Open Collections

Improving and bounding asymptotic approximations for diversity combiners in correlated generalized rician fading

University of British Columbia

Although relatively simple exact error rate expression are available for selection combining (SC) and equal gain combining (EGC) with independent fading channels, results for correlated channels are highly complex, requiring multiple levels of integration when more than two branches are involved. Not only does the complexity make numeric computation resource intensive, it obscures how channel statistics and correlation affect system performance. Asymptotic analysis has been used to derive simple error expressions valid in high signal-to-noise ratio (SNR) regimes. However, it is not clear at what SNR value the asymptotic results are an accurate approximation of the exact solution. In this thesis, we derive asymptotic results for SC, EGC, and maximal ratio combining (MRC) in correlated generalized Rician fading channels. By assuming generalized Rician fading, our results incorporate Rician, Rayleigh, and Nakagami-m fading scenarios as special cases. Furthermore, the asymptotic results for SC are expanded into an exact infinite series. Although this series grows quickly in complexity as more terms are included, truncation to even two or three terms has much greater accuracy than the first (asymptotic) term alone. Finally, we derive asymptotically tight lower and upper bounds on the error rate for EGC. Using these bounds, we are able to show at what SNR values the asymptotic results are valid.

Text

2013-02-18

Attribution 3.0 Unported

http://hdl.handle.net/2429/43932

Electrical and Computer Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by/3.0/