Indoor acoustic mapping and geolocation

Thomas Shyy, Purdue University

Abstract

The Global Positioning System (GPS) is widely used in navigations of flights, vehicles and travelers nowadays. However, the signals are not always available and there are some cases where the receptions of the GPS signal can be lost because of the presence of urban canyons, buildings, tunnels and forests. Another down side for the GPS is that indoor positioning can be really difficult because of the greater signal interference from the indoor. On the other hand, the concept of echolocation has also been used for positioning purposes. Echolocation offers a promising approach to improve the quality of life of people with blindness using the "searching" and "scanning" strategies. Scanning consists of head movement from a person to detect the object while searching relies heavily on information within the echo rather than headmovement. Motivated by the concept of echolocation positioning technique for blinds, we have developed a model system to estimate the geolocation within our designated testing facility, an echo chamber room. The echo chamber is designed to have its surrounding to have the least acoustic impact absorbance to allow the least reduction of refelcted sound waves' intensities. In the room, a sound speaker is used as the sound source to carry the pulse wave we created to represent the "snap" from blinds. During the measurement, the position of our sound source always stays stationary while a sound receiver equipped with 64 measurement microphones is collecting the sound intensities for different sound waves from the source. We have chosen sixteen receiving positions to represent the room due to the limitation of the our cords' length of the receiver. Measurements will be performed one position at a time for the 16 positions. For every microphone, the sound intensities from the source are recorded and displayed in plots at MATLAB with sound intensity (V) on the y axis and time (s) on the x axis. On the data plots, we are to find the first seven major peaks on the y axis and more importantly, the corresponded elapsed time on the x axis. The focuses for our study are on direct sound waves and first order sound waves. The first peak, with the least elapsed time, will be the direct wave which travels directly from the source to the microphone. The second to the seventh peaks are all first order waves which reflect once from the source to the microphone. Since the echo chamber is rectangular shaped that consists of six planes, there are going to be six first order waves. The sound source is set as the origin for the measurement coordinate system. The relative positions of the receiver and the microphones to the source are measured. One simple assumption is made for the study assuming that the travelled distance of a single acoustic wave is going to be the multiplication of the speed of sound and the travelled time because the source is a simple pulse wave. By applying the mirror source method, we are able to develop a preliminary mathematical model to estimate the geolocation within the coordinate system with the data information acquired. For the model, we are capable of estimating the length, width, and height of the room with the known coordinates of the source and receiving points; and the travelled distances for the direct and first order acoustic waves. The average error percentage of the estimation from the preliminary model is around 15 percent compared to the actual measured dimensions. To improve the accuracy of the theoretical model, the method of least square is applied using all the data points to improve the model for better error percentage. Our geolocation estimating system is limited in two conditions: The position of the sound source remains constant; it is applicable for indoor environment. For future development and study, the model can be modified to adapt with non- constant source and greater influences by the outdoor noises.

Degree

M.S.M.E.

Advisors

Ariyur, Purdue University.

Subject Area

Mechanical engineering|Computer science

COinS