Wireless mesh networks (WMNs) are an alternative technology for last-mile broadband Internet access. In WMNs, similar to ad hoc networks, each user node operates not only as a host but also as a router; user packets are forwarded to and from an Internet-connected gateway in multihop fashion. The meshed topology provides good reliability, market coverage and scalability, as well as low upfront investments. Despite the recent start-up surge in WMNs, much research remains to be done before WMNs realize their full potential. This paper tackles the problem of determining the exact capacity of a WMN. The key concept that we introduce to enable this calculation is the bottleneck collision domain that is defined as the geographical area of the network that bounds from above the amount of data that can be transmitted in the network. We show that for WMNs the throughput of each node decreases as O(1/n),wheren is the total number of nodes in the network. In contrast with most existing work on ad hoc network capacity, we do not limit our study to the asymptotic case. In particular, for a given topology and the set of active nodes, we provide exact upper-bounds on the throughput of any node. The calculation can be used to provision the network, to ensure quality of service and fairness, etc. The theoretical results are validated by detailed simulations.

Modern enterprise networks are of sufficient complexity that even simple faults can be difficult to diagnose — let alone transient outages or service degradations. Nowhere is this problem more apparent than in the 802.11-based wireless access networks now ubiquitous in the enterprise. In addition to the myriad complexities of the wired network, wireless networks face the additional challenges of shared spectrum, user mobility and authentication management. Not surprisingly, few organizations have the expertise, data or tools to decompose the underlying problems and interactions responsible for transient outages or performance degradations. In this paper, we present a set of modeling techniques for automatically characterizing the source of such problems. In particular, we focus on data transfer delays unique to 802.11 networks — media access dynamics and mobility management latency. Through a combination of measurement, inference and modeling we reconstruct sources of delay — from the physical layer to the transport layer — as well as the interactions among them. We demonstrate our approach using comprehensive traces of wireless activity in the UCSD Computer Science building.

The growing popularity of wireless networks has led to cases of heavy utilization and congestion. In heavily utilized wireless networks, the wireless portion of the network is a major performance bottleneck. Understanding the behavior of the wireless portion of such networks is critical to ensure their robust operation. This understanding can also help optimize network performance. In this paper, we use link layer information collected from an operational, large-scale, and heavily utilized IEEE 802.11b wireless network deployed at the 62 nd Internet Engineering Task Force (IETF) meeting to study congestion in wireless networks. We motivate the use of channel busy-time as a direct measure of channel utilization and show how channel utilization along with network throughput and goodput can be used to define highly congested, moderately congested, and uncongested network states. Our study correlates network congestion and its effect on link-layer performance. Based on these correlations we find that (1) current rate adaptation implementations make scarce use of the 2 Mbps and 5.5 Mbps data rates, (2) the use of Request-to-Send/Clear-to-Send (RTS–CTS) prevents nodes from gaining fair access to a heavily congested channel, and (3) the use of rate adaptation, as a response to congestion, is detrimental to network performance. 1

The growing deployment and concomitant rise in wireless network usage necessitates the comprehensive understanding of its behavior. More importantly, as networks grow in size and number of users, congestion in the wireless portion of the network is likely to increase. We believe there is a strong need to understand the intricacies of the wireless portion of a congested network by interpreting information collected from the network. Congestion in a wireless network can be best analyzed by studying the transmission of frames at the link layer. To this end, we use vicinity sniffing techniques to analyze the link layer in an operational IEEE 802.11b wireless network. In this paper, we discuss how congestion in a network can be estimated using point-to-point link reliability. We then show how link reliability is correlated with the behavior of link-layer properties such as frame retransmissions, frame sizes, and data rates. Based on the results from these correlations, our hypothesis is that the performance of the link layer in congested networks can be improved by (1) sending smaller frames, and/or (2) using higher data rates with a fewer number of frames sent.

WLANs are indispensable for providing Internet access to users at locations such as universities, corporate offices, conferences, airports, and coffee shops. Many of these environments often experience flash crowds, which we define to be a sudden surge in the number of users simultaneously attempting to access the WLAN. When flash crowds occur, WLANs are likely to suffer from destructive interference, excessive channel load, and unsustainable packet processing at access points (APs). These conditions lead to a plethora of problems, such as a deterioration in network throughput, heavy packet loss, intermittent connectivity, overwhelmed APs, and sometimes, a network collapse. To verify these claims, we present two case studies of operational WLANs that experienced the aforementioned problems. The two WLANs each consisted of over 100 APs and more than 1000 simultaneous users, deployed at recently held 62 nd and 64 th Internet

Testing and deploying mobile wireless networks and applications are very challenging tasks, due to the network size and administration as well as node mobility management. Well known simulation tools provide a more flexible environment but they do not run in real time and they rely on models of the developed system rather than on the system itself. Emulation is a hybrid approach allowing real application and tra#c to be run over a simulated network, at the expense of accuracy when the number of nodes is too important. In this paper, emulation is split in two stages: first, the simulation of network conditions is precomputed so that it does not undergo real-time constraints that decrease its accuracy; second, real applications and tra#c are run on an emulation platform where the precomputed events are scheduled in soft real-time. This allows the use of accurate models for node mobility, radio signal propagation and communication stacks. An example shows that a simple situation can be simply tested with real applications and tra#c while relying on accurate models.

When a roadside 802.11-based wireless access point is shared by more than one vehicle, the vehicle with the lowest transmission rate reduces the effective transmission rate of all other vehicles. This performance anomaly [9] degrades both individual and overall throughput in such multi-vehicular environments. Observing that every vehicle eventually receives good performance when it is near the access point, we propose MV-MAX (Multi-Vehicular Maximum), a medium access protocol that opportunistically grants wireless access to vehicles with the maximum transmission rate. Mathematical analysis and trace-driven simulations based on real data show that MV-MAX not only improves overall system throughput, compared to 802.11, by a factor of almost 4, but also improves on the previously proposed time-fairness scheme [20, 22, 15] by a factor of more than 2. Moreover, despite being less fair than 802.11, almost every vehicle benefits by using MV-MAX over the more equitable 802.11 access mechanism. Finally, we show that our results are consistent across different data sets.

In multihop wireless networks, fair allocation of bandwidth among different nodes is one of the critical problems that affects the serviceability of the entire system. Although there is significant research on the fairness issues in single-hop wireless networks, research on multihop fairness is rarely found in the literature. We study various queuing schemes for multihop wireless networks and examine the fairness and throughput performance of each scheme. Each scheme offers a different degree of fairness. While relatively simple queuing schemes require less hardware and processing budget, they inevitably lack good fairness and performance. In contrast, the scheme that provides fairness requires per-flow (i.e., network-layer flow) queuing. Furthermore, we show that in order to achieve the optimal bandwidth utilization, the medium access control (MAC) layer should be able to support different priorities. Without such a MAC-layer QoS scheme, in the worst case, the bandwidth utilization can be degraded by O(N),whereN is the number of end users. We theoretically investigate the pros and cons of different queuing schemes and verify the analytical results with detailed simulations.

We present a new metric, Expected Data Rate (EDR), for accurately finding high-throughput paths in multihop ad hoc wireless networks. Our metric is based upon a new model for transmission interference which is a critical factor in determining path throughput. We construct a realistic and practical transmission interference model by (1) determining transmission contention degree of each link as a function of the wireless link loss, (2) quantifying the impact of the wireless link loss on medium access backoff, and (3) considering possible concurrent transmissions when two links do not interfere with each other. Our transmission interference model also takes the non-optimality of IEEE 802.11 medium access scheduling into account. Using extensive ns-2 simulations of IEEE 802.11 ad hoc networks, we find that EDR can accurately determine the achievable data rates of ad hoc paths, thereby significantly outperforming the other existing metrics.

We argue that carrier sense in 802.11 and other wireless protocols leads to scheduling decisions that are overly pessimistic and hence waste capacity. As an alternative, we propose interference cancellation, in which simultaneous signals are modeled and decoded together rather than treating all but one as random noise. This method greatly expands the conditions under which overlapping transmissions can be successfully received, even by a single receiver. We demonstrate the practicality of these better receivers via a proof-of-concept experiment with USRP software radios. We argue that supporting concurrent transmissions enables new and more effective wireless MACs in which carrier sense is disabled.