Uniform recovery from subgaussian multi-sensor measurements

Parallel acquisition systems are employed successfully in a variety of different sensing applications when a single sensor cannot provide enough measurements for a high-quality reconstruction. In this paper, we consider compressed sensing (CS) for parallel acquisition systems when the individual sensors use subgaussian random sampling. Our main results are a series of uniform recovery guarantees which relate the number of measurements required to the basis in which the solution is sparse and certain characteristics of the multi-sensor system, known as sensor profile matrices. In particular, we derive sufficient conditions for optimal recovery, in the sense that the number of measurements required per sensor decreases linearly with the total number of sensors, and demonstrate explicit examples of multi-sensor systems for which this holds. We establish these results by proving the so-called Asymmetric Restricted Isometry Property (ARIP) for the sensing system and use this to derive both nonuniversal and universal recovery guarantees. Compared to existing work, our results not only lead to better stability and robustness estimates but also provide simpler and sharper constants in the measurement conditions. Finally, we show how the problem of CS with block-diagonal sensing matrices can be viewed as a particular case of our multi-sensor framework. Specializing our results to this setting leads to a recovery guarantee that is at least as good as existing results.     

Real-time data gathering in sensor networks

Wireless sensor networks represent a new generation of real-time traffic communications and high data rate sensor applications, such as structural health monitoring and control. We study some problems related to data gathering in sensor networks when the sensors collect the sensed data about their environment and this information should be delivered to a collecting central Base Station. We prove that scheduling messages through the network to minimize the maximal delivery time with restrictions on the total idle time allowed is NP-hard. We also refer to a special case of linear network topology for which we present two polynomial time optimization algorithms: One is for minimizing the maximal lateness and maximal delay, while the other is for minimizing the number of tardy messages.     

A note on power domination in grid graphs

The problem of monitoring an electric power system by placing as few measurement devices in the system as possible is closely related to the well known vertex covering and dominating set problems in graphs (see [T.W. Haynes, S.M. Hedetniemi, S.T. Hedetniemi, M.A. Henning, Power domination in graphs applied to electrical power networks, SIAM J. Discrete Math. 15(4) (2002) 519-529]). A set S of vertices is defined to be a power dominating set of a graph if every vertex and every edge in the system is monitored by the set S (following a set of rules for power system monitoring). The minimum cardinality of a power dominating set of a graph is its power domination number. In this paper, we determine the power domination number of an n×m grid graph.     

On the number of range queries in k-space

A range query on a set of points in a k-dimensions coordinate space asks for all points lying within a hyper-rectangle specified by ranges of permissible values for each of the coordinates. In this paper we regard as identical any two range queries which return the same set of points. We then investigate the number of range queries possible on a set—given a set of N points in k-space, what is the maximum number of distinct subsets that may be specified by giving bounding hyper-rectangles? The bounds we find for this number (as a function of N and k) are substantial improvements over previous results, and tighten a lower bound on the computational time required to process range queries. We also report some results concerning the expected number of range queries on random distributions of points.     

Coloring Geometric Range Spaces

We study several coloring problems for geometric range-spaces. In addition to their theoretical interest, some of these problems arise in sensor networks. Given a set of points in ?² or ?³, we want to color them so that every region of a certain family (e.g., every disk containing at least a certain number of points) contains points of many (say, k) different colors. In this paper, we think of the number of colors and the number of points as functions of k. Obviously, for a fixed k using k colors, it is not always possible to ensure that every region containing k points has all colors present. Thus, we introduce two types of relaxations: either we allow the number of colors used to increase to c(k), or we require that the number of points in each region increases to p(k).Symmetrically, given a finite set of regions in ?² or ?³, we want to color them so that every point covered by a sufficiently large number of regions is contained in regions of k different colors. This requires the number of covering regions or the number of allowed colors to be greater than k.The goal of this paper is to bound these two functions for several types of region families, such as halfplanes, halfspaces, disks, and pseudo-disks. This is related to previous results of Pach, Tardos, and Tóth on decompositions of coverings.     

Power domination in graphs

The problem of monitoring an electric power system by placing as few measurement devices in the system as possible is closely related to the well-known domination problem in graphs. In 1998, Haynes et al. considered the graph theoretical representation of this problem as a variation of the domination problem. They defined a set S to be a power dominating set of a graph if every vertex and every edge in the system is monitored by the set S (following a set of rules for power system monitoring). The power domination number γ_P(G) of a graph G is the minimum cardinality of a power dominating set of G. In this paper, we present upper bounds on the power domination number for a connected graph with at least three vertices and a connected claw-free cubic graph in terms of their order. The extremal graphs attaining the upper bounds are also characterized.     

Maximum lifetime connected coverage with two active-phase sensors

A sensor with two active phrases means that active mode has two phases, the full-active phase and the semi-active phase, which require different energy consumptions. A full-active sensor can sense data packets, transmit, receive, and relay the data packets. A semi-active sensor cannot sense data packets, but it can transmit, receive, and relay data packets. Given a set of targets and a set of sensors with two active phrases, find a sleep/active schedule of sensors to maximize the time period during which active sensors form a connected coverage set. In this paper, this problem is showed to have polynomial-time ${(7.875+\varepsilon)}$ -approximations for any ${\varepsilon >0 }$ when all targets and sensors lie in the Euclidean plane and all sensors have the same sensing radius R _s and the same communication radius R _c with R _c ≥ 2R _s .     

Optimal placement of stereo sensors

Autonomous wireless devices such as sensor nodes and stereo cameras, due to their low cost of operation coupled with the potential for remote deployment, have found a plethora of applications ranging from monitoring air, soil and water to seismic detection and military surveillance. Typically, such a network spans a region of interest with the individual nodes cooperating to detect events and disseminate information. Given a deployment of sensors and targets over a region, a sensor pairing is desired for each target that optimizes the coverage under certain constraints. This problem can be modeled as an integer programming problem and solved using branch-and-cut. For larger problems, it is necessary to limit the number of variables, and a GRASP routine was developed for this purpose. Valid cutting planes are developed and computational results presented. Research supported in part by NSF grant numbers DMS–0317323 and CMS–0301661.     

Random weights, robust lattice rules and the geometry of the cbcrc algorithm

In this paper we study lattice rules which are cubature formulae to approximate integrands over the unit cube [0,1] ^s from a weighted reproducing kernel Hilbert space. We assume that the weights are independent random variables with a given mean and variance for two reasons stemming from practical applications: (i) It is usually not known in practice how to choose the weights. Thus by assuming that the weights are random variables, we obtain robust constructions (with respect to the weights) of lattice rules. This, to some extend, removes the necessity to carefully choose the weights. (ii) In practice it is convenient to use the same lattice rule for many different integrands. The best choice of weights for each integrand may vary to some degree, hence considering the weights random variables does justice to how lattice rules are used in applications. In this paper the worst-case error is therefore a random variable depending on random weights. We show how one can construct lattice rules which perform well for weights taken from a set with large measure. Such lattice rules are therefore robust with respect to certain changes in the weights. The construction algorithm uses the component-by-component (cbc) idea based on two criteria, one using the mean of the worst case error and the second criterion using a bound on the variance of the worst-case error. We call the new algorithm the cbc2c (component-by-component with 2 constraints) algorithm. We also study a generalized version which uses r constraints which we call the cbcrc (component-by-component with r constraints) algorithm. We show that lattice rules generated by the cbcrc algorithm simultaneously work well for all weights in a subspace spanned by the chosen weights ?? ⁽¹⁾, . . . , ?? ^(r). Thus, in applications, instead of finding one set of weights, it is enough to find a convex polytope in which the optimal weights lie. The price for this method is a factor r in the upper bound on the error and in the construction cost of the lattice rule. Thus the burden of determining one set of weights very precisely can be shifted to the construction of good lattice rules. Numerical results indicate the benefit of using the cbc2c algorithm for certain choices of weights.     

α-Coverage to extend network lifetime on wireless sensor networks

An important problem in the context of wireless sensor networks is the Maximum Network Lifetime Problem (MLP): find a collection of subset of sensors (cover) each covering the whole set of targets and assign them an activation time so that network lifetime is maximized. In this paper we consider a variant of MLP, where we allow each cover to neglect a certain fraction (1 ? α) of the targets. We analyze the problem and show that the total network lifetime can be hugely improved by neglecting a very small portion of the targets. An exact approach, based on a Column Generation scheme, is presented and a heuristic solution algorithm is also provided to initialize the approach. The proposed approaches are tested on a wide set of instances. The experimentation shows the effectiveness of both the proposed problems and solution algorithms in extending network lifetime and improving target coverage time when some regularity conditions are taken into account.     

Approximation schemes for parallel machine scheduling with availability constraints

We investigate the problems of scheduling n weighted jobs to m parallel machines with availability constraints. We consider two different models of availability constraints: the preventive model, in which the unavailability is due to preventive machine maintenance, and the fixed job model, in which the unavailability is due to a priori assignment of some of the n jobs to certain machines at certain times. Both models have applications such as turnaround scheduling or overlay computing. In both models, the objective is to minimize the total weighted completion time. We assume that m is a constant, and that the jobs are non-resumable.For the preventive model, it has been shown that there is no approximation algorithm if all machines have unavailable intervals even if w_i=p_i for all jobs. In this paper, we assume that there is one machine that is permanently available and that the processing time of each job is equal to its weight for all jobs. We develop the first polynomial-time approximation scheme (PTAS) when there is a constant number of unavailable intervals. One main feature of our algorithm is that the classification of large and small jobs is with respect to each individual interval, and thus not fixed. This classification allows us (1) to enumerate the assignments of large jobs efficiently; and (2) to move small jobs around without increasing the objective value too much, and thus derive our PTAS. Next, we show that there is no fully polynomial-time approximation scheme (FPTAS) in this case unless P=NP.For the fixed job model, it has been shown that if job weights are arbitrary then there is no constant approximation for a single machine with 2 fixed jobs or for two machines with one fixed job on each machine, unless P=NP. In this paper, we assume that the weight of a job is the same as its processing time for all jobs. We show that the PTAS for the preventive model can be extended to solve this problem when the number of fixed jobs and the number of machines are both constants.     

Coprime ordering of cyclic planar difference sets

Motivated by a question about uniform dessins d’enfants, it is conjectured that every cyclic planar difference set of prime power order m≠4 can be cyclically ordered such that the difference of every pair of neighbouring elements is coprime to the module v?m²+m+1. We prove that this is the case whenever the number ω(v) of different prime divisors of v is less than or equal to 3. To achieve this we consider a graph related to the difference set and show that it is Hamiltonian.     

Convex solution of a permutation problem

In this paper, we show that a problem of finding a permuted version of k vectors from R^N such that they belong to a prescribed rank r subset, can be solved by convex optimization. We prove that under certain generic conditions, the wanted permutation matrix is unique in the convex set of doubly-stochastic matrices. In particular, this implies a solution of the classical correspondence problem of finding a permutation that transforms one collection of points in R^k into the another one. Solutions to these problems have a wide set of applications in Engineering and Computer Science.     

Calculating the Number of Functions with a Given Endomorphism

An iterative procedure is proposed for calculating the number of k-valued functions of n variables such that each one has an endomorphism different from any constant and permutation. Based on this procedure, formulas are found for the number of three-valued functions of n variables such that each one has nontrivial endomorphisms. For any arbitrary semigroup of endomorphisms, the power is found of the set of all three-valued functions of n variables such that each one has endomorphisms from a specified semigroup.     

Maximum transversal in partial Latin squares and rainbow matchings

In a partial Latin square P a set of distinct entries, such that no two of which are in the same row or column is called a transversal. By the size of a transversal T, we mean the number of its entries. We define a duplex to be a partial Latin square of order n containing 2n entries such that exactly two entries lie in each row and column and each of n symbols occurs exactly twice. We show that determining the maximum size of a transversal in a given duplex is an NP-complete problem. This problem relates to independent sets in certain subfamilies of cubic graphs. Generalizing the concept of transversals in edge coloring of graphs we are led to introduce the concept of rainbow matching. We show that if each color appears at most twice then it is a polynomial time problem to know whether there exists a rainbow matching of size at least ⌊n/2⌋-t for each fixed t, where n is the order of the graph. As an application we show that for any fixed t, there is a polynomial time algorithm which decides whether α(G)?n-t, for any graph G on 2n vertices containing a perfect matching. At the end we mention some other applications of rainbow matching.     

Approximation algorithms for free-label maximization

Inspired by air-traffic control and other applications where moving objects have to be labeled, we consider the following (static) point-labeling problem: given a set P of n points in the plane and labels that are unit squares, place a label with each point in P in such a way that the number of free labels (labels not intersecting any other label) is maximized. We develop efficient constant-factor approximation algorithms for this problem, as well as PTASs, for various label-placement models.     

The N-policy of a discrete time Geo/G/1 queue with disasters and its application to wireless sensor networks

We analyze an N-policy of a discrete time Geo/G/1 queue with disasters. We obtain the probability generating functions of the queue length, the sojourn time, and regeneration cycles such as the idle period and the busy period. We apply the queue to a power saving scheme in wireless sensor networks under unreliable network connections where data packets are lost by external attacks or shocks. We present various numerical experiments for application to power consumption control in wireless sensor networks. We investigate the characteristics of the optimal N-policy that minimizes power consumption and derive practical insights on the operation of the N-policy in wireless sensor networks.     

Unconditionally secure key assignment schemes

In this paper we propose an information-theoretic approach to the access control problem in a scenario where a group of users is divided into a number of disjoint classes. The set of rules that specify the information flow between different user classes in the system defines an access control policy. An access control policy can be implemented by using a key assignment scheme, where a trusted central authority (CA) assigns an encryption key and some private information to each class.We consider key assignment schemes where the key assigned to each class is unconditionally secure with respect to an adversary controlling a coalition of classes of a limited size. Our schemes are characterized by a security parameter r, the size of the adversary coalition. We show lower bounds on the size of the private information that each class has to store and on the amount of randomness needed by the CA to set up any key assignment scheme. Finally, we propose some optimal constructions for unconditionally secure key assignment schemes.     

Stability of the Feasible Set Mapping of Linear Systems with an Exact Constraint Set

This paper deals with the stability of the feasible set mapping of linear systems of an arbitrary number (possibly infinite) of equations and inequalities such that the variable x ranges on a certain fixed constraint set X?? ⁿ (X could represent the solution set of a given constraint system, e.g., the positive cone of ? ⁿ in the case of sign constraints). More in detail, the paper provides necessary as well as sufficient conditions for the lower and upper semicontinuity (in Berge sense), and the closedness, of the set-valued mapping which associates, with any admissible perturbation of the given (nominal) system its feasible set. The parameter space is formed by all the systems having the same structure (i.e., the same number of variables, equations and inequalities) as the nominal one, and the perturbations are measured by means of the pseudometric of the uniform convergence.