Local Coverage Probability Estimation in Single Frequency Networks

  • Authors:

  • Agnes Ligeti;Slimane Ben Slimane

  • Affiliations:

  • Radio Communication Systems Laboratory, Department of Signals, Sensors and Systems, Royal Institute of Technology, 100 44 Stockholm, Sweden E-mail: agnesl@radio.kth.se;Radio Communication Systems Laboratory, Department of Signals, Sensors and Systems, Royal Institute of Technology, 100 44 Stockholm, Sweden E-mail: slimane@radio.kth.se

  • Venue:
  • Wireless Personal Communications: An International Journal
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

In OFDM based Single Frequency Networks (SFNs) such as Digital AudioBroadcasting (DAB) and Digital Video Broadcasting (DVB), the mobilereceiver is provided by the same signal from several synchronoustransmitters at the same carrier frequency.This transmitter diversity structure makes SFNs very efficientin fading channels with an excellent spectrum saving, and very goodpower utilization over the service area.However, depending on the position of the mobile within the servicearea, these signals arrive with different delays causing possibleechoes at the receiver.These echoes contribute to both the useful and the interferingparts of the signal.As the average received power in radio channels is a stochasticprocess, such contribution creates correlation between the differentcomponents of the Signal-to-Interference Ratio (SIR).In broadcasting practice, correlation was not taken into accountwhen estimating the distribution of the SIR and the local coverageprobability.This paper considers the effect of correlation on the estimation ofthe SIR and the coverage probability in OFDM based SFNs.Analyses are provided to show that in some situations the estimationerror can be considerable when neglecting this correlation.To obtain a better estimation, the correlation coefficient between thetotal useful and total interfering parts of the received signal isfirst evaluated and then used as a correction factor in estimating themean and standard deviation of the SIR.The obtained results show that such method can significantly reducethe estimation error of local coverage probability without increasingthe computation complexity.