Customizing methods for global optimization-a geometric viewpoint

A new class of global optimization algorithms, extending the multidimensional bisection method of Wood, is described geometrically. New results show how the geometry of the global minimum relates to performance. Remarkably, the epigraph of the objective function, turned upside down, plays a key role. Algorithms customized to take advantage of special information about the objective function belong to the class. A number of algorithms in the literature, including those of Piyavskii-Shubert, Mladineo, Wood and Breiman & Cutler, also belong, and simple modifications of them produce customized algorithms. Comparison of various algorithms in the class is provided.Paper presented at the II. IIASA-workshop on Global Optimization, December 9–12, 1990, Sopron (Hungary).     

THE COSTATE VARIABLE IN A STOCHASTIC RENEWABLE RESOURCE MODEL

ABSTRACT. In this paper we discuss the costate variable in a stochastic optimal control model of a renewable natural resource, which we call a fishery. The role of the costate variable in deterministic control models has been discussed extensively in the literature. See for example Lyon [1999], Clark [1990, pp. 102 107] and Arrow and Kurz [1970, pp. 35 37]; however, there is little discussion of this variable for stochastic models, even though thecostate variable has similar roles in the two models. In both models the costate variableis a shadow value of the associated state variable, and as such has the role of rationingthe use of the state variable. In addition, as has been shown in Lyon [1999], in natural resource problems the costate variable can be partitioned into a scarcity effect and a cost effect. We show that this same partitioning can be done in the stochastic renewable resource problem. We discuss and contrast the similarities and differences in these concepts for deterministic and stochastic models. In addition, we present a numerical example to help solidify the results.     

Gauge-invariant lattice gas for the microcanonical Ising model

We introduce a lattice gas for particles with discrete momenta (1, 0, –1) and local deterministic microdynamics, which exactly reproduces Creutz's microcanonical algorithm for the ferromagnetic Ising model. However, because of the manifest gauge invariance of our variables, both the Ising ferromagnetic and spin-glass systems share precisely the same dynamics with different initial conditions. Additional conservation laws in the 1D Ising case result in a completely integrable system in the limit of zero or unbounded demon energy cutoff. Numerical investigations of ergodicity are presented for the pure Ising lattice gas in one and two dimensions.     

Detection and Evaluation of Deterministic Jitter Causes in CP-PLL’s Due to Macro Level Faults and Pre-Detection Using Simple Methods

Charge-Pump Phase-locked loops are currently used in a variety of SoC signal generation applications. They ultimately determine performance of other SoC blocks, such as ADCs, DACs, RF and synchronisation functions. In many situations, only simple frequency lock tests are carried out on the CP-PLL portion of a circuit, with other complex direct jitter tests being carried out indirectly at a higher system level. Although these higher level system tests must generally be carried out at some point they can be time consuming. In addition, if the PLL is designed and operating correctly the PLL system will generally have far better performance than the system it is driving. This paper investigates typical jitter output responses of CP-PLLs when subjected to selected forward path leakage faults. The evaluation platform consists of a macro level mixed signal based PLL-Model. Degradation of the PLL output is evaluated from the phase noise spectrum, jitter spectrum and sideband spur degradation. Further evaluations and analysis are supplied relating block level effects to jitter and phase noise. Investigations are made as to the efficacy of detection of these errors with simple measurement techniques. The crux of the work is thus initially to develop techniques to aid evaluation of the likely jitter performance of a CP-PLL system without resorting to direct measurement techniques.Martin John Burbidge Is currently employed as a lecturer for the Microsystems research group at Lancaster University. He worked towards a HND in mechanical and electrical engineering from 1985 to 1987. From 1993 to 1996 he studied technology, mathematical modelling, and embedded micro controller design with the Open University. He received a first class honours degree in Electronic and Communications Engineering from Lancaster University in 1999. From March 2000, he was employed as a Research associate at Lancaster University and has been investigating and developing innovative test techniques and test paradigm shifts for Phase locked loops and frequency synthesis systems. During this period he has produced eleven conference and six journal publications. He also carries out his own consultancy activities in this area. His general interests include RF board level and ASIC design, design for test, and built in self-test relating to RF circuits.     

Service Curve Based Routing Subject to Deterministic QoS Constraints

We consider the problem of routing in packet-switching networks subject to deterministic QoS constraints with the objective of minimizing induced cost. In particular, we propose a novel blend of generic service curves as abstraction of network transport services, of the associated network calculus under min–plus algebra and of existing heuristics for the restricted shortest path problem. Since the minimum cost objective requires an unambiguous definition of link service cost, we derive a cost model for the deployed generic service curve abstraction. Furthermore we describe the details of the different steps of the developed routing algorithm in pseudo-code notation. Our approach is subdivided into a pre-computation and an on-demand phase. Within the on-demand phase we deploy tabu-search as local search heuristic to exploit the neighbourhood of a pre-computed minimum weight path. The presented approach provides a generic technical solution for minimum-cost routing subject to deterministic QoS constraints. It is specifically designed to provide a solution path whenever the computation is stopped. Although such solutions may be non-optimal and even infeasible with respect to the QoS demand, obtaining a possible path is important in the trade-off of required processing time and accuracy of results.     

'T-RANS' Simulation of Deterministic Eddy Structure in Flows Driven by Thermal Buoyancy and Lorentz Force

The paper reports on the application of the Time-dependent Reynolds-Averaged Navier–Stokes (T-RANS) approach to analysing the effects of magnetic force and bottom-wall configuration on the reorganisation of a large coherent structure and its role in the transport processes in Rayleigh–Bénard convection. The large-scale deterministic motion is fully resolved in time and space, whereas the unresolved stochastic motion is modelled by a `subscale' model for which the conventional algebraic stress/flux expressions were used, closed with the low-Re number (k)-(ε)-(θ²) three-equation model. The applied method reproduces long-term averaged mean flow properties, turbulence second moments, and all major features of the coherent roll/cell structure in classic Rayleigh–Bénard convection in excellent agreement with the available DNS and experimental results. Application of the T-RANS approach to Rayleigh–Bénard convection with wavy bottom walls and a superimposed magnetic field yielded the expected effects on there organisation of the eddy structure and consequent modifications of the mean and turbulence parameters and wall heat transfer. This revised version was published online in August 2006 with corrections to the Cover Date.     

Book review of advances in gabor analysis,applied and numerical harmonic analysis,Hans G. Feichtinger and Thomas Strohmer (Eds.), Birkhäuser,Boston-Basel-Berlin, 2002, 356 pages. ISBN 0-8176-4239-0

In some recent publications it was shown that certain stationary stochastic dynamic programming problems with general state and action spaces can be solved by generalized linear programming. It Is the main aim of the present paper to demonstrate that a similar linear programming approach is feasible even in the non-stationary case. For this end, we formulate a programming problem (D^?) and show that (D^?) is equivalent to the problem of finding a p=optimal policy for the stochastic dynamic program, whereas a modification of (D^?) turns out to be the dual program of a pair of general linear programs.     

Dynamic behaviour of a discrete-time SIR model with information dependent vaccine uptake

This paper proposes and analyzes a discrete-time deterministic SIR model with information dependent immunization behaviour, where vaccination coverage at birth during any period of time is a general phenomenological function of the risk of infection that is perceived at the beginning of the period. Results on existence of equilibria, their local stability, and system persistence are proved. Then, by considering the noteworthy subcase of a piecewise linear ‘prevalence-dependent’ coverage function, the local stability of the endemic state is proved and conditions for its global asymptotic stability are given. Some insight on both Neimarck-Sacher and period-doubling bifurcations are provided. Overall we show that prevalence-dependent coverage is an essentially stabilising force. However period-doubling bifurcations are possible though under stressed parameter constellations.     

无线多媒体传感器网络部署控制算法

已有的无线多媒体传感器网络（WMSNs）研究针对传感器放置在目标区域内的情况进行,且没有考虑节点通过云台的转动获得的整个可能感知区域对覆盖率的影响。本文首先针对节点放置点高于目标区域的应用进行研究,综合考虑节点感知区域和可能感知区域,建立了延时和无延时感知模型,并针对不同的感知模型提出了传感器网络部署控制算法（IVPDCA）,算法中改进了虚拟势场算法,定义了节点质量的概念来表示节点间覆盖重叠的大小,建立受力模型,使得节点在合力作用下进行重新部署,同时关闭冗余节点,既延长了网络寿命,又提高了区域覆盖率。仿真结果验证了算法的有效性。     

A useful technique for piecewise deterministic Markov decision processes

This note presents a technique that is useful for the study of piecewise deterministic Markov decision processes (PDMDPs) with general policies and unbounded transition intensities. This technique produces an auxiliary PDMDP from the original one. The auxiliary PDMDP possesses certain desired properties, which may not be possessed by the original PDMDP. We apply this technique to risk-sensitive PDMDPs with total cost criteria, and comment on its connection with the uniformization technique.