1. GSU Indoor Navigation Senior Project Fall Semester 2013 Michael W Tucker

2. Development Development: Maya: = Sample Building Layout for Navigation GUI Development: Photoshop, Android SDK= Languages: Java, Android operating system IDE: IntelliJ IDEA12.1.6, Eclipse

3. Purpose I am attempting to build an indoor navigation application. The GSU application app would primarily be used by incoming students that tend to have a hard time finding their way around the campus or specific buildings. The application would then pinpoint the students location using their device, launch a 3d render of the building, followed by guiding them to the assigned classroom using a built in navigation system.

4. How Indoor Navigation Works and why it’s so difficult. An indoor positioning system (IPS) is a network of devices used to wirelessly locate objects or people inside a building. Instead of using satellites, an IPS relies on nearby anchors (nodes with a known position), which either actively locate tags or provide environmental context for devices to sense. The localized nature of an IPS has resulted in design fragmentation, with systems making use of various optical, radio, or even acoustic technologies. GPS does not work indoors. The satellite signals are not strong enough to penetrate any building a student may enter. All modern smartphones have Wi-Fi built in, and wireless networks are common enough in indoor spaces that an app could easily scan for known access points and calculate your position using trilateration. Unlike the wide open world of Android, developers on the iPhone side aren’t allowed to perform these Wi-Fi “signal scans.” Fortunately, there are alternatives. One approach is to make the building do the work instead of the device. Some Wi-Fi installations, such as the Cisco MSE, can determine the location of any wireless device in the building. The access points themselves listen for the Wi-Fi signals created by your phone, then estimate its position via trilateration. This solution has been deployed successfully at a few locations, including at the American Museum of Natural History.

5. GSU Navigation Prototype #1 Idea was scrapped

6. GSU Navigation Prototype 2

7. GSU Navigation Prototype 2 Continued

8. GSU Navigation Functionality There is an open source application called “Red pin“. Redpin consist of two basic components: a sniffer component that gathers and collects information about different wireless devices in range in order to create a fingerprint, and a locator component, which stores measured fingerprints in a repository and contains the algorithm to locate a mobile device. To allow easy, reliable and fast user collaboration, the locator is run as a central server. It is implemented using Java SE 6.0 and SQLite (or MySQL). The sniffer component the system uses is an (iPhone, Android) application that gathers information about different wireless devices in range. Interval Labeling for the Redpin Client application works as follows. When a user adds a new location, the current location is stored, and an interval scanner is instantiated. The interval scanner runs in the background, and periodically performs a scan of Wi-Fi networks, creating fingerprints by attaching the current location to the new Wi-Fi measurements. The fingerprint is then being sent to the server and is stored there. Interval scanning is stopped whenever a significant movement is detected through the built-in (iPhone, Android) accelerometer, or when the users try to either add a new location or locate themselves.

9. Additional “Redpin” info Instead of relying on a static fingerprint model, Redpin does not make a difference between training and usage. Redpin can be used right away and allows its user to enter the symbolic labels if necessary. To further increase the accuracy and the adaptability of the locator, Redpin allows multiple measurements taken consecutively to be added to the same fingerprint. We call this process 'Interval Labeling'. In addition, Redpin employs a two stage positioning mechanism: if fingerprints consist of only a few measurements, Redpin uses a kNN variant while SVM is used for big fingerprints.

10. GSU Navigation Flow Chart Launch GSU Application Log in New User Student Info Needed Class Entry/ Lookup Submit Indoor Navigation Directions List Submit Classes Displayed Previous Indoor Navigation Active Directions Text Based Displayed Email Directions Indoor Navigation Stop Return Classes Displayed Multiple Classed Stored Stop and Close GSU Navigation App Stop Return Classes Displayed Previous

11. GSU Flow Chart Continued

12. GSU Navigation System Example: of the RedPin Engine GSU Navigation GSU Text Navigation Red pin would run underneath the GSU app when the navigation is launched.