In this paper we investigate the effect of local error recovery vs. end-to-end error recovery in reactive protocols. For this purpose, we analyze and compare the performance of two protocols: the Dynamic Source Routing protocol (DSR[2]), which does end-to-end error recovery when a route fails and the Witness Aided Routing protocol (WAR[I]), which uses local correction mechanisms to recover from route failures. We show that the performance of DSR degrades extremely fast as the route length increases (that is, DSR is not scalable), while WAR maintains both low latency and low resource consumption regardless of the route length.

Wireless systems, and mobile ad hoc networks in particular, are more likely to experience transmission and routing errors than their wired counterpart. Factors li ke the lack of infrastructure, node mobility, and random radio link quality can contribute to significantly higher error rates in these networks. In addition, errors have a more serious impact on the network's resources, due to limitations in bandwidth and battery power inherent to the wireless ad hoc environment. This further complicates the task of designing scalable routing protocols, since larger networks are likely to experience even more errors, which may lead to slower convergence, longer end-to-end delay and unnacceptably high number of retransmissions. In this paper, we focus on the impact of error prevention and recovery on the scaling properties of on-demand protocols for ad hoc networks. Our analytical study, based on the evaluation of the Witness Aided Routing (WAR) and the Dynamic Source Routing (DSR) protocols, shows that the lack of localized intervention in handling errors translates eventually into lack of scalability, both in terms of performance and resource consumption. As route length increases, the performance of DSR degrades dramatically, especially in the presence of fluctuating wireless link quality. Even for small routes, DSR's lack of an error handling mechanism leads to very low probability of success when there is a non-zero probability that links are not bidirectional. On the other hand, WAR remains relatively insensitive both to the length of the route and to variations in mobility and call rates, and has a higher tolerance to radio link instability. This indicates that localized error correction can increase route effectiveness and alleviate the effects of short-lived radio link problems to an extent that allows the protocol to scale with the network size.2 On The Scalability of On-Demand Routing Protocols further be classified as satellite, cellular or packet radio (ad hoc) network. If both routers and endpoints are mobile, the network is referred to as mobile ad hoc network (or MANET, a term adopted by the IEEE 802.11 9 committee). This type of network is the most challenging in terms of providing connections between any two nodes, because it does not rely on any existing wired infrastructure and all nodes are mobile. The communication between nodes in a network is provided by a routing protocol. The routing protocol is responsible for maintaining a logical infrastructure which will allow communication between any two remote hosts as if they were directly connected. In mobile networks, a routing protocol must also be able to hide adverse effects of mobility and problems in radio links from the mobile nodes.An ideal routing protocol should be able to compute and recompute routes quickly, with minimum computational and transmission overhead. The routes should be the shortest possible, with no loops and should provide a good balancing of traffic in the entire network (to avoid quality of service deterioration). However, a...

One of the central trends in the optimization community over the past several years has been the steady improvement of general-purpose solvers. A logical next step in this evolution is to combine mixed-integer linear programming, constraint programming, and global optimization in a single system. Recent research in the area of integrated problem solving suggests that the right combination of different technologies can simplify modeling and speed up computation substantially. Nevertheless, integration often requires special-purpose coding, which is time consuming and error prone. We present a general-purpose solver, SIMPL, that allows its user to replicate (and sometimes improve on) the results of custom implementations with concise models written in a high-level language. We apply SIMPL to production planning, product configuration, machine scheduling, and truss structure design problems on which customized integrated methods have shown significant computational advantage. We obtain results that either match or surpass the original codes at a fraction of the implementation effort.

No abstract

We describe a specialized dynamic programming algorithm for factory crane scheduling. An innovative data structure controls the memory requirements of the state space and enables solution of problems of realistic size. The algorithm finds optimal space-time trajectories for two factory cranes or hoists that move along a single overhead track. Each crane is assigned a sequence of pickups and deliveries at specified locations that must be performed within given time windows. The cranes must not interfere with each other, although one may yield to the other. The state space representation permits a wide variety of constraints and objective functions. It is stored in a compressed data structure that uses a cartesian product of intervals of states and an array of two-dimensional circular queues. We also show that only certain types of trajectories need be considered. The algorithm found optimal solutions in less than a minute for medium-sized instances of the problem (160 tasks, spanning four hours). It can also be used to create benchmarks for tuning heuristic algorithms that solve larger instances.