Glenn Judd

Self-management in chaotic wireless deployments

Over the past few years, wireless networking technologies have made vast forays into our daily lives. Today, one can find 802.11 hardware and other personal wireless technology employed at homes, shopping malls, coffee shops and airports. Present-day wireless network deployments bear two important properties: they are unplanned, with most access points (APs) deployed by users in a spontaneous manner, resulting in highly variable AP densities; and they are unmanaged, since manually configuring and managing a wireless network is very complicated. We refer to such wireless deployments as being chaotic.In this paper, we present a study of the impact of interference in chaotic 802.11 deployments on end-client performance. First, using large-scale measurement data from several cities, we show that it is not uncommon to have tens of APs deployed in close proximity of each other. Moreover, most APs are not configured to minimize interference with their neighbors. We then perform trace-driven simulations to show that the performance of end-clients could suffer significantly in chaotic deployments. We argue that end-client experience could be significantly improved by making chaotic wireless networks self-managing. We design and evaluate automated power control and rate adaptation algorithms to minimize interference among neighboring APs, while ensuring robust end-client performance.

Efficient channel-aware rate adaptation in dynamic environments

Increasingly, 802.11 devices are being used by mobile users. This results in very dynamic wireless channels that are difficult to use efficiently. Current rate selection algorithms are dominated by probe-based approaches that search for the best transmission rate using trial-and-error. In mobile environments, probe-based techniques often perform poorly because they inefficiently search for the moving target presented by the constantly changing channel. We have developed a channel-aware rate adaptation algorithm CHARM - that uses signal strength measurements collected by the wireless cards to help select the transmission rate. Moreover, unlike previous approaches CHARM leverages channel reciprocity to obtain channel information, so the information is available to the transmitter without incurring RTS/CTS overhead. This combination of techniques allows CHARM to respond quickly to dynamic channel changes. We implemented CHARM in the Madwifi driver for wireless cards using the Atheros chipset. Our evaluation both in the real world and on a controlled testbed shows that channel-aware rate selection can significantly outperform probe-based rate adaptation, especially over dynamic channels.

Providing contextual information to pervasive computing applications

Pervasive computing applications are increasingly leveraging contextual information from several sources to provide users with behavior appropriate to the environment in which they reside. If these sources of contextual information are used and deployed in an ad hoc manner however they may provide overlapping functionality, fail to provide needed functionality, and require the use of inconsistent interfaces by applications. To overcome these problems, we introduce a contextual information service that provides applications with contextual information via a virtual database. Unlike previous efforts, our service provides applications a consistent, lightweight, and powerful mechanism for obtaining contextual information, and includes explicit support for the on demand computation of contextual information. We show, via example applications and a contextual information service prototype that we have implemented, how this approach can be used to allow proactive applications to adapt their behavior to match a users current environment.

Repeatable and realistic wireless experimentation through physical emulation

In wireless networking research, there has long existed a fundamental tension between experimental realism on one hand, and control and repeatability on the other hand. Hardware-based experimentation provides realism, but is tightly coupled to the physical environment and circumstances under which experiments are carried out. To overcome this, researchers have understandably embraced simulation as a means of evaluation. Unfortunately, wireless simulation is plagued with inherent inaccuracies. To overcome the stark tradeoff between the realism of hardware-based experimentation and the repeatability of simulation-based experimentation, we are developing a wireless emulator that enables both realistic and repeatable experimentation. Unlike previous emulators, our approach simultaneously achieves both a high degree of realism and fine-grained repeatability by leveraging physical layer emulation.

Fixing 802.11 access point selection

As 802.11 deployment and use become ubiquitous, it becomes increasingly important to make efficient use of the bandwidth provided. Unfortunately, 802.11’s current access point selection model fails to enable efficient use of access point bandwidth. We discuss why the current 802.11 access point selection model fails, and evaluate alternative models via simulation. Our results indicate that with a small amount of enhanced functionality, 802.11 access point selection can be vastly improved.

FPGA-Based Channel Simulator for a Wireless Network Emulator

Wireless channel emulators are important tools for testing radio devices, especially in mobile environments. Wireless network emulators give the same accuracy and control for testing radio network systems that traditional channel emulators give to point to point radio links. Network emulators require many more independent channels than traditional channel emulators. This problem is particularly challenging for the real time channel simulator in the emulator. The challenges of designing a wireless network channel simulator are discussed and a design is presented on a Xilinx Virtex-II Pro FPGA. This channel simulator can model 210 independent channels between 15 nodes with a bandwidth of 90MHz. The performance of the design was verified by measuring transport-layer throughput between 802.11b radios transmitting through the channel simulator.

Characterizing 802.11 wireless link behavior

Since wireless signals propagate through the ether, they are significantly affected by attenuation, fading, multipath, and interference. As a result, it is difficult to measure and understand fundamental wireless network behavior. This creates a challenge for both network researchers, who often rely on simulators to evaluate their work, and network managers, who need to deploy and optimize operational networks. Given the complexity of wireless networks, both communities often rely on simplifying rules, which frequently have not been validated using today’s wireless radios. In this paper, we undertake a detailed characterization of 802.11 link-level behavior using commercial 802.11 cards. Our study uses a wireless testbed that provides signal propagation emulation, giving us complete control over the signal environment. In addition, we use our measurements to analyze the performance of an operational wireless network. Our work contributes to a more accurate understanding of link-level behavior and enables the development of more accurate wireless network simulators.

A simple mechanism for capturing and replaying wireless channels

Physical layer wireless network emulation has the potential to be a powerful experimental tool. An important challenge in physical emulation, and traditional simulation, is to accurately model the wireless channel. In this paper we examine the possibility of using on-card signal strength measurements to capture wireless channel traces. A key advantage of this approach is the simplicity and ubiquity with which these measurements can be obtained since virtually all wireless devices provide the required metrics. We show that for low delay spread environments wireless traces gathered using this method can be replayed in a physical wireless emulator to produce higher layer network behavior that is similar to the behavior that would have occurred in the real world. Thus, wireless channel traces gathered using on-card metrics are an effective means of enabling existing low delay spread wireless testbeds to be emulated.

Low-overhead channel-aware rate adaptation

Current rate selection algorithms are dominated by probe-based approaches that search for the best transmission rate using trial-and-error. When operating over a dynamic channel, probe-based techniques can perform poorly since they inefficiently search for the moving target presented by the constantly changing channel. We have developed a channel-aware rate adaptation algorithm - CHARM - that responds quickly to dynamic channel changes, and significantly outperforms probe-based algorithms in many instances. Unlike previous approaches, CHARM leverages channel reciprocity to obtain channel information without incurring RTS/CTS overhead. Our work shows that channel-aware rate selection is viable, and can significantly outperform probe-based rate adaptation over both static and dynamic channels.

A software architecture for physical layer wireless network emulation

Despite their widespread deployment, many aspects of wireless network performance are poorly understood, and there is great room from improvement in wireless network reliability and performance. A key obstacle to understanding and improving wireless networks has been the lack of a realistic yet flexible experimental methodology. Physical layer wireless network emulation promises to achieve much of the flexibility of wireless simulators while maintaining much of the realism of real wireless networks. We have developed a software architecture that tames the complexity of physical layer wireless network emulation, and presents users with a powerful yet ease-to-use interface. We present several case studies showing how this software architecture allows complex wireless experiments to be conducted in an efficient manner while still enabling novice users to quickly run simple experiments.