect it when they are operating outside of normal parameters and require some sort of location information to be provided by the component so the cause of the irregularity can be located and remedied.

Many systems and system components generate some sort of operating noise that is transmitted through air, machine components or building structures. Such characteristics often change when operating parameters are no longer within the standard range. Examples: fan and disk noise of computers (louder fans when overheating or fans go bad), gear/friction noises in cars and other machines, sounds of

wildlife eating crops in a farmer's field, vibrations in building structures due to overload conditions or because a state of dangerous self-oscillation / natural frequency oscillation was reached/induced.

Noise cancelling and sound dampening often prevent the operators to notice subtitle changes. e.g. in cars. In the older days it was very apparent for the operators to hear

when some engine or transmission component became worn and was about to fail but

with noise reduction and noise dampening is becomes more difficult for a human operator to notice. Sound and vibration sensors distributed in a machine/car can detect such changes much earlier.

The idea to use sound or vibrations to detect some behavior or phenomenon is of course not new. For example seismic activity detection (tsunami warning) is well known. Intrusion detection through sound or other sensors is also a common usage and finally the monitoring of specific mechanic components (e.g. jet engines or train wheels to generate alarms before failure) exists today. One key differentiator is that in all these applications the sensors are often "designed into the component" and become part of the machine. They cannot be re-located, are hardwired, cannot be deployed in an ad-hoc fashion and hence are difficult/expensive to deploy and cannot easily adapt to new environments or conditions.

In an extreme case the proposed system would be implemented in a mobile smart phone application. It would be carried by a user to different places, detect its location (through well known smart phone mechanisms) and monitors the noises at the current location and compares them to a profile. For example in the car, in the data center, at home in the kids bedroom (baby phone), .... whenever the noise deviates from a previously learned profile or when very distinct sounds are recorded an action can be triggered. The application is in contact with other applications (on other mobile smart phones that ar...