Skip to Main content Skip to Navigation

Amélioration de la sélectivité de vitesse pour l'acquisition de signaux ENG par l'analyse spectrale spatio-temporelle

Abstract : Improving the lives of people with disabilities is a major challenge in our society. In the case of a handicap due to deficiencies of the sensory-motor system, surgical or drug solutions can rarely restore the altered functions. More technological solutions may in some cases be used to bypass problematic malfunction of the nervous system. One solution is to control certain functions by functional electrical stimulation. For this to be effective stimulation and approaches the natural control an interesting approach is to use natural touch sensors in the system control loop. As part of this thesis, we focus on electrical systems acquisition neurological signals from the peripheral nervous system. The idea is that the information from the afferent nerve fibers obtained using extra-neural electrodes (eg. Cuff) could help to "capture" the sensory information from several sensors natural without being too invasive. The main difficulty of this technique and to discriminate the activity of a fiber or group of fibers from the composite signal received by the electrode. This signal includes, in fact, information from a large number of afferent and efferent very different origin. One possible approach is to discriminate the information of different fibers from their physiological and anatomical features. The work presented in this manuscript are looking to improve the selectivity of the collection based on the propagation speed of the different fibers. Note that no truly viable solution exists in the literature on this principle of selectivity. The first objective was therefore to prove that in an ideal case (ideal filter), it would be possible to make effective discrimination. Our work is based on studies of temporal, spatial and spatio-temporal order to assess the possibilities of extracting useful information from the simulated extracellular potentials. The simulations were performed from a realistic acquisition system model developed from the parameters of the model and biophysiological electrode model and associated preamplifier. Based on the promising results obtained in a case of an ideal filter, IIR filter implementations and spatiotemporal FIR of first and second order have been proposed. The proposed filter stability parameters have been studied and possible to envisage an effective physical implementation.
Document type :
Complete list of metadatas

Cited literature [191 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Friday, March 15, 2019 - 10:59:08 AM
Last modification on : Wednesday, December 11, 2019 - 3:39:57 AM


Version validated by the jury (STAR)


  • HAL Id : tel-02068694, version 1



Mariam Abdallah. Amélioration de la sélectivité de vitesse pour l'acquisition de signaux ENG par l'analyse spectrale spatio-temporelle. Autre. Université Montpellier, 2015. Français. ⟨NNT : 2015MONTS239⟩. ⟨tel-02068694⟩



Record views


Files downloads