- [Show abstract] [Hide abstract] ABSTRACT: Multihop wireless mesh networks are becoming a new at- tractive communication paradigm owing to their low cost and ease of deployment. Routing protocols are critical to the performance and reliability of wireless mesh networks. Traditional routing protocols send traffic along predetermined paths and face difficulties in co ping with unreliable and unpredictable wireless medium. In this paper, we propose a Simple Opportunistic Adaptive Routing protocol (SOAR) to explic- itly support multiple simultaneous flows in wireless mesh networks. SOAR incorporates the following four major components to achieve high throughput and fairness: (i) adaptive forwarding path selection to lever- age path diversity while minimizing duplicate transmissions, (ii) priority timer-based forwarding to let only the best forwarding node forward the packet, (iii) local loss recovery to efficiently detect a nd retransmit lost packets, and (iv) adaptive rate control to determine an appropriate sending rate according to the current network conditions. We implement SOAR in both NS-2 simulation and an 18-node wireless mesh testbed. Our extensive evaluation shows that SOAR significantly outperforms traditional routing and a seminal opportunistic routing protocol, ExOR, under a wide range of scenarios.
Conference Paper: Traffic-Aware Channel Assignment in Enterprise Wireless LANs[Show abstract] [Hide abstract] ABSTRACT: Campus and enterprise wireless networks are increasingly characterized by ubiquitous coverage and rising traffic demands. Efficiently assigning channels to access points (APs) in these networks can significantly affect the performance and capacity of the WLANs. The state-of-the-art approaches assign channels statically, without considering prevailing traffic demands. In this paper, we show that the quality of a channel assignment can be improved significantly by incorporating observed traffic demands at APs and clients into the assignment process. We refer to this as traffic-aware channel assignment . We conduct extensive trace-driven and synthetic simulations and identify deployment scenarios where traffic-awareness is likely to be of great help, and scenarios where the benefit is minimal. We address key practical issues in using traffic-awareness, including measuring an interference graph, handling non-binary interference, collecting traffic demands, and predicting future demands based on historical information. We present an implementation of our assignment scheme for a 25-node WLAN testbed. Our testbed experiments show that traffic-aware assignment offers superior network performance under a wide range of real network configurations. On the whole, our approach is simple yet effective. It can be incorporated into existing WLANs with little modification to existing wireless nodes and infrastructure.
- [Show abstract] [Hide abstract] ABSTRACT: Campus and enterprise wireless networks are in- creasingly characterized by ubiquitous coverage and rising traffic demands. Efficiently assigning channels to access points (A Ps) in these networks can significantly affect the performance a nd capacity of the wireless LAN. Several research studies have tackled this issue. However, even the state-of-the-art approaches assign channels without considering prevailing traffic dem ands. The channel assignment problem has a parallel in the wire- line world, where recent work has established the tremendous effectiveness of using traffic demands in network engineeri ng decisions. Motivated by this, our paper explores whether the qual- ity of a channel assignment can be improved by incorporating observed traffic demands at APs and clients into the assignme nt process. Using extensive simulations over publicly-available wire- less traffic traces, as well as synthetic settings, we show th at being traffic-aware could substantially improve the overall quality of a channel assignment. We develop and evaluate practical traffic- aware assignment algorithms that predict future demands based on historical information and use the predicted demands for assigning channels. Finally we demonstrate the effectiveness of traffic-aware assignment using testbed experiments. I. I NTRODUCTION In the past few years, wireless networks have made signifi- cant in-roads into the common workplace. Today, most enter- prises and campuses - large or small - offer near-ubiquitous wireless coverage. There is also anecdotal evidence that the traffic volumes in workplace WLANs have grown significantly in the matter of just a few months (11). Ensuring good wireless performance in these modern set- tings is challenging. The broadcast nature of wireless commu- nication implies that WLANs are plagued by severe interfer- ence issues. Growing densities of deployment together with increasing traffic volumes only exacerbate these problems. Traditionally, careful channel assignmenthas provided some respite from this problem. In the common case, network ad- ministrators conduct detailed site surveys, and try out var ious configurations to manually determine the right channel and placement for each AP. The state-of-the-art research (15), (17) also offers similar static solutions. However, the ever-ch anging nature of the wireless frontier, with newer devices and user applications contending for the wireless medium (11), will soon render these manual, one-time approaches ineffective. Researchers in the wireline world have faced similar issues when static routing weights were deemed insufficient for managing the resources of large ISP networks. As a solution, researchers advocated adapting the routing weights to observed traffic demands. In the past few years, several operational a nd research papers have shown the tremendous effectiveness of this approach. Motivated by the success of these approaches in the IP world, our paper asks the following question: Does the quality of a channel assignment improve when dynamic traffic demands in the WLAN are taken into account? To answer this question, we develop and systematically study the notion of traffic-aware channel assignment for WLANs. We espouse traditional objectives for optimizing the channel assignment, and show how they can be modified to incorporate the observed traffic demands of wireless APs and clients, as well as their locations. We outline simple approaches for collecting current demand information in prac- tice. Obtaining optimal channel assignments that satisfy t hese objectives is NP-Hard. Therefore, we develop a simple set of techniques for efficiently obtaining channel assignments t hat can track the prevailing network conditions very closely. To evaluate the benefits of our traffic-aware approach, we use extensive simulations over both real topologies and traffic demands (available publicly at (10) and (12)), as well as over several synthetic settings. We also conduct many small- scale testbed experiments. In either case - simulations or experiments - we first evaluate a setting where perfect infor- mation about current and future demands is available. These baseline analyses help establish the potential benefits of t raffic- aware channel assignment algorithms. Our simulations and experiments show that being traffic-aware could substantia lly improve the quality of a channel assignment in terms of total network throughput. We find that approaches that incorporat e the traffic demands of both clients and APs are often superior than those that solely rely on AP traffic demands. The exact level of improvement from traffic-awareness depends on the deployment scenario - e.g. the density of wireless nodes, the traffic volumes, and the number of "hot-spot" APs. We care- fully evaluate the operating conditions where traffic-awar eness can offer the maximum benefit. We also observe that traffic- aware channel assignment offers similar fairness as existi ng traffic-agnostic approaches. In practice, the assumption of perfect demand information is unrealistic. To address this, we propose several approac hes for traffic demand prediction, and we extend our traffic-aware channel assignment algorithms to use predicted demands. We show that the performance from the resulting channel assignments is very reasonably close to the ones obtained with access to perfect information (within 5%). The rest of the paper is organized as follows. In Section II, we survey related work. We introduce traffic-agnostic and traffic-aware performance metrics used for channel assign- ment in Section III, and develop assignment algorithms in Section IV. In Section V, we describe prediction algorithms to estimate traffic demands. We introduce our evaluation metho d- ology and datasets in Section VI. We present simulation and testbed results in Section VII and Section VIII, respective ly. We discuss practical issues in Section IX and conclude in Section X.
Conference Paper: ER: Efficient retransmission scheme for wireless LANs[Show abstract] [Hide abstract] ABSTRACT: Wireless LANs (WLANs) have been deployed at a remark- able rate at university campuses, office buildings, airport s, hotels, and malls. Providing efficient and reliable wireles s communications is challenging due to inherent lossy wire- less medium and imperfect packet scheduling that results in packet collisions. In this paper, we develop an efficient retransmission scheme (ER) for wireless LANs. Instead of retransmitting the lost packets in their original forms, ER codes packets lost at different destinations and uses a sing le retransmission to potentially recover multiple packet los ses. We develop a simple and practical protocol to realize the idea and implement it in both simulation and testbed, and our results demonstrate the effectiveness of this approach .
- [Show abstract] [Hide abstract] ABSTRACT: Multihop wireless mesh networks are becoming a new attractive communication paradigm. Many cities and public places have deployed or are planning to deploy mesh networks to provide Internet access to residents and local businesses. Routing protocol design is critical to the performance and reliability of wireless mesh networks. Traditional routing protocols send traffic along pre-determined paths and have been shown ineffective in coping with unreliable and unpredictable wireless medium. In this paper, we develop a simple opportunistic adaptive routing protocol (SOAR) for wireless mesh networks. SOAR maximizes the progress each packet makes by using priority-based timers to ensure that the most preferred node forwards the packet with little coordination overhead. Moreover, SOAR minimizes resource consumption and duplicate transmissions by judiciously selecting forwarding nodes to prevent routes from diverging. To further protect against packet losses, SOAR uses local recovery to retransmit a packet when an ACK is not received within a specified time. SOAR uses a combination of selective ACKs, piggyback ACKs, and ACK compression to protect against ACK loss while minimizing ACK overhead. We evaluate SOAR using NS-2 simulations. Our preliminary results show that SOAR is promising to achieve high efficiency and effectively support multiple simultaneous flows.