Regularization techniques for ill-posed inverse problems in data assimilation

Optimal state estimation from given observations of a dynamical system by data assimilation is generally an ill-posed inverse problem. In order to solve the problem, a standard Tikhonov, or L2, regularization is used, based on certain statistical assumptions on the errors in the data. The regularization term constrains the estimate of the state to remain close to a prior estimate. In the presence of model error, this approach does not capture the initial state of the system accurately, as the initial state estimate is derived by minimizing the average error between the model predictions and the observations over a time window. Here we examine an alternative L1 regularization technique that has proved valuable in image processing. We show that for examples of flow with sharp fronts and shocks, the L1 regularization technique performs more accurately than standard L2 regularization.

An inverse method for determining source characteristics for emergency response applications

Following a malicious or accidental atmospheric release in an outdoor environment it is essential for first responders to ensure safety by identifying areas where human life may be in danger. For this to happen quickly, reliable information is needed on the source strength and location, and the type of chemical agent released. We present here an inverse modelling technique that estimates the source strength and location of such a release, together with the uncertainty in those estimates, using a limited number of measurements of concentration from a network of chemical sensors considering a single, steady, ground-level source. The technique is evaluated using data from a set of dispersion experiments conducted in a meteorological wind tunnel, where simultaneous measurements of concentration time series were obtained in the plume from a ground-level point-source emission of a passive tracer. In particular, we analyze the response to the number of sensors deployed and their arrangement, and to sampling and model errors. We find that the inverse algorithm can generate acceptable estimates of the source characteristics with as few as four sensors, providing these are well-placed and that the sampling error is controlled. Configurations with at least three sensors in a profile across the plume were found to be superior to other arrangements examined. Analysis of the influence of sampling error due to the use of short averaging times showed that the uncertainty in the source estimates grew as the sampling time decreased. This demonstrated that averaging times greater than about 5min (full scale time) lead to acceptable accuracy.

Retrieval of mesospheric electron densities using an optimal estimation inverse method

We present a new method to determine mesospheric electron densities from partially reflected medium frequency radar pulses. The technique uses an optimal estimation inverse method and retrieves both an electron density profile and a gradient electron density profile. As well as accounting for the absorption of the two magnetoionic modes formed by ionospheric birefringence of each radar pulse, the forward model of the retrieval parameterises possible Fresnel scatter of each mode by fine electronic structure, phase changes of each mode due to Faraday rotation and the dependence of the amplitudes of the backscattered modes upon pulse width. Validation results indicate that known profiles can be retrieved and that χ2 tests upon retrieval parameters satisfy validity criteria. Application to measurements shows that retrieved electron density profiles are consistent with accepted ideas about seasonal variability of electron densities and their dependence upon nitric oxide production and transport.

Inverse problems in neural field theory

We study inverse problems in neural field theory, i.e., the construction of synaptic weight kernels yielding a prescribed neural field dynamics. We address the issues of existence, uniqueness, and stability of solutions to the inverse problem for the Amari neural field equation as a special case, and prove that these problems are generally ill-posed. In order to construct solutions to the inverse problem, we first recast the Amari equation into a linear perceptron equation in an infinite-dimensional Banach or Hilbert space. In a second step, we construct sets of biorthogonal function systems allowing the approximation of synaptic weight kernels by a generalized Hebbian learning rule. Numerically, this construction is implemented by the Moore–Penrose pseudoinverse method. We demonstrate the instability of these solutions and use the Tikhonov regularization method for stabilization and to prevent numerical overfitting. We illustrate the stable construction of kernels by means of three instructive examples.

Speech-like and non-speech lip kinematics and coordination in aphasia

Background and aims: In addition to the well-known linguistic processing impairments in aphasia, oro-motor skills and articulatory implementation of speech segments are reported to be compromised to some degree in most types of aphasia. This study aimed to identify differences in the characteristics and coordination of lip movements in the production of a bilabial closure gesture between speech-like and nonspeech tasks in individuals with aphasia and healthy control subjects. Method and procedure: Upper and lower lip movement data were collected for a speech-like and a nonspeech task using an AG 100 EMMA system from five individuals with aphasia and five age and gender matched control subjects. Each task was produced at two rate conditions (normal and fast), and in a familiar and a less-familiar manner. Single articulator kinematic parameters (peak velocity, amplitude, duration, and cyclic spatio-temporal index) and multi-articulator coordination indices (average relative phase and variability of relative phase) were measured to characterize lip movements. Outcome and results: The results showed that when the two lips had similar task goals (bilabial closure) in speech-like versus nonspeech task, kinematic and coordination characteristics were not found to be different. However, when changes in rate were imposed on the bilabial gesture, only speech-like task showed functional adaptations, indicated by a greater decrease in amplitude and duration at fast rates. In terms of group differences, individuals with aphasia showed smaller amplitudes and longer movement durations for upper lip, higher spatio-temporal variability for both lips, and higher variability in lip coordination than the control speakers. Rate was an important factor in distinguishing the two groups, and individuals with aphasia were limited in implementing the rate changes. Conclusion and implications: The findings support the notion of subtle but robust differences in motor control characteristics between individuals with aphasia and the control participants, even in the context of producing bilabial closing gestures for a relatively simple speech-like task. The findings also highlight the functional differences between speech-like and nonspeech tasks, despite a common movement coordination goal for bilabial closure.

Digital predistorter design using B-Spline neural network and inverse of De Boor algorithm

This contribution introduces a new digital predistorter to compensate serious distortions caused by memory high power amplifiers (HPAs) which exhibit output saturation characteristics. The proposed design is based on direct learning using a data-driven B-spline Wiener system modeling approach. The nonlinear HPA with memory is first identified based on the B-spline neural network model using the Gauss-Newton algorithm, which incorporates the efficient De Boor algorithm with both B-spline curve and first derivative recursions. The estimated Wiener HPA model is then used to design the Hammerstein predistorter. In particular, the inverse of the amplitude distortion of the HPA's static nonlinearity can be calculated effectively using the Newton-Raphson formula based on the inverse of De Boor algorithm. A major advantage of this approach is that both the Wiener HPA identification and the Hammerstein predistorter inverse can be achieved very efficiently and accurately. Simulation results obtained are presented to demonstrate the effectiveness of this novel digital predistorter design.

To eat or not to eat? Kinematics and muscle activity of reach-to-grasp movements are influenced by the action goal, but observers do not detect these differences

Recent evidence suggests that the mirror neuron system responds to the goals of actions, even when the end of the movement is hidden from view. To investigate whether this predictive ability might be based on the detection of early differences between actions with different outcomes, we used electromyography (EMG) and motion tracking to assess whether two actions with different goals (grasp to eat and grasp to place) differed from each other in their initial reaching phases. In a second experiment, we then tested whether observers could detect early differences and predict the outcome of these movements, based on seeing only part of the actions. Experiment 1 revealed early kinematic differences between the two movements, with grasp-to-eat movements characterised by an earlier peak acceleration, and different grasp position, compared to grasp-to-place movements. There were also significant differences in forearm muscle activity in the reaching phase of the two actions. The behavioural data arising from Experiments 2a and 2b indicated that observers are not able to predict whether an object is going to be brought to the mouth or placed until after the grasp has been completed. This suggests that the early kinematic differences are either not visible to observers, or that they are not used to predict the end-goals of actions. These data are discussed in the context of the mirror neuron system

Uniqueness results for direct and inverse scattering by infinite surfaces in a lossy medium

We consider the Dirichlet boundary-value problem for the Helmholtz equation, Au + x2u = 0, with Imx > 0. in an hrbitrary bounded or unbounded open set C c W. Assuming continuity of the solution up to the boundary and a bound on growth a infinity, that lu(x)l < Cexp (Slxl), for some C > 0 and S~< Imx, we prove that the homogeneous problem has only the trivial salution. With this resnlt we prove uniqueness results for direct and inverse problems of scattering by a bounded or infinite obstacle.

Complex-valued b-spline neural networks for modelling and inverse of wiener systems

Communication signal processing applications often involve complex-valued (CV) functional representations for signals and systems. CV artificial neural networks have been studied theoretically and applied widely in nonlinear signal and data processing [1–11]. Note that most artificial neural networks cannot be automatically extended from the real-valued (RV) domain to the CV domain because the resulting model would in general violate Cauchy-Riemann conditions, and this means that the training algorithms become unusable. A number of analytic functions were introduced for the fully CV multilayer perceptrons (MLP) [4]. A fully CV radial basis function (RBF) nework was introduced in [8] for regression and classification applications. Alternatively, the problem can be avoided by using two RV artificial neural networks, one processing the real part and the other processing the imaginary part of the CV signal/system. A even more challenging problem is the inverse of a CV

The inverse domino effect: are economic reforms contagious?

This article examines whether a country's economic reforms are affected by reforms adopted by other countries. Our theoretical model predicts that reforms are more likely when factors of production are internationally mobile and reforms are pursued in other economies. Using the change in the Index of Economic Freedom as the measure of market-liberalizing reforms and panel data (144 countries, 1995–2006), we test our model. We find evidence of the spillover of reforms. Moreover, consistent with our model, international trade is not a vehicle for the diffusion of economic reforms; rather the most important mechanism is geographical or cultural proximity.

Control and analysis of oriented thin films of lipid inverse bicontinuous cubic phases using grazing incidence small-angle X‑ray scattering

Lipid cubic phases are complex nanostructures that form naturally in a variety of biological systems, with applications including drug delivery and nanotemplating. Most X-ray scattering studies on lipid cubic phases have used unoriented polydomain samples as either bulk gels or suspensions of micrometer-sized cubosomes. We present a method of investigating cubic phases in a new form, as supported thin films that can be analyzed using grazing incidence small-angle X-ray scattering (GISAXS). We present GISAXS data on three lipid systems: phytantriol and two grades of monoolein (research and industrial). The use of thin films brings a number of advantages. First, the samples exhibit a high degree of uniaxial orientation about the substrate normal. Second, the new morphology allows precise control of the substrate mesophase geometry and lattice parameter using a controlled temperature and humidity environment, and we demonstrate the controllable formation of oriented diamond and gyroid inverse bicontinuous cubic along with lamellar phases. Finally, the thin film morphology allows the induction of reversible phase transitions between these mesophase structures by changes in humidity on subminute time scales, and we present timeresolved GISAXS data monitoring these transformations.

Experimental confirmation of transformation pathways between inverse double diamond and gyroid cubic phases

A macroscopically oriented double diamond inverse bicontinuous cubic phase (QIID) of the lipid glycerol monooleate is reversibly converted into a gyroid phase (QIIG). The initial QIID phase is prepared in the form of a film coating the inside of a capillary, deposited under flow, which produces a sample uniaxially oriented with a ⟨110⟩ axis parallel to the symmetry axis of the sample. A transformation is induced by replacing the water within the capillary tube with a solution of poly(ethylene glycol), which draws water out of the QIID sample by osmotic stress. This converts the QIID phase into a QIIG phase with two coexisting orientations, with the ⟨100⟩ and ⟨111⟩ axes parallel to the symmetry axis, as demonstrated by small-angle X-ray scattering. The process can then be reversed, to recover the initial orientation of QIID phase. The epitaxial relation between the two oriented mesophases is consistent with topologypreserving geometric pathways that have previously been hypothesized for the transformation. Furthermore, this has implications for the production of macroscopically oriented QIIG phases, in particular with applications as nanomaterial templates.