Performance of N^th-best user selection in multiuser diversity systems with co-channel interference and outdated channel information

In this paper, we present the N^th-best user selection scheme to multiuser diversity systems and evaluate its performance in the presence of co-channel interference (CCI) and outdated channel information (OCI). This scheme is efficient in OCI conditions where the user which was the best at the selection time instant could not be the best at the transmission time instant. Also, it is useful when the scheduling unit fails in error in selecting the best user among all the available users. Furthermore, the studied scheme is efficient in situations where while the best user is waiting to be scheduled by a certain scheduling unit, it gets scheduled by other unit. In the analysis, we first obtain the statistics of the end-to-end (e2e) signal-to-interference plus noise ratio (SINR) and then use it to derive closed-form expressions for the e2e outage probability and symbol error probability. The ergodic channel capacity is also numerically calculated in this paper. All links are assumed to follow Rayleigh distribution. Also, the asymptotic behavior of the system is studied and the diversity order and coding gain are derived. Monte-Carlo simulations are provided to validate the achieved analytical and asymptotic expressions. Main results illustrate that with fixed interference power, having more desired users can still enhance the system performance via increasing the diversity order. Moreover, findings show that the diversity order of the system linearly increases with decreasing the order of the scheduled desired user, and vice versa decreases. Finally, results illustrate that the higher the correlation coefficient between the signal-to-noise ratios (SNRs) of users at the selection and transmission time instances is, the better the achieved performance.