Abstract:
Due to di erent atmospheric conditions, signals transmitted through wireless medium
experience several distortions. As a result, the received signal randomly gets attenu-
ated in amplitude and distortion in phase. Therefore, irregularity is observed in received
signal strength. The performance of the wireless communication system gets degraded
in fading environment, relative to the additive white Gaussian noise (AWGN) environ-
ment. The mobile user experiences di erent fading scenarios, which can be characterized
as multi-path fading, shadow fading and mixture of both (composite fading). There are
some known fading models such as Rayleigh, Rician, Nakagami-m, Weibull and composite
fading channel (Rayleigh/Log-Normal (R/L), Rician/Log-Normal (Rice/L),Weibull/Log-
Normal (W/L)or Nakagami/Log Normal (N/L)).
In this dissertation, di erent fading scenarios are discussed and analyzed, and the
corresponding probability distribution function (PDF) has been analysed. The Gaussian
Q function has been approximated using the trapezoidal rule.
The computational complexity involved in the calculation of the double integral in the
expression of SEP for composite N/L channel is very high. Therefore, two closed form
approximations for the PDF of the composite N/L channel are obtained using the approx-
imations proposed by Holtzman and Gauss Hermite. Further, these two approximations
are incorporated in the expression of SEP to reduce the computational complexity. In
order to evaluate the performance of these two approximations, they are compared with
the exact SEP expression. It is found that SEP result from the approximations proposed
by Holtzman and Gauss Hermite are very close to the exact SEP obtained for low and
high SNR, respectively.
Further, the performance of the two approximations are evaluated in a combined (time
shared) N/L and Rician fading channel. Afterwards, the sensitivity analysis is carried
out to study the e ect of Rician k factor, Nakagami-m factor and time shared factor in
the combined channel.
