Change search
ReferencesLink to record
Permanent link

Direct link
Indoor mobile classification and coverage analysis
2009 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

As a broader market of high bitrate services is introduced within the mobile radio network, an increasing fraction of usage will be indoor. This drastically changes the demands on the network infrastructure, which will need to handle both indoor and outdoor users at higher bitrates. Since indoor users in general experience worse signal conditions than clients located outdoors, differences between usage in typical indoor (obstructed by building material) and outdoor (unobstructed clear sky view) environments become highly interesting. Therefore, methods to classify measurement samples as typical indoor are needed. A study of present indoor and outdoor signal coverage has been performed. Data collected by TEMS measurement tools was used as a training set for discriminant function classification of clients as experiencing typical indoor or outdoor signal conditions, respectively. It was concluded that the best variable for classification is RxLev (Received signal level) for cells within the macro cell layer, normalized with respect to an approximated client to antenna tower distance. Using the identified classification procedure, data samples of commercial traffic obtained by RPMO (Real-time Performance Monitoring) of two different networks - one within the center of a large city and one much less dense - were analyzed. In both regions, typical indoor environment usage was found to represent 40% of the traffic. This excludes in-building locations where signal conditions are more outdoor-like, such as high structures or buildings with large sections of glass. The total current use of mobiles from within buildings in the center of the investigated city was estimated to be about 60%. Studying pico (indoor) cell classification results more closely, three typical groups of building types were identified among the studied data: Concrete, glass and high buildings (ordered by descending fraction of typical indoor environment classification). Clients within concrete structures were classified as experiencing a typical indoor environment, while clients in high buildings were found to experience signal conditions very similar to those experienced by a typical outdoor client. Usage conditions in buildings with large sections of glass in their exterior are in general in-between the two extremes. Also, micro cells were verified as handling mainly typical outdoor traffic. The average improvement in typical indoor environment signal strength resulting from reinforcement by pico cell installation was found to be 12dB. It was also predicted that average RxLev in the network, given the indoor usage of today, could be increased by up to about 4dB by improving coverage in typical indoor areas by solutions similar to pico cells (from the current 8% up to 100%). Conversely, the current network average RxLev, keeping the indoor infrastructure at the level of today, would be reduced by 5dB if typical indoor traffic were to increase from 40% to 70%, which is the current proportion of in-building pico cell data classified as experiencing a typical indoor scenario. It was also concluded that in areas with large cells, a denser macro network could improve coverage in typical indoor environments more efficiently than specific indoor solutions.

Place, publisher, year, edition, pages
Keyword [en]
Physics Chemistry Maths, indoor, mobiles, radio, networks, signals, coverage
Keyword [sv]
Fysik, Kemi, Matematik
URN: urn:nbn:se:ltu:diva-44080ISRN: LTU-EX--09/190--SELocal ID: 1e2ec86c-a771-4bbc-b3de-3beeafa1f645OAI: diva2:1017355
Subject / course
Student thesis, at least 30 credits
Educational program
Engineering Physics, master's level
Validerat; 20101217 (root)Available from: 2016-10-04 Created: 2016-10-04Bibliographically approved

Open Access in DiVA

fulltext(1913 kB)0 downloads
File information
File name FULLTEXT01.pdfFile size 1913 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

ReferencesLink to record
Permanent link

Direct link