Thesis (Ph. D.)--Hong Kong University of Science and Technology, 2003. / Includes bibliographical references (leaves 171-179). Also available in electronic version. Access restricted to campus users.

Not available / text

Not available

Conventional data networks are based on a layered architecture, in which a layer implements some aspect of the network while hiding the detailed implementation from the other layers. The introduction of wireless networks has created a need to violate this layered discipline to create cross-layer designs or adaptations. Such cross-layer adaptations optimize the performance of wireless networks by using information from any layer in the network. The key problem is that ad-hoc implementations of cross-layer adaptations introduce complex interactions between layers and thus reduce the level of modularity and abstraction in the network's implementation. This gives rise to a significant increase in complexity. We demonstrate that a new software architecture is able to provide a systematic framework that helps us to implement a wide variety of cross-layer adaptations while preserving to a significant degree the modularity found in the existing network's implementation. To develop such an architecture, we first create a taxonomy of possible cross-layer adaptations. The taxonomy allows a precise description of a wide variety of cross-layer adaptations. Thus our taxonomy can serve as a framework for developing a cross-layer architecture. We develop the software architecture by creating two architectures, a conceptual one and a concrete one. We first develop a conceptual architecture, which shows the key mechanisms that are required to implement cross-layer adaptations. This architecture helps us to understand how we can implement cross-layer adaptations by using our architectural framework. We then develop a concrete architecture, which shows how we can implement such a conceptual architecture on real wireless systems. This architecture addresses more detailed implementation issues. We design the concrete architecture for Hydra, which is a flexible wireless network testbed. We then show that our architecture is generic enough to allow us to support a wide set of cross-layer architectures. We evaluate the proposed architecture by performing three case studies, each of which implements a cross-layer adaptation within Hydra based on the concrete architecture. The case studies allow us to implement and evaluate the key mechanisms provided by our architectural framework. We also implement each cross-layer adaptation by using a conventional approach, in which one layer performs the cross-layer adaptation directly communicating with other layers and other nodes. Comparing both the implementation techniques allows us to evaluate how our architectural framework supports a wide variety of cross-layer adaptations while reducing the complexity of implementation of cross-layer adaptations. / text

M. Tech. Electrical Engineering. / Widespread deployment of wireless local area networks (WLANs) and a gradual increase in streaming applications have brought about a demand for improved Quality of Service (QoS) in wireless networks. The IEEE 802.11e standard was proposed to provide QoS mechanisms for assigning high priority to delay-sensitive applications. However, Internet traffic is still dominated by TCP based applications, and the negative effects of the IEEE 802.11e service differentiation scheme on TCP performance in the presence of high priority traffic are becoming a challenging issue. TCP has been found to perform poorly in wireless networks, including IEEE 802.11e; more applications with higher QoS demands use UDP in the transport layer than TCP. Therefore, the QoS of low priority traffic in 802.11e is not guaranteed in networks highly loaded with high priority traffic. This is aggravated by the class differentiation introduced in current QoS protocols, which results in TCP applications being starved during high traffic load. The motivation of this work is to enhance the interaction between the TCP and MAC protocols in order to improve TCP performance in WLANs.

No description available.

No description available.

No description available.

This thesis focuses on the issues in transport protocol performance in wireless ad-hoc networks.From the review of existing research we identify the following three main factors contributing to the poor performance of transport protocols such as TCP. These are: the inability of transport protocol to detect/identify and respond to network events in wireless ad-hoc networks; the inability of TCP window-based flow control mechanism to avoid overloading the wireless channel; and the frequent route failures due to mobility, causing interruptions in transmission. / We focus on 2 mechanisms to improve the performance of transport protocols. Recognising the many benefits of multipath routing in wireless ad-hoc networks, we propose a multipath transport design framework, incorporating a number of transport protocol components that are suitable for wireless ad-hoc networks, with different degrees of cross-interference. We first consider the use of independent multiple paths, and show that the throughput performance of multiple paths is no more than 55-60%, due to a spatial reuse of the wireless channel at the source. We then show that the overall improvement resulting from the use of independent multiple paths is further limited by the scarcity of independent multiple paths between a given pair of source and destination nodes. We conclude that independent multipaths are worth consideration only in sufficiently dense and large networks. Then we propose metrics to characterise node-disjoint multipaths with different degrees of cross-interference, and investigate correlation of these metrics with the throughput performance of multipath. / Our simulation results show that the throughput performance of multiple paths is determined by both the number of cross-interference links between paths and the local density of cross path links. These observations can be used in selecting multiple paths for good throughput performance of multipath transport. / The second mechanism we focus on is the rate-based flow control over a multi-hop path. Recognising the lack of suitable bandwidth estimation techniques for multi-hop wireless networks, we propose a new bandwidth estimation technique (termed Saturation Throughput Estimate or STE) suitable for wireless ad-hoc networks. It is based on a nearly linear relationship between channel occupancy (a local measurement of the channel activity level) and throughput available to a node. It also accounts for the traffic forwarded by the neighbouring nodes on behalf of the node in question. We show that our estimation technique is capable of providing good estimates of bandwidth available to a node in small to medium size ad-hoc networks, over a broad range of traffic loads. We also develop and verify an analytical model of IEEE 802.11 suitable for studying 802.11 performance under limited load conditions. Our analytical studies confirm the key assumption of a nearly linear relationship between channel occupancy and throughput underlying the proposed STE bandwidth estimation technique, and provide useful insights into selecting a suitable operating point to achieve high channel utilisation and avoid congestion in the wireless channel. Subsequently, the proposed STE bandwidth estimation technique is fine-tuned to meet the requirements of rate-based flow control over a multi-hop path. Finally, we present and develop a design of a feedback rate-based flow control framework to evaluate the merit of our STE technique and other rate estimation techniques proposed in the subject literature. We show that ATP (Ad-Hoc Transport Protocol) rate estimation technique tends to saturate the network, resulting in long queuing delay. In contrast, the rate-based flow control scheme using our STE rate estimate as rate feedback is effective in regulating and maintaining traffic just below the congestion point, thus achieving high network utilisation in static and low mobility networks. / Thesis (PhDTelecommunications)--University of South Australia, 2006.

Multicast is an efficient way of disseminating information to a group of receivers on the Internet and other computer networks. To combat losses, transmission repetition is common in both best effort and reliable multicast protocols. In the latter, repetition is performed by an automatic repeat request (ARQ) scheme. In either case, forward error correction (FEC) coding can be used to replace repetition, which may result in remarkably improved performance in terms of user throughput. This improvement is especially significant for wireless networks where bandwidth is valuable and loss rate is high. The main goal of this research is to find efficient FEC and hybrid FEC/ARQ (known as HARQ) schemes for a class of message-based multicast protocols, which are compatible with the protocols operating in a wireless environment. / In this thesis, firstly, general multicast and error control techniques are reviewed and examined. A message-based multicast protocol P_Mul is used as a base for performance studies and protocol modifications. P_Mul is designed to handle receivers operating in responsive and silent (no feedback) operational modes. An OPNET model of the standard P_Mul error recovery scheme, originally based on ARQ, is developed. Various schemes such as an ideal (rateless) FEC, Reed-Solomon and LDPC codes are incorporated in the model and their behaviours simulated, discussed and compared with the standard P_Mul performance. / Novel message-based mathematical models are derived that allow accurate estimation of P_Mul performance in the case of finite-length message multicast. The mathematical models take into account certain scenarios and schemes, e.g. when loss is random and all receivers non-responsive or responsive with error recovery based on ARQ or its hybrid with ideal FEC. / Furthermore, the simulation model is expanded so that P_Mul can efficiently manage complex traffic with different priorities; seamlessly serve multicast groups whose members may change operational modes frequently; and adapt to the channel conditions without saturating the network with an excessive number of packets. The performance of the P_Mul model with various enhancements is progressively simulated, compared and argued, pointing to the influence of the various protocol and network parameters. / Finally, a novel method for improving the performance of short messages at high error rates is proposed and the preliminary performance analysed. The method applies erasure decoding on smaller payload portions, called packet segments, to better match with the high error rates, yet without shortening the packet to the lower layer. With some increase in processing complexity and error detection overhead, the chance of short message reception improves considerably / Thesis (PhDTelecommunications)--University of South Australia, 2005.