Edward W. Knightly

Opportunistic media access for multirate ad hoc networks

The IEEE 802.11 wireless media access standard supports multiple data rates at the physical layer. Moreover, various auto rate adaptation mechanisms at the medium access layer have been proposed to utilize this multi-rate capability by automatically adapting the transmission rate to best match the channel conditions. In this paper, we introduce the Opportunistic Auto Rate (OAR) protocol to better exploit durations of high-quality channels conditions. The key mechanism of the OAR protocol is to opportunistically send multiple back-to-back data packets whenever the channel quality is good. As channel coherence times typically exceed multiple packet transmission times for both mobile and non-mobile users, OAR achieves significant throughput gains as compared to state-of-the-art auto-rate adaptation mechanisms. Moreover, over longer time scales, OAR ensures that all nodes are granted channel access for the same time-shares as achieved by single-rate IEEE 802.11. We describe mechanisms to implement OAR on top of any existing auto-rate adaptation scheme in a nearly IEEE 802.11 compliant manner. We also analytically study OAR and characterize the gains in throughput as a function of the channel conditions. Finally, we perform an extensive set of ns-2 simulations to study the impact of such factors as node velocity, channel conditions, and topology on the throughput of OAR.

Endpoint admission control: architectural issues and performance

The traditional approach to implementing admission control, as exemplified by the Integrated Services proposal in the IETF, uses a signalling protocol to establish reservations at all routers along the path. While providing excellent quality-of-service, this approach has limited scalability because it requires routers to keep per-flow state and to process per-flow reservation messages. In an attempt to implement admission control without these scalability problems, several recent papers have proposed various forms of endpoint admission control. In these designs, the hosts (the endpoints) probe the network to detect the level of congestion; the host admits the flow only if the detected level of congestion is sufficiently low. This paper is devoted to the study of endpoint admission control. We first consider several architectural issues that guide (and constrain) the design of such systems. We then use simulations to evaluate the performance of endpoint admission control in various settings. The modest performance degradation between traditional router-based admission control and endpoint admission control suggests that a real-time service based on endpoint probing may be viable.

Modeling per-flow throughput and capturing starvation in CSMA multi-hop wireless networks

Multi-hop wireless networks employing random access protocols have been shown to incur large discrepancies in the throughputs achieved by the flows sharing the network. Indeed, flow throughputs can span orders of magnitude from near starvation to many times greater than the mean. In this paper, we address the foundations of this disparity. We show that the fundamental cause is not merely differences in the number of contending neighbors, but a generic coordination problem of CSMA-based random access in a multi-hop environment. We develop a new analytical model that incorporates this lack of coordination, identifies dominating and starving flows and accurately predicts per-flow throughput in a large-scale network.We then propose metrics that quantify throughput imbalances due to the MAC protocol operation. Our model and metrics provide a deeper understanding of the behavior of CSMA protocols in arbitrary topologies and can aid the design of effective protocol solutions to the starvation problem.

Low-rate TCP-targeted denial of service attacks: the shrew vs. the mice and elephants

Denial of Service attacks are presenting an increasing threat to the global inter-networking infrastructure. While TCPs congestion control algorithm is highly robust to diverse network conditions, its implicit assumption of end-system cooperation results in a well-known vulnerability to attack by high-rate non-responsive flows. In this paper, we investigate a class of low-rate denial of service attacks which, unlike high-rate attacks, are difficult for routers and counter-DoS mechanisms to detect. Using a combination of analytical modeling, simulations, and Internet experiments, we show that maliciously chosen low-rate DoS traffic patterns that exploit TCPs retransmission timeout mechanism can throttle TCP flows to a small fraction of their ideal rate while eluding detection. Moreover, as such attacks exploit protocol homogeneity, we study fundamental limits of the ability of a class of randomized timeout mechanisms to thwart such low-rate DoS attacks

Admission control for statistical QoS: theory and practice

In networks that support quality of service, an admission control algorithm determines whether or not a new traffic flow can be admitted to the network such that all users will receive their required performance. Such an algorithm is a key component of future multiservice networks because it determines the extent to which network resources are utilized and whether the promised QoS parameters are actually delivered. The goals in this article are threefold. First, we describe and classify a broad set of proposed admission control algorithms. Second, we evaluate the accuracy of these algorithms via experiments using both on-off sources and long traces of compressed video; we compare the admissible regions and QoS parameters predicted by our implementations of the algorithms with those obtained from trace-driven simulations. Finally, we identify the key aspects of an admission control algorithm necessary for achieving a high degree of accuracy and hence a high statistical multiplexing gain.

Denial of service resilience in ad hoc networks

Significant progress has been made towards making ad hoc networks secure and DoS resilient. However, little attention has been focused on quantifying DoS resilience: Do ad hoc networks have sufficiently redundant paths and counter-DoS mechanisms to make DoS attacks largely ineffective? Or are there attack and system factors that can lead to devastating effects? In this paper, we design and study DoS attacks in order to assess the damage that difficult-to-detect attackers can cause. The first attack we study, called the JellyFish attack, is targeted against closed-loop flows such as TCP; although protocol compliant, it has devastating effects. The second is the Black Hole attack, which has effects similar to the JellyFish, but on open-loop flows. We quantify via simulations and analytical modeling the scalability of DoS attacks as a function of key performance parameters such as mobility, system size, node density, and counter-DoS strategy. One perhaps surprising result is that such DoS attacks can increase the capacity of ad hoc networks, as they starve multi-hop flows and only allow one-hop communication, a capacity-maximizing, yet clearly undesirable situation.

Opportunistic fair scheduling over multiple wireless channels

Emerging spread spectrum high-speed data networks utilize multiple channels via orthogonal codes or frequency-hopping patterns such that multiple users can transmit concurrently. In this paper, we develop a framework for opportunistic scheduling over multiple wireless channels. With a realistic channel model, any subset of users can be selected for data transmission at any time, albeit with different throughputs and system resource requirements. We first transform selection of the best users and rates from a complex general optimization problem into a decoupled and tractable formulation: a multiuser scheduling problem that maximizes total system throughput and a control-update problem that ensures long-term deterministic or probabilistic fairness constraints. We then design and evaluate practical schedulers that approximate these objectives.

Measurement driven deployment of a two-tier urban mesh access network

Multihop wireless mesh networks can provide Internet access over a wide area with minimal infrastructure expenditure. In this work, we present a measurement driven deployment strategy and a data-driven model to study the impact of design and topology decisions on network-wide performance and cost. We perform extensive measurements in a two-tier urban scenario to characterize the propagation environment and correlate received signal strength with application layer throughput. We find that well-known estimates for pathloss produce either heavily overprovisioned networks resulting in an order of magnitude increase in cost for high pathloss estimates or completely disconnected networks for low pathloss estimates. Modeling throughput with wireless interface manufacturer specifications similarly results in severely underprovisioned networks. Further, we measure competing, multihop flow traffic matrices to empirically define achievable throughputs of fully backlogged, rate limited, and web-emulated traffic. We find that while fully backlogged flows produce starving nodes, rate-controlling flows to a fixed value yields fairness and high aggregate throughput. Likewise, transmission gaps occurring in statistically multiplexed web traffic, even under high offered load, remove starvation and yield high performance. In comparison, we find that well-known noncompeting flow models for mesh networks over-estimate network-wide throughput by a factor of 2. Finally, our placement study shows that a regular grid topology achieves up to 50 percent greater throughput than random node placement.

QoS-driven server migration for Internet data centers

Many organizations have chosen to host Internet applications at Internet data centers (IDCs) located near network access points of the Internet to take advantage of their high availability, large network bandwidths and low network latencies. Current IDCs provide for a dedicated and static allocation of resources to each hosted application. Unfortunately, workloads for these sites are highly variable, leading to poor resource utilization, poor application performance, or both. In this paper, we develop a framework for QoS-driven dynamic resource allocation in IDCs. Termed QuID (quality of service infrastructure on demand), the frameworks contributions are threefold. First, we develop a simple adaptive algorithm to reduce the average number of servers used by an application while satisfying its QoS objectives. Second, we develop an optimal off-line algorithm that bounds the advantage of any dynamic policy and provides a benchmark for performance evaluation. Finally, we perform an extensive simulation study using traces from large-scale E-commerce and search-engine sites. We explore the gains of the QuID algorithms as a function of the system parameters (such as server migration time), algorithm parameters (such as control time scale), and workload characteristics (such as peak-to-mean ratio and autocorrelation function of the request rate).

Distributed multi-hop scheduling and medium access with delay and throughput constraints

Providing quality of service in random access multi-hop wireless networks requires support from both medium access and packet scheduling algorithms. However, due to the distributed nature of ad hoc networks, nodes may not be able to determine the next packet that would be transmitted in a (hypothetical) centralized and ideal dynamic priority scheduler. In this paper, we develop two mechanisms for QoS communication in multi-hop wireless networks. First, we devise distributed priority scheduling a technique that piggybacks the priority tag of a nodes head-of-line packet onto handshake and data packets; e.g., RTS/DATA packets in IEEE 802.11. By monitoring transmitted packets, each node maintains a scheduling table which is used to assess the nodes priority level relative to other nodes. We then incorporate this scheduling table into existing IEEE 802.11 priority back-off schemes to approximate the idealized schedule. Second, we observe that congestion, link errors, and the random nature of medium access prohibit an exact realization of the ideal schedule. Consequently, we devise a scheduling scheme termedmulti-hop coordinationso that downstream nodes can increase a packets relative priority to make up for excessive delays incurred upstream. We next develop a simple analytical model to quantitatively explore these two mechanisms. In the former case, we study the impact of the probability of overhearing another packets priority index on the schemes ability to achieve the ideal schedule. In the latter case, we explore the role of multi-hop coordination in increasing the probability that a packet satisfies its end-to-end QoS target. Finally, we perform a set of ns-2 simulations to study the schemes performance under more realistic conditions.