Paper detail

Stochastic Approximation Algorithm for Optimal Throughput Performance of Wireless LANs

Throughput improvement of the Wireless LANs has been a constant area of research. Most of the work in this area, focuses on designing throughput optimal schemes for fully connected networks (no hidden nodes). But, we demonstrate that the proposed schemes, though perform optimally in fully connected network, achieve significantly lesser throughput even than that of standard IEEE 802.11 in a network with hidden nodes. This motivates the need for designing schemes that provide near optimal performance even when hidden nodes are present. The primary reason for the failure of existing protocols in the presence of hidden nodes is that these protocols are based on the model developed by Bianchi. However this model does not hold when hidden nodes exist. Moreover, analyzing networks with hidden nodes is still an open problem. Thus, designing throughput optimal schemes in networks with hidden nodes is particularly challenging. The novelty of our approach is that it is not based on any underlying mathematical model, rather it directly tunes the control variables so as to maximize the throughput. We demonstrate that this model independent approach achieves maximum throughput in networks with hidden terminals as well. Apart from this major contribution, we present stochastic approximation based algorithms for achieving weighted fairness in a connected networks. We also present a throughput optimal exponential backoff based random access algorithm. We demonstrate that the exponential backoff based scheme may outperform an optimal p-persistent scheme in networks with hidden terminals. This demonstrates the merit of exponential backoff based random access schemes which was deemed unnecessary by results shown by Bianchi.