Spectral Efficiency Analysis in Presence of Correlated Gamma-Lognormal Desired and Interfering Signals

Abstract

Spectral efficiency analysis in presence of correlated interfering signals is very important in modern generation wireless networks where there is aggressive frequency reuse with a dense deployment of access points. However, most works available in literature either address the effect of correlated interfering signals or include interferer activity, but not both. Further, available literature has also addressed the effect of large scale fading (shadowing and distance-dependent path loss) only, however has fallen short of including the composite effect of line of sight and non-line of sight multipath small scale fading. The correlation of desired signals with interfering signals due to shadowing has also not been considered in existing literature.In this work, we present a comprehensive analytical signal to interference power ratio evaluation framework addressing all the above mentioned important components of the model in a holistic manner. In this analysis we extend and apply the Moment Generating Function-matching method to such systems so that correlation and activity of lognormal random variables can be included with high accuracy. We compare the analytical results against realistic channel model based extensive Monte-Carlo simulation for mmWave and sub-6 GHz in both indoor and outdoor scenarios. The performance of the model is depicted in terms of mean, α% outage spectral efficiency and Kullback-Leibler divergence and Kolmogorov-Smirnov distance.Index Terms-Co-channel interference, signal to interference power ratio, sum of log-normal distributed random variables, outage spectral efficiency, correlated lognormal, LOS/NLOS.

Effect of Epidemic Interference on the Performance of M-ASK, M-PSK and M-QAM Modulation Schemes

“…One type of non-stationary interference that occurs in communications systems is the co-channel interference, which has been modeled as a random process with Lognormal power [13]- [15]. In this case, the total interference is formed as the sum of several signals with Lognormal power distribution.…”