wireless mesh network

One of the triumphs of wireline network research of the last decade has been the casting of the Internet congestion control problem within an optimization framework based on utility functions. Such an approach provides a sound understanding of the underlying stability and fairness issues, as well as a post-facto justification of TCP-like additiveincrease multiplicative-decrease (AIMD) algorithms. This paper provides a counter-example showing that the same result cannot be extended to wireless networks, at least not in a straightforward manner. The fundamental difference is that wireless networks are of a broadcast nature. There is no strict notion of a “link,” since transmissions from nearby nodes interfere with each other. Using a simple model of interference in wireless networks, a counter-example of a wireless network is presented in which the congestion control mechanism has an unstable equilibrium point at the desired fair solution. Further, ns-2 simulations of this counter-example manifest an oscillatory behavior. Surprisingly, this oscillatory behavior appears to be fairly typical in wireless networks, with most randomly chosen network examples manifesting it. This loss of stability suggests a possible need for the re-design of wireless TCP and wireless queue management to explicitly account for the wireless nature of the effects of interference.

Many previous papers have pointed out that TCP performance in multi-hop ad hoc networks is not optimal. This is due to several TCP design principles that reflect the characteristics of wired networks dominant at the time when TCP was designed, that are not adequate for multi-hop ad hoc networks. For example, congestion phenomena in multi-hop networks are very different than in traditional wired networks, and route failures and route changes may be frequent events. To overcome these problems, in a previous work we presented a novel transport protocol – named TPA – specifically tailored to multi-hop ad hoc networks. In this chapter we present the implementation in a GNU/Linux environment of a TPA prototype. This prototype has been implemented at the user level to make the debugging process easier and to do not affect the system stability. This chapter also reports some results of the experimental evaluation of TPA in a static multi-hop scenario. Specifically, we used our prototype to compare TCP and TPA performance in a chain topology network. Our experimental results show that, in the analyzed scenarios, TPA always outperforms TCP in a significant way both in terms of throughput and energy consumption. 1.

Many previous papers have pointed out that TCP performance in multi-hop ad hoc networks (MANETs) is sub-optimal. This is due to several TCP design principles that reflect the characteristics of wired networks dominant at the time when TCP was designed that do not hold in MANETs. Based on this evidence, several TCP variants have been proposed in the literature. However, little effort has been devoted to investigate the performance of TCP in a real environment, even in a static scenario. Most of the work relies on simulation. In this chapter we provide an experimental analysis of TCP in static multi-hop ad hoc networks. We investigate the TCP performance in a simple, but interesting, scenario, i.e., a chain topology with different number of hops. We highlight some results contrasting with simulations and show that these discrepancies are due to the different protocols-- or different protocol implementations-- used in practice with respect to simulation tools. 1.

Although it is well-known that TCP throughput is suboptimal in multihop wireless networks, little performance data is available for TCP in realistic wireless environments. In this paper, we present the results of an extensive experimental study of TCP performance on a 32-node wireless mesh network testbed deployed on the Purdue University campus. Contrary to prior work which considered a single topology with equal-length links and only 1-hop neighbors within transmission range of each other, our study considers more realistic heterogeneous topologies. We vary the maximum TCP window size, in correlation with two important MAC layer parameters: the use of RTS/CTS and the MAC data rate. Based on our TCP throughput results, we give recommendations on configuring TCP and MAC parameters, which in many cases contradict previous proposals (which had themselves contradicted each other).

Abstract—Wireless multi-hop access networks are an increasingly popular option to provide cost-efficient last-mile Internet access. However, despite extensive research, performance of even basic communication services, such as TCP, is still problematic. Measurements collected on a wireless testbed indicate that the poor performance of multi-hop access networks is caused by poor interactions between TCP congestion control and linklayer bit-rate adaptation resulting in severely reduced network efficiency even over short wireless paths (< 6 hops). However, bit-rate adaptation improves fairness across TCP flows. The same principal observations hold for hybrid wireless/wireline paths. To investigate approaches to improve TCP performance, we present a simple model that captures the cause for the inefficiency of TCP over autorate links. We then examine several techniques at both the TCP level and the link layer (TCP Vegas, clamping, limiting the buffer size at the wireless routers) to alleviate contention. None of these techniques works for all scenarios, but the simple approach to limit the buffer size is attractive in many settings that include four or more wireless hops. I.

This paper investigates the interaction between end-to-end flow control and MAC-layer scheduling on wireless links. We consider a wireless network with multiple users receiving information from a common access point; each user suffers fading, and a scheduler allocates the channel based on channel quality, but subject to fairness and latency considerations. We show that the fairness property of the scheduler is compromised by the transport layer flow control of TCP NewReno. We provide a receiver-side control algorithm, CLAMP, that remedies this situation. CLAMP works at a receiver to control a TCP sender by setting the TCP receiver’s advertised window limit. CLAMP provides control over wireless link utilisation and queueing delay at the access point buffer, and it provides the scheduler control over the allocation of bandwidth between the users. The paper shows that physical and link layer effects have a strong impact on transport layer performance, and vice versa, both for TCP and for CLAMP-controlled TCP.

wireless networks

The advent of static wireless mesh networks (WMNs) has largely shifted the design goals of wireless routing protocols from maintaining connectivity among routers to providing high throughput. The change of design goals has led to the creation of many “exotic ” optimization techniques such as opportunistic routing and network coding, that promise a dramatic increase in overall network throughput. These “exotic” techniques have also moved many mechanisms such as reliability and rate control, that used to be below or above the routing layer in traditional protocols, to the routing layer. In this paper, we first review the above evolution of routing protocol design and show that the consolidation of mechanisms from multiple layers into the routing layer poses new challenges to the methodology for evaluating and comparing this new generation of routing protocols. We then discuss the diverse set of current practices in evaluating recently proposed protocols and their strengths and weaknesses. Our discussion suggests that there is an urgent need to carefully rethink the implications of the new merged-layer routing protocol design and develop effective methodologies for meaningful and fair comparison of these protocols. Finally, we make several concrete suggestions on the desired evaluation methodology. In particular, we show that the traffic sending rate plays a fundamental role and should be carefully controlled.

Abstract not found