Distributed maximum likelihood for self-localization in sensor networks

We show that the sensor localization problem can be cast as a static parameter estimation problem for Hidden Markov Models and we develop fully decentralized versions of the Recursive Maximum Likelihood and the Expectation-Maximization algorithms to localize the network. For linear Gaussian models, our algorithms can be implemented exactly using a distributed version of the Kalman filter and a message passing algorithm to propagate the derivatives of the likelihood. In the non-linear case, a solution based on local linearization in the spirit of the Extended Kalman Filter is proposed. In numerical examples we show that the developed algorithms are able to learn the localization parameters well.

Self-similarity limits of genomic signatures

It is shown that metric representation of DNA sequences is one-to-one. By using the metric representation method, suppression of nucleotide strings in the DNA sequences is determined. For a DNA sequence, an optimal string length to display genomic signature in chaos game representation is obtained by eliminating effects of the finite sequence. The optimal string length is further shown as a self-similarity limit in computing information dimension. By using the method, self-similarity limits of bacteria complete genomic signatures are further determined.

A generalized model for the effective thermal conductivity of porous media based on self-similarity

A generalized model for the effective thermal conductivity of porous media is derived based on the fact that statistical self-similarity exists in porous media. The proposed model assumes that porous media consist of two portions: randomly distributed non-touching particles and self-similarly distributed particles contacting each other with resistance. The latter are simulated by Sierpinski carpets with side length L = 13 and cutout size C = 3, 5, 7 and 9, respectively, depending upon the porosity concerned. Recursive formulae are presented and expressed as a function of porosity, ratio of areas, ratio of component thermal conductivities and contact resistance, and there is no empirical constant and every parameter has a clear physical meaning. The model predictions are compared with the existing experimental data, and good agreement is found in a wide range of porosity of 0.14-0.80, and this verifies the validity of the proposed model.

Experimental studies on self-ignition of hydrogen/air supersonic combustion

Self-ignition tests of a model scramjet combustor were conducted by using parallel sonic injection of gaseous hydrogen from the base of a blade-like strut into a supersonic airstream, The vitiated air was produced by burning H-2, O-2, and air to a stagnation temperature of 1000-2100 K and a stagnation pressure of 0.8-1.6 MPa, The effects of different parameters on the self-ignition limits were analyzed, In addition, the effects of the combustor's different wall configurations on self-ignition limits were specifically studied. It was found that the wall configurations of the combustor had a significant effect on self-ignition limits, which might have variations of 420-840 K deg in stagnation temperature; however, the local static temperature in the recirculation zones for different wall configurations remained the same at approximately 1100 K, It was found that self-ignition could initiate at the exit of the combustor and this can be considered as a weak self-ignition characteristic.