Cross-Layer MAC-Routing Protocol Design for Indoor WLANs with Integrated Channel Modeling under Dynamic Load Conditions

Abstract

High density environments like smart buildings, offices, and other public places are deploying their indoor Wireless Local Area Networks (WLANs) which in turn have problems of user mobility, dynamic traffic demands, as well as the complex propagation of RF networks and can no longer follow standard communication protocols. Performance can be seriously impaired by multipath fading, interference and congestion, in these environments and this is compounded by the strict design of the network protocol layers in older network architectures. This paper brings forward a new cross-layer MAC-Routing Protocol (CLMRP) that will include RF-aware channel modeling and adaptive control techniques that will allow the network in question to work optimally in a dynamic indoor environment. At the physical layer, it uses a lightweight Hidden Markov Model (HMM) to classify in real time the channel (through utilised metrics such as received signal strength indicator (RSSI), signal-to-noise ratio (SNR), and packet success rates). Such forecasts are used to update Medium Access Control (MAC) layer parameters such as body mass index (BMI) to tune contention window parameters (CWmin, CWmax) and minimize collisions and optimize medium access. At the same time, the routing layer adjusts the metric, called Expected Transmission Count (ETX) to take into consideration the real-time queue occupancy along with projected link quality, which leads to an increase in the stability of the path and a decrease in the end-to-end delay. Most importantly, the schema presented proposes to incorporate antenna behavior and the RF front-end properties to the channel prediction and optimization scheme taking into account of the effect of the beamwidth, gain and directional patterns on the estimation of link quality. Results of simulations of the NS-3 environment based on IEEE 802.11 ax PHY models demonstrate that the CLMRP protocol is much more efficient in comparison with other conventional MAC and routing protocols such as IEEE 802.11 DCF with AODV and DSR, that results in 40% increased throughput, 30% decrease in delays and high efficiency in reliability of packet delivery. Such results demonstrate the strengths of integrating RF propagation-aware simulation in a complement with smart cross-layer adaptation, thus identifying CLMRP as a promising solution to the next generation WLAN deployments that need robust and low latency and antenna-enhanced communications.