Cross-Layer Analysis of Rate Adaptation, DCF and TCP in Multi-Rate WLANs

摘要: Wireless Internet access is facilitated by IEEE 802.11 WLANs that, in addition to realizing a specific form of CSMA/CA-distributed coordination function (DCF)- implement range performance enhancement features such as multi-rate adaptation that induce cross-layer protocol coupling. Recent works empirical WLAN evaluation have shown interactions can be subtle, sometimes leading unexpected outcomes. Two instances are: significant throughput degradation (a bell-shaped curve) resulting from automatic rate fallback (ARF) having difficulty distinguishing collision channel noise, and scalable TCP over DCF able curtail effective multiple contention the presence many contending stations. The latter also mitigates negative effect ARF. In this paper, we present station-centric Markov chain models aimed at capturing complex between ARF, DCF, TCP. Our analyses may viewed multi-protocol extensions Bianchi's model despite significantly increased complexity, lead tractable accurate predictions due modular results complement simulation-based findings, demonstrating versatility efficacy analysis for dynamics.