11 - Asymptotic performance analysis for channel-aware scheduling

Summary

To obtain multi-user diversity gain, adaptive modulation and channel-aware scheduling must be used. However, the channel variance and the opportunistic nature of channel-aware scheduling make throughput analysis very difficult. An asymptotic analysis of signal-to-noise ratio (SNR) for multi-user diversity is presented in [200]. Capacity analyses for Rayleigh and Nakagami fading channels are addressed in [217] and [51], respectively. However, the results of those capacity analyses are too complicated to obtain insights.

In this chapter, we provide an asymptotic performance analysis of channel-aware packet scheduling based on extreme value theory, including throughput and delay analysis for both single carrier and multi-carrier networks. In Section 11.1, we briefly describe the main results of extreme value theory used in this chapter. In Section 11.2, we use an asymptotic analysis of throughput of single-carrier systems with channel-aware scheduling. We first address the average throughput of systems with an homogeneous average SNR and obtain its asymptotic expression. Compared to the exact throughput expression, the asymptotic one, which is applicable to a broader range of channel fading distributions, is more concise and easier to obtain insights. Furthermore, we confirm the accuracy of the asymptotic results by numerical simulation. For a system with heterogeneous SNRs, normalized-SNR-based scheduling needs to be used for fairness. We also investigate the asymptotic average throughput of normalized-SNR-based scheduling and prove that the average throughput in this case is less than that in the homogeneous case with a power constraint. In Section 11.3, we provide a closed-form asymptotic average packet delay analysis for single-carrier networks exploiting multi-user diversity. In Section 11.4, asymptotic analysis of throughput and delay is extended into multi-carrier networks. The asymptotic analysis for mean packet delay demonstrates that the multi-user diversity gain in multi-carrier networks is not limited by slow-fading as in single-carrier networks.