Channel Allocation and post-CCA based Bandwidth Adaptation in Wireless Local Area Networks with Heterogeneous Bandwidths

Abstract

The new prominent 802.11ac standard aims at achieving Gbps data throughput for individual users while at the same time guaranteeing backward compatibility. The approaches to achieving this goal use enhanced physical-layer features, such as higher modulation levels, MIMO (Multiple Input Multiple Output), and wider bandwidth. As for the bandwidth, the channel bonding technique that makes use of multiple 20MHz channels in 5GHz band is adopted. However, the heterogeneity of bandwidth in a network can cause asymmetric interferences in which some transmissions are not sensed by some nodes. As a result, the conventional CSMA/CA (Carrier Sense Multiple Access with Collision Avoidance) may not work well in 802.11ac networks and the Gbps throughput, which is attainable for a single link, is not achievable network-widely, which we call the Hidden Channel (HC) problem.

In this dissertation, we illustrate the HC problem with a 802.11ac network as a reference system. Then we analyze the problem using Markov chain technique and show how the contention parameters and transmission time affect collision probability and fairness in some deployment scenarios. The validity of the analysis is verified through simulation in the same chapter.

As a solution to the HC problem, a centralized and heuristic channel allocation algorithm, PCA (Primary Channel Allocation), in an enterprise local area network is proposed in the next part of this dissertation. The PCA algorithm, an extended version of well known ``University Timetabling'' algorithm for incorporating multi-channel purpose, is designed to avoid HC problem effectively. Through simulations, we demonstrate that our proposed channel allocation algorithm lowers the packet error rate (PER) compared to uncoordinated and RSSI (Received Signal Strength Indicator) based allocation schemes and increases the network-wide throughput as well as the throughput of a station that experiences poor performance. This implies improved fairness performance among transmission pairs with various channel bandwidths.

Then, simple experiments are conducted with USRP and WARP boards to show that the problem is real and to prove that the validity of our next solution. Based on that, we argue for the need of bandwidth adaptation based on post-CCA, which is another clear channel assessment (CCA) procedure after finishing a transmission. The post-CCA helps mimic the CSMA/CD (CSMA with Collision Detection) mechanism in the wired Ethernet, thus enhancing channel assessment capability. Then, we propose PoBA (Post-CCA based Bandwidth Adaptation) that alters bandwidth and channel configuration dynamically. Using simulation, we confirm that the PoBA increases network-wide throughput, channel utilization and fairness, and decreases packet error probability.