There is a high degree of uncertainty regarding the nature of traffic on future integrated service networks. This uncertainty motivates the use of adaptive resource allocation policies that can take advantage of the statistical fluctuations in the traffic demands. The adaptive control mechanisms must be 'lightweight', in terms of their overheads, and scale to potentially large networks with many traffic flows. Adaptive routing is one form of adaptive resource allocation, and this thesis considers the application of Stochastic Learning Automata (SLA) for distributed, lightweight adaptive routing in future integrated service communication networks. The thesis begins with a broad critical review of the use of Artificial Intelligence (AI) techniques applied to the control of communication networks. Detailed simulation models of integrated service networks are then constructed, and learning automata based routing is compared with traditional techniques on large scale networks. Learning automata are examined for the 'Quality-of-Service' (QoS) routing problem in realistic network topologies, where flows may be routed in the network subject to multiple QoS metrics, such as bandwidth and delay. It is found that learning automata based routing gives considerable blocking probability improvements over shortest path routing, despite only using local connectivity information and a simple probabilistic updating strategy. Furthermore, automata are considered for routing in more complex environments spanning issues such as multi-rate traffic, trunk reservation, routing over multiple domains, routing in high bandwidth-delay product networks and the use of learning automata as a background learning process. Automata are also examined for routing of both 'real-time' and 'non-real-time' traffics in an integrated traffic environment, where the non-real-time traffic has access to the bandwidth 'left over' by the real-time traffic. It is found that adopting learning automata for the routing of the real-time traffic may improve the performance to both real and non-real-time traffics under certain conditions. In addition, it is found that one set of learning automata may route both traffic types satisfactorily. Automata are considered for the routing of multicast connections in receiver-oriented, dynamic environments, where receivers may join and leave the multicast sessions dynamically. Automata are shown to be able to minimise the average delay or the total cost of the resulting trees using the appropriate feedback from the environment. Automata provide a distributed solution to the dynamic multicast problem, requiring purely local connectivity information and a simple updating strategy. Finally, automata are considered for the routing of multicast connections that require QoS guarantees, again in receiver-oriented dynamic environments. It is found that the distributed application of learning automata leads to considerably lower blocking probabilities than a shortest path tree approach, due to a combination of load balancing and minimum cost behaviour.

The problem of edge disjoint path routing arises from applications in distributed memory parallel computing. We examine this problem in both the directed hypercube and two-dimensional grid topologies. Complexity results are obtained for these problems where the routing must consist entirely of shortest length paths. Additionally, approximation algorithms are presented for the case when the routing request is of a special form known as a permutation. Permutations simply require that no vertex in the graph may be used more than once as either a source or target for a routing request. Szymanski conjectured that permutations are always routable in the directed hypercube, and this remains an open problem.

IP address lookup is an important processing function of Internet routers. The challenge lies in finding the longest prefix that matches the packets destination address. One of the issues concerning IP address lookups is the average lookup time. In previous works, caching was shown to be an effective method to minimize the average lookup time. Caching involves storing information on recent IP lookup results in order to decrease average lookup times. In this thesis, we present two architectures that contain a prefix cache and a dynamic substride cache. The dynamic substride cache stores longest possible substrides from previous lookups, and is used in conjunction with a prefix cache. Successful hits in both the caches help reduce the number of worst-case lookups in the low level memory containing the IP routing table in a trie data structure. From simulations, we show that the two architectures show up to 99.9%global hit rate. Furthermore we present analytical models to find optimal designs for the two architectures. We also show that the architectures can support incremental updates once appropriate modifications are made to the trie data structure.

Multi-hop wireless mesh networks (WMNs) provide a cost-effective means to enable broadband wireless access (BWA) services to end users. Such WMNs are required to support different classes of traffic where each class requires certain quality of service (QoS) levels. The research direction undertaken in this thesis considers the development of enhanced routing and scheduling algorithms that enable WMNs to support various QoS metrics for the served traffic. A fundamental class of routing problems in WMNs asks whether a given end-to-end flow that requires certain bandwidth, and benefits from routing over a single path (also called non-bifurcated routing), can be routed given that some ongoing flows are being served in the network. In the thesis, we focus on the development of combinatorial algorithms for solving such incremental non-bifurcated problems for two types of WMNs: 1. WMNs where mesh routers use contention-based protocol for medium access control (MAC), and 2. WMNs where mesh routers use time division multiple access (TDMA) for MAC. For WMNs employing contention-based MAC protocols, we present a novel non-bifurcated routing algorithm that employs techniques from the theory of network flows. The main ingredient in our algorithm is a method for computing interference-constrained flow augmenting paths for routing subscriber demands in the network. For WMNs employing TDMA, we develop a number of joint routing and scheduling algorithms, and investigate the use of such algorithms to maximize the number of served flows. In chapter 4, we consider a throughput maximization problem in the well-known class of grid WMNs. We present an iterative algorithm that strives to achieve high throughput by considering routing and scheduling a pair of distinct flows simultaneously to the gateway in each iteration. In chapter 5, we explore joint routing and scheduling in TDMA-based WMNs with arbitrary topologies, and devise an algorithm that can deal with arbitrary interference relations among pairs of transmission links. In particular, our devised algorithm solves a generalized problem where a cost value is associated with using any possible time-slot on any transmission link, and a minimum cost route is sought along which a new flow can be scheduled without perturbing existing slot assignments.

Due to the rapid development of today¡¦s multimedia communication systems, the complexity and scale of the systems increase day after day. For real-time computing of the systems which become more and more complicated, not only can we use VLSI chips, with the growth of manufacturing techniques of Integrated Circuit, we can apply the Reconfigurable Process Unit to improve real-time computing. Reconfigurable Process Unit is characterized by less cost in research and production as well as less time spent in research and development. Simultaneously, it processes more flexibility than VLSI chips and more suitability in taking advantageous position of real-time computing on an unspecified multimedia communication system. Fine-grain Multi-context Reconfigurable Process Unit has a mechanism of multi-context; therefore, it will take less time when the system reconfigures. This thesis deals with system environment of Computer-Aided Design under the structure of FMRPU, focusing on the placement and routing based on block partition method and designing an automatic optimization mechanism in accordance with historical records to elevate the rate of routable circuit. With the spirit from various existing algorithm of circuit, we add the factors of block partition, which forms the implements of placement and routing based on block partition. Combined clustering and the limit caused by the hardware structure of FMRRPU, we can have an accurate block partition on FMRPU. Through the continual increase of historical records, the assessment for the upper limit of the argument of routable circuit will get closer to the actual figure. Simultaneously, after the Logic Block Partition, the probability of routable circuit will get great assurance, and the time consumed in lots of repetitious computing on un-routable circuit will decrease. The experimental result reveals that the modified placement cost function can obtain enormous improvement under the comparison with that mentioned the master thesis of Tzu-che Huang. Not only the routability steps up, the unnecessary consumption also reduces largely. In routing, the negotiated congestion-delay algorithm produced on the basis of the transformation of maze routing algorithm has great suitability in the operation on FMRPU, which has many optimization goals and limited routing resource. After the redefinition of the cost function and expenditure for routing, we can operate with accuracy and the time spent on the delayed circuit will decrease.

The work presented in this thesis is motivated by the dual goal of developing a scalable and efficient approach for IP lookup using both hardware and software approach. The work involved designing algorithms and techniques to increase the capacity and flexibility of the Internet. The Internet is comprised of routers that forward the Internet packets to the destination address and the physical links that transfer data from one router to another. The optical technologies have improved significantly over the years and hence the data link capacities have increased. However, the packet forwarding rates at the router have failed to keep up with the link capacities. Every router performs a packet-forwarding decision on the incoming packet to determine the packet??s next-hop router. This is achieved by looking up the destination address of the incoming packet in the forwarding table. Besides increased inter-packet arrival rates, the increasing routing table sizes and complexity of forwarding algorithms have made routers a bottleneck in the packet transmission across the Internet. A number of solutions have been proposed that have addressed this problem. The solutions have been categorized into hardware and software solutions. Various lookup algorithms have been proposed to tackle this problem using software approaches. These approaches have proved more scalable and practicable. However, they don??t seem to be able to catch up with the link rates. The first part of my thesis discusses one such software solution for routing lookup. The hardware approaches today have been able to match up with the link speeds. However, these solutions are unable to keep up with the increasing number of routing table entries and the power consumed. The second part of my thesis describes a hardware-based solution that provides a bound on the power consumption and reduces the number of entries required to be stored in the routing table.

This paper proposes a hybrid routing protocol based on the k-hop clustering structure for MANETs. The source sends packets to the destination directly if it is in the source¡¦s neighbor table; otherwise the source reactively sends the route request packet (RREQ) to trigger the routing process. Instead of adopting the nodes that forwards the RREQ as the route between the source and destination, the route reply packet (RREP) is broadcasted via the nodes in the clusters that the RREQ has ever passed to find the route. The route constructed in this way can avoid the clusterheads always being in the transmission route and collapsing due to overloading. In comparison with the Cluster Based Routing Protocol (CBRP), the proposed protocol can distribute the communication workload from the clusterheads to member nodes. Simulation results demonstrate that the proposed protocol has better packet delivery ratio and end-to-end delay time than that of CBRP.

The trend to integrate more and more processing cores and memory cores into a single module has increased the overall size of chips to the point where global interconnects between sub-units are becoming harder and harder to route and meet timing rules and requirements. The traditional way of routing interconnects and the use of uniform, unidirectional, point to point busses may no longer be optimal for certain designs where metal layers and chip area for interconnects are limited. The need for a more flexible routing methodology is necessary and can be achieved by using routing and calibration techniques currently being implemented at board level design. This report proposes the use of non-uniform, bidirectional, and possibly multi-point loads global interconnects within a single chip module through the use of on chip calibration techniques to compensate for less restrictive wiring rules for certain chip designs. This report will also apply a widely used digital calibration technique to simulate the implementation on a field programmable gate array. / text

No description available.

Peer-to-Peer(P2P) systems have emerged as a promising paradigm to structure large scale distributed systems. They provide a robust, scalable and decentralized way to share and publish data.The unstructured P2P systems have gained much popularity in recent years for their wide applicability and simplicity. However efficient resource discovery remains a fundamental challenge for unstructured P2P networks due to the lack of a network structure. To effectively harness the power of unstructured P2P systems, the challenges in distributed knowledge management and information search need to be overcome. Current attempts to solve the problems pertaining to knowledge management and search have focused on simple term based routing indices and keyword search queries. Many P2P resource discovery applications will require more complex query functionality, as users will publish semantically rich data and need efficiently content location algorithms that find target content at moderate cost. Therefore, effective knowledge and data management techniques and search tools for information retrieval are imperative and lasting. In my dissertation, I present a suite of protocols that assist in efficient content location and knowledge management in unstructured Peer-to-Peer overlays. The basis of these schemes is their ability to learn from past peer interactions and increasing their performance with time.My work aims to provide effective and bandwidth-efficient searching and data sharing in unstructured P2P environments. A suite of algorithms which provide peers in unstructured P2P overlays with the state necessary in order to efficiently locate, disseminate and replicate objects is presented. Also, Existing approaches to federated search are adapted and new methods are developed for semantic knowledge representation, resource selection, and knowledge evolution for efficient search in dynamic and distributed P2P network environments. Furthermore,autonomous and decentralized algorithms that reorganizes an unstructured network topology into a one with desired search-enhancing properties are proposed in a network evolution model to facilitate effective and efficient semantic search in dynamic environments.