Marco Balsi

Enhanced synchronized switch harvesting: a new energy harvesting scheme for efficient energy extraction

This paper presents a new technique for optimized energy harvesting using piezoelectric microgenerators called enhanced synchronized switch harvesting (ESSH). This technique is based on the concept of synchronized switch harvesting (SSH), a nonlinear technique developed for energy harvesting from structural vibration. Compared with the standard technique of energy harvesting, the new technique dramatically increases the harvested power by almost 300% at resonance frequencies in the same vibration conditions, and also ensures an optimal harvested power whatever the load connected to the microgenerator. Furthermore, the new technique (ESSH) in this paper can be truly self-powered; a self-powered circuit which implements the technique is proposed. In addition, the overall power dissipation for the control circuitry is relatively constant (only about 121 µW), which is more attractive especially at high excitation. Because the new technique (ESSH) in this paper can be truly self-powered, no external power supply is needed, making the system suitable for more application fields, especially in remote operation.

Functional Source Separation from Magnetoencephalographic Signals

We propose a novel cerebral source extraction method (functional source separation, FSS) starting from extra‐cephalic magnetoencephalographic (MEG) signals in humans. It is obtained by adding a functional constraint to the cost function of a basic independent component analysis (ICA) model, defined according to the specific experiment under study, and removing the orthogonality constraint, (i.e., in a single‐unit approach, skipping decorrelation of each new component from the subspace generated by the components already found). Source activity was obtained all along processing of a simple separate sensory stimulation of thumb, little finger, and median nerve. Being the sources obtained one by one in each stage applying different criteria, the a posteriori “interesting sources selection” step is avoided. The obtained solutions were in agreement with the homuncular organization in all subjects, neurophysiologically reacting properly and with negligible residual activity. On this basis, the separated sources were interpreted as satisfactorily describing highly superimposed and interconnected neural networks devoted to cortical finger representation. The proposed procedure significantly improves the quality of the extraction with respect to a standard BSS algorithm. Moreover, it is very flexible in including different functional constraints, providing a promising tool to identify neuronal networks in very general cerebral processing. Hum Brain Mapp, 2006.

Somatosensory dynamic gamma-band synchrony: a neural code of sensorimotor dexterity.

To investigate neural coding characteristics in the human primary somatosensory cortex, two fingers with different levels of functional skill were studied. Their dexterity was scored by the Fingertip writing test. Each finger was separately provided by a passive simple sensory stimulation and the responsiveness of each finger cortical representation was studied by a novel source extraction method applied to magnetoencephalographic signals recorded in a 14 healthy right handed subject cohort. In the two hemispheres, neural oscillatory activity synchronization was analysed in the three characteristic alpha, beta and gamma frequency bands by two dynamic measures, one isolating the phase locking between neural network components, the other reflecting the total number of synchronous recruited neurons. In the dominant hemisphere, the gamma band phase locking was higher for the thumb than for the little finger and it correlated with the contra-lateral finger dexterity. Neither in the dominant nor in the non-dominant hemisphere, any effect was observed in the alpha and beta bands. In the gamma band, the amplitude showed similar tendency to the phase locking, without reaching statistical significance. These findings suggest the dynamic gamma band phase locking as a code for finger dexterity, in addition to the magnification of somatotopic central maps.

Robot vision with cellular neural networks: a practical implementation of new algorithms

Cellular neural networks (CNNs) are well suited for image processing due to the possibility of a parallel computation. In this paper, we present two algorithms for tracking and obstacle avoidance using CNNs. Furthermore, we show the implementation of an autonomous robot guided using only real-time visual feedback; the image processing is performed entirely by a CNN system embedded in a digital signal processor (DSP). We successfully tested the two algorithms on this robot. Copyright

Guiding a mobile robot with cellular neural networks

We show how cellular neural networks (CNNs) are capable of providing the necessary signal processing needed for visual navigation of an autonomous mobile robot. In this way, even complex feature detection and object recognition can be obtained in real time by analogue hardware, making fully autonomous real-time operation feasible. An autonomous robot was first simulated and then implemented by simulating the CNN with a DSP. The robot is capable of navigating in a maze following lines painted on the floor. Images are processed entirely by a CNN-based algorithm, and navigation is controlled by a fuzzy-rule-based algorithm. Copyright

Design and training of multilayer discrete time cellular neural networks for antipersonnel mine detection using genetic algorithms

In this work we present a novel strategy for the simultaneous design and training of multilayer discrete-time cellular neural networks. This methodology is applied to the detection of surface-laid antipersonnel mines in infrared imaging. The procedure is based on the application of genetic algorithms for both network design and learning task.

Optimizing ICA in fMRI using information on spatial regularities of the sources

Spatial independent component analysis (ICA) is a well-established technique for multivariate analysis of functional magnetic resonance imaging (fMRI) data. It blindly extracts spatiotemporal patterns of neural activity from functional measurements by seeking for sources that are maximally independent. Additional information on one or more sources (e.g., spatial regularity) is often available; however, it is not considered while looking for independent components. In the present work, we propose a new ICA algorithm based on the optimization of an objective function that accounts for both independence and other information on the sources or on the mixing model in a very general fashion. In particular, we apply this approach to fMRI data analysis and illustrate, by means of simulations, how inclusion of a spatial regularity term helps to recover the sources more effectively than with conventional ICA. The improvement is especially evident in high noise situations. Furthermore we employ the same approach on data sets from a complex mental imagery experiment, showing that consistency and physiological plausibility of relatively weak components are improved.

Morphological Analysis of ECG Holter Recordings by Support Vector Machines

A new method of automatic shape recognition of heartbeats from ECG Holter recordings is presented. The mathematical basis of this method is the theory of support vector machine, a new paradigm of learning machine. The method consists of the following steps: signal preprocessing by digital filters, segmentation of the Holter recording into a series of heartbeats by wavelet technique, support vector approximation of each heartbeat with the use of Gaussian kernels, support vector classification of heartbeats. The learning sets for classification are prepared by physician. Hence, we offer a learning machine as a computer-aided tool for medical diagnosis. This tool is flexible and may be tailored to the interest of physicians by setting up the learning samples. The results shown in the paper prove that our method can classify pathologies observed not only in the QRS alterations but also in P (or F), S and T waves of electrocardiograms. The advantages of our method are numerical efficiency and very high score of successful classification.

Real time vision by FPGA implemented CNNs

In order to get real time image processing for mobile robot vision, we propose to use a discrete time cellular neural network implementation by a convolutional structure on Altora FPGA using VHDL language. We obtain at least 9 times faster processing than other emulations for the same problem.

Fuzzy reasoning for the design of CNN-based image processing systems

Fuzzy reasoning in image processing has been proved to be a very effective way to formalize complex inference techniques based on heuristics or experience, taking perceptual quality criteria into account. In this paper, we discuss implementation of fuzzy reasoning image processing on the standard cellular neural network universal machine. In this way, it is possible to employ such powerful massively parallel chips to speed up use of known algorithms, and to systematize design of new perceptual-quality driven CNN applications.