This paper concerns the definition of an analytical model and distributed protocols for determining the routes of a mobile data collector (sink) traveling through the nodes of a wireless sensor network (WSN). The routes are determined with the overall aim of maximizing the network lifetime. The contribution of our work is twofold. First, we introduce a novel mixed integer linear programming formulation for determining the sink's route and the sojourn time at the different "sink sites." The model takes into account realistic parameters such as the maximum distance the sink can travel between sites, different sink mobility rates, as well as the costs to support and perform data routing. Solutions to the model provide the route of the sink as a sequence of sites and the sojourn times at those sites that induce the maximum network lifetime. We then propose the Greedy Maximum Residual Energy (GMRE) protocol for sink mobility. GMRE is distributed and localized, thus being suitable for wireless sensor networking. In GMRE the sink greedily keeps moving toward those areas in the network where there is the most residual energy, as if "drawn" to them. This heuristic is then compared with a very simple and energy-unaware protocol where the next site in the sink route is chosen randomly and uniformly each time the sink moves. Simulation results show that GMRE leads to improvements in network lifetime that are four times as much as the lifetime when the sink is kept static, while balancing energy consumption throughout the network.

Using bi-criteria decision making analysis, a new model for test suite minimization has been developed that pursues two objectives: minimizing a test suite with regard to a particular level of coverage while simultaneously maximizing error detection rates. This new representation makes it possible to achieve significant reductions in test suite size without experiencing a decrease in error detection rates. Using the all-uses inter-procedural data flow testing criterion, two binary integer linear programming models were evaluated, one a single-objective model, the other a weighted-sums bi-criteria model. The applicability of the bi-criteria model to regression test suite maintenance was also evaluated. The data show that minimization based solely on definition-use association coverage may have a negative impact on the error detection rate as compared to minimization performed with a bi-criteria model that also takes into account the ability of test cases to reveal error. Results obtained with the bi-criteria model also indicate that test suites minimized with respect to a collection of program faults are effective at revealing subsequent program faults.

This paper addresses the problem of determining a single facility location on an undirected tree with n nodes and edges that fail independently with probability (1 − pe) so that service provided to demand points can be achieved with reliability. Termed the relisum problem, the objective is to find a network location that maximizes the expected number of nodes reachable by operational paths. A decomposition formula is developed for which several properties are analytically derived and realistic vertex optimality conditions are identified. Two polynomial algorithms are presented: an O(n3) algorithm which is an adaptation of the Floyd-Warshall algorithm for finding all pairwise shortest paths in a graph and an O(n2) algorithm based on a depth-first node traversal and the decomposition nature of an operational path. Computational results are provided, and sensitivity analysis ranges and marginal values for pe are analytically derived. Properties and algorithms for the weighted version of the problem are discussed, as well as a polynomial heuristic for finding a relisum node on a cyclic network. © 1996 John Wiley & Sons, Inc.