AAU Student Projects - visit Aalborg University's student projects portal
A master's thesis from Aalborg University
Book cover


Localization Information Retrieval Exploiting Cooperation Among Mobile Devices

Author

Term

10. term

Publication year

2008

Pages

80

Abstract

Determining a device’s location in cellular networks has become increasingly important. Many existing methods work best with a clear line of sight between the base station and the phone, so accuracy drops in dense urban areas. GPS is an attractive option, but it needs a clear line of sight to satellites, can take several minutes to obtain the first position fix (time-to-first-fix), and adds both cost and battery drain to a handset. This project explores a GPS-independent approach where mobile devices cooperate to determine their geographic position. The method is fingerprint-based: it uses patterns in Cell ID information (identifiers of nearby cell towers) to narrow down where a device is likely located. In this context, Google has launched MyLocation; we enhance this service by adding cooperation among devices. We developed a prototype system, discussed its limitations, and suggested possible improvements. The main result is that with three cooperating phones, the estimated area in which the device is located can be reduced by more than a factor of five compared to a non-cooperative setup. With three cooperating devices, the distance between the GPS position and the center of the estimated area is about 70 m, and the extent of the estimated area is about 145 m.

At bestemme en enheds position i mobilnet er blevet stadig vigtigere. Mange eksisterende metoder virker bedst, når der er fri sigtelinje mellem basestation og telefon; derfor falder præcisionen i tætbebyggede byer. GPS er et attraktivt alternativ, men kræver fri sigtelinje til satellitter, kan bruge flere minutter på at få første positionslås (time-to-first-fix), og øger både omkostninger og batteriforbrug i en telefon. Dette projekt undersøger en GPS-uafhængig metode, hvor mobile enheder samarbejder om at finde deres geografiske position. Tilgangen er fingerprint-baseret: den udnytter mønstre i Cell ID‑oplysninger (identifikatorer for nærliggende mobilmaster) for at afgrænse, hvor enheden sandsynligvis befinder sig. I denne sammenhæng har Google lanceret MyLocation; vi forbedrer denne tjeneste ved at tilføje samarbejde mellem enheder. Vi har udviklet et prototypsystem, beskrevet dets begrænsninger og foreslået mulige forbedringer. Hovedresultatet er, at når tre telefoner samarbejder, kan det anslåede område, hvor enheden befinder sig, reduceres med mere end en faktor fem sammenlignet med ingen samarbejde. For tre samarbejdende telefoner er afstanden mellem GPS-positionen og centrum af det anslåede område omkring 70 m, og udstrækningen af det anslåede område er omkring 145 m.

[This apstract has been rewritten with the help of AI based on the project's original abstract]