This thesis presents the contribution to the design of an Ultrasonic Local Positioning
System (ULPS) for expanded coverage employing encoding techniques. One of the main
objectives of the thesis is the design of an ultrasonic beacon to be placed on the ceiling
of an indoor space, by means of a selected commercial transducer, in order to improve
its radiation pattern to increase the area of coverage on the floor. Another objective is
the development of a complete ULPS with encoding and detection signal based on Time
Code Division Multiple Access, which improves the detection capability of the emitted
ultrasonic signal by the beacons, without using an external synchronism between the
mobile and the beacons.
An important feature in any LPS is its coverage. The choice of an optimal transducer is
not an easy task, and more difficult when it has to meet special requirements, such as in
this case a large bandwidth, in order to use spread spectrum techniques. Once a
commercial transducer is selected, it is proposed the design of a conical reflector in order
to improve its emission pattern and to have a larger coverage area. This permits to keep
a certain consistency in the received signal level, so a lower number of beacons can be
used for implementing the positioning system.
The behavior of the ULPS has been theoretically analyzed, and then, it has been
simulated from the mobile standpoint (it has to detect the signals transmitted by the
beacons), considering that ultrasonic beacons transmit continuously at regular intervals.
For the study, several types of noise and different effects introduced by the transmitter,
channel, and receiver have been taken into account. Several time and frequency domain
techniques have been proposed, analyzing their more outstanding features. In addition, a
MAI and ISI interference analysis has been carried out, including usual mitigation
techniques. In order to improve the performance of the system against these
interferences, a new emission technique has been proposed.
For comparing simulated results with real data, a real-ULPS has been developed based
on the mentioned proposals. All the algorithms have been validated by means of
measurements carried out in fixed environment positions and on a moving robot.
|