Hearing loss of various types is increasingly plaguing our modern world (Geneva: World Health Organization 2018). As the life expectancy increased in the industrialized world, age-related hearing loss (ARHL) has become more prevalent. The wars and terrorism of the modern world also created a significant body of blast-induced hearing loss (BIHL) patients. Both types of hearing loss present significant challenges for listeners even at suprathreshold sound levels. However, increasing bodies of clinical and laboratory evidence have suggested that the difficulties in the processing of time-varying auditory features in speech and other natural sounds may not be sufficiently diagnosed by threshold changes and simple auditory electrophysiological measures (Snell and Frisina 2000; Saunders et al. 2015; Bressler et al. 2017; Guest et al. 2018).<br>Studies have emphasized that excitatory/inhibitory neurotransmission imbalance plays important roles in ARHL (Caspary et al. 2008) and may also be key in BIHL, as hinted by the strong presence of GABA regulation in non-blast TBI (O’Dell et al. 2000; Cantu et al. 2015; Guerriero et al. 2015). The current studies focus on age-related and blast-induced hearing deficits by examining changes in the processing of simple, brief stimuli and complex, sustained, temporally modulated sounds.<br>Through post hoc circular analysis of single-unit, in vivo recording of young and aged inferior colliculus (IC) neurons responding to amplitude modulation (AM) stimuli and modulation depth changes, we observed evidence of central compensation in the IC manifesting as increased sensitivity to presynaptic input, which was measured via local field potentials (LFPs). We also found decreased sensitivity to decreasing modulation depth. Age-related central gain in the IC single units, while preserving and even overcompensating for temporal phase coding in the form of vector strength, was unable to make up for the loss of envelope shape coding.<br>Through careful, longitudinal measurements of auditory evoked potential (AEP) responses towards simple sounds, AM and speech-like iterated rippled noise (IRN), we documented the development and recovery of BIHL induced by a single mild blast in a previously established (Song et al. 2015; Walls et al. 2016; Race et al. 2017) rat blast model over the course of two months. We identified crucial acute (day 1-4 post-exposure) and early recovery (day 7-14) time windows in which drastic changes in electrophysiology take place. Challenging conditions and broadband, speech-like stimuli can better elucidate mild bTBI-induced auditory deficits during the sub-acute period. The anatomical significance of the aforementioned time windows was demonstrated with immunohistochemistry methods, showing two distinct waves of GABA inhibitory transmission changes taking place in the auditory brainstem, the IC, and the auditory thalamus. These changes were in addition to axonal and oxidative damage evident in the acute phase. We examined the roles and patterns of excitatory/inhibitory imbalance in BIHL, its distinction compared to that of ARHL, and demonstrated the complexity of its electrophysiological consequences. Blast traumatizes the peripheral auditory system and auditory brainstem, evident through membrane damage and acrolein-mediated oxidative stress. These initial traumas kickstart a unique, interlocking cascade of excitatory/inhibitory imbalances along the auditory neuraxis that is more complex and individually varied than the gradual, non-traumatic degradations in ARHL. Systemic treatment with the FDA-approved acrolein scavenger Hydralazine (HZ) was attempted with limited effects.<br>Taken together, the current study provided insights into the similarities and distinctions between the mechanisms of ARHL and BIHL and called for innovative and individual diagnostic and therapeutic measures.<br>

Research in vocal vibrato has established that vocal tract filtering is primarily responsible for the amplitude modulation (AM) present in Western classical vibrato. Using electroglottography (EGG) and the EGG speed quotient, which is sensitive to fluctuations in the amplitude of vocal fold vibration, AM was detected at the laryngeal (source) level, in addition to the subsequent AM which results from vocal tract filtering. Seventeen classically-trained opera singers sang vowels in three pitch and loudness conditions. EGG and microphone measurements of FM and AM and their rates, extents, and periodicity were made. Airflow was also measured, and the samples were rated by voice professors for vibrato consistency, speed, and width. Physiologic and acoustic data revealed that AM from vocal tract filtering, or the resonance-harmonics interaction (RHI) described by Horii and associates, was present throughout the vibrato samples. Laryngeal-level AM was also present throughout, with soft conditions having the highest mean extents. Singers with lower degrees of laryngeal-level AM were also those rated highest for vibrato consistency. Vibrato rate increased as pitch increased, and, to a lesser extent, as intensity increased. These findings document, in addition to the AM resulting from the RHI, the concurrent presence of laryngeal-level AM in a group of singers representing a range of training and experience.

Amplitude and phase modulated optical OFDM (Orthogonal Frequency Division Multiplexing) are analyzed in a 50GBit/s single channel and 40GBit/s 5 channel 512 subcarrier non-ideal dispersion-compensated fiber optic communication systems. PM-OFDM is investigated as an alternative to AM-OFDM to alleviate the problem associated with amplitude-modulated signals in a nonlinear medium. The inherent dispersion compensation capability in OFDM (using a cyclic prefix) allows transmission over a link whose dispersion map is not exactly known. OFDM also mitigates the effects of dispersion slope in wavelength-division multiplexed (WDM) systems. Moreover, the overall dispersion throughout the transmission link may vary due to environmental effects and aging. OFDM is inherently tolerant to over- or under-compensation and dispersion slope mismatch. OFDM transmission over dispersive, non-dispersion managed fiber links using OFDM requires an overhead in excess of the maximum accumulated dispersion. Existing WDM systems usually employ periodic dispersion management. OFDM in these systems requires a smaller overhead. It is, however, more susceptible to nonlinearity due to the coherent beating of subcarriers after each dispersion-compensated span. The large variation in intensity associated with amplitude-modulated OFDM makes this modulation format more susceptible to nonlinear effects in fiber compared to phase-modulated signals. This holds true unless dispersion and EDFA noise lead to amplitude variations strong enough for PM-OFDM to be degraded by nonlinear effects as well. In conclusion OFDM is beneficial for non-ideal dispersion managed systems. PM-OFDM can further improve the performance.

No description available.

No description available.

No description available.

No description available.

[ES] Los sistemas de transmisión óptica no-coherente se emplean actualmente en las redes ópticas de corto alcance (< 80 km), como son las redes de ámbito metropolitano. La implementación más común en el estado del arte se basa en sistemas que emplean multiplexación por división en longitud de onda (WDM, wavelength division multiplexing) de cuatro longitudes de onda (¿) proporcionando un régimen binario de 100 Gbps (4¿×25 Gbps). En los últimos años, los sistemas de transmisión ópticos no-coherentes están evolucionando desde 100 Gbps a 400 Gbps (4¿×100 Gbps). Dado que este mercado comprende un gran número de sistemas, el coste es un parámetro importante que debe ser lo más bajo posible. El objetivo de esta tesis es investigar distintos aspectos del procesado de señal en general y, específicamente, investigar nuevas técnicas de procesado digital de señal (DSP, digital signal processing) que puedan ser utilizadas en sistemas de transmisión óptica no-coherentes empleando la modulación por amplitud de pulsos (PAM, pulse-amplitude modulation). Para que una técnica DSP sea interesante en el contexto de una red óptica WDM no-coherente, esta debe mitigar de manera efectiva al menos una de las tres limitaciones principales que afectan a estos sistemas: limitaciones de ancho de banda, limitaciones por dispersión cromática (CD), y el ruido. En esta tesis se proponen y examinan una serie de algoritmos cuyo su rendimiento es analizado mediante simulación y experimentalmente en laboratorio: - Feed-forward equalizer (FFE): este es el esquema de ecualización más común que se emplea principalmente en las transmisiones ópticas no-coherentes de alto régimen binario. Puede compensar grandes limitaciones en el ancho de banda. - Estimación de la secuencia de máxima verosimilitud (MLSE): el MLSE es un detector óptimo y, por lo tanto, proporciona las mejores prestaciones en detección cuando se abordan las limitaciones por CD y de ancho de banda. - Conformación geométrica de la constelación: en los esquemas de modulación de intensidad óptica multinivel, la distancia entre los niveles de amplitud puede ajustarse adecuadamente (de manera que no son equidistantes) a fin de aumentar la tolerancia de la señal frente al ruido. - Conformación probabilística: técnica diseñada específicamente para esquemas de modulación multinivel. Esta técnica ajusta la probabilidad de cada nivel de amplitud de modo que se incrementa la tolerancia al ruido óptico. - Señalización de respuesta parcial (PRS, partial signaling response): este es un enfoque basado en DSP donde una interferencia entre símbolos (ISI, inter-symbol interference) controlada es introducida intencionalmente de tal manera que la señal resultante requiere menos ancho de banda. La técnica PRS puede adaptarse para combatir también el efecto de CD. - Pre-énfasis digital (DPE, digital pre-emphasis): esta técnica consiste en aplicar el inverso de la función de transferencia del sistema a la señal en el transmisor, lo que reduce el impacto de las limitaciones de ancho de banda en el receptor. - Modulación con codificación Trellis (TCM, Trellis-coded modulation): esquema de modulación que combina elementos de corrección de errores (FEC, forward error correction) con técnicas de partición en conjuntos y modulación multidimensional para generar una señal más resistente al ruido. - Modulación multidimensional por partición en conjuntos: muy similar a TCM, pero sin ningún elemento FEC. Tiene menos ganancias que TCM en términos de tolerancia al ruido, pero no es tan sensible al ISI. Utilizando estas técnicas, esta tesis demuestra que es posible lograr una transmisión óptica con régimen binario de 100 Gbps/¿ empleando componentes de bajo coste. En esta tesis también demuestra regímenes binarios de más de 200 Gbps, lo que indica que la transmisión óptica no-coherente con modulación PAM puede ser una solución viable y eficiente en coste / [CA] Actualment, s'utilitzen sistemes òptics no coherents en xarxes òptiques de curt abast ( < 80 km), com són les xarxes d'àmbit metropolità. La implementació més comuna que podem trobar en l'estat de l'art es correspon amb sistemes emplenant multiplexació per divisió en longitud d'ona (WDM, wavelength division multiplexing) de quatre longituds d'ona (¿) proporcionant un règim binari de 100 Gbps (4¿×25 Gbps). En els últims anys, els sistemes de transmissió òptica no-coherents han evolucionat des de 100 Gbps cap a 400 Gbps (100 Gbps/¿). Atès que el mercat de sistemes de curt abast compren un gran volum de dispositius òptics instal·lats, el cost unitari és molt important i ha de ser el més baix possible. L'objectiu d'aquesta tesi és analitzar aspectes del processament de senyal en general i, específicament, investigar noves tècniques de processament digital de senyal (DSP, digital signal processing) que puguen ser utilitzades en sistemes de transmissió òptica no-coherent que utilitzen la modulació per amplitud d'impulsos (PAM, pulse-amplitude modulation). Per tal que una tècnica DSP es considere interessant per a una xarxa òptica WDM no-coherent, aquesta ha de mitigar efectivament almenys una de les tres principals limitacions que afecten aquests sistemes: limitacions d'ample de banda, limitacions per dispersió cromàtica (CD), i el soroll. En aquesta tesi s'examinen una sèrie d'algoritmes, el seu rendiment s'analitza per simulació i experimentalment en laboratori: - Feed-forward equalizer (FFE): aquest és l'esquema d'equalització més comú i s'utilitza bàsicament en les transmissions òptiques no coherents d'alt règim binari. Pot compensar grans quantitats de limitacions d'ample de banda. - Estimació de la seqüència de probabilitat màxima (MLSE): el MLSE és un detector òptim i, per tant, proporciona el millor rendiment quan es tracta de limitacions d'ample de banda i de CD. - Conformació geomètrica de la constel·lació: en esquemes de modulació òptica d'intensitat multinivell es pot ajustar la distància entre els nivells d'amplitud (de manera que ja no són equidistants) per augmentar la tolerància del senyal al soroll. - Conformació probabilística: una tècnica dissenyada específicament per als esquemes de modulació multinivell; ajusta la probabilitat de cada nivell d'amplitud de manera que augmenta la tolerància al soroll òptic. - Senyalització de resposta parcial (PRS, partial signaling response): és un enfocament basat en DSP on la interferència entre símbols (ISI, inter-symbol interference) controlada s'introdueix intencionalment de manera que el senyal resultant requereix menys ample de banda. La tècnica PRS es pot adaptar per combatre els efectes del CD. - Pre-èmfasi digital (DPE, digital pre-emphasis): aquesta tècnica consisteix a aplicar la inversió de la funció de transferència del sistema a la senyal en el transmissor de manera que es redueix l'impacte de les limitacions d'ample de banda en la senyal en el receptor. - Modulació amb codificació Trellis (TCM, Trellis-coded modulation): esquema de modulació que combina els elements de correcció d'errors avançats (FEC, forward error correction) amb tècniques de partionament de conjunts i modulació multidimensional per generar un senyal més resistent al soroll. - Modulació multidimensional per partició en conjuntes: molt similar a TCM però sense elements FEC. Té guanys menors que TCM en termes de tolerància al soroll, però no és tan sensible a l'ISI. Mitjançant l'ús d'aquestes tècniques, aquesta tesi demostra que és possible aconseguir una transmissió òptica amb un règim binari de 100 Gbps/¿ utilitzant components de baix cost. Esta tesi també demostra règims binaris de més de 200 Gbps, el que indica que la tecnologia no-coherent amb modulació PAM és una solució viable i eficient en cost per a una nova generació de sistemes transceptors òptics WDM funcionant a 800 Gbps (4¿×200 G / [EN] Non-coherent optical transmission systems are currently employed in short-reach optical networks (reach shorter than 80 km), like metro networks. The most common implementation in the state-of-the-art is the four wavelength (¿) 100 Gbps (4¿×25 Gbps) wavelength division multiplexing (WDM) transceiver. In recent years non-coherent optical transmissions are evolving from 100 Gbps to 400 Gbps (4¿×100 Gbps). Since in the short-reach market the volume of optical devices being deployed is very large, the cost-per-unit of the devices is very important, and it should be as low as possible. The goal of this thesis is to investigate some general signal processing aspects and, specifically, digital signal processing (DSP) techniques required in non-coherent pulse-amplitude modulation (PAM) optical transmission, and also to investigate novel algorithms which could be applied to this application scenario. In order for a DSP technique to be considered an interesting solution for non-coherent WDM optical networks it has to effectively mitigate at least one of the three main impairments affecting such systems: bandwidth limitations, chromatic dispersion (CD) and noise (in optical or electrical domain). A series of algorithms are proposed and examined in this thesis, and their performance is analyzed by simulation and also experimentally in the laboratory: - Feed-forward equalization (FFE): this is the most common equalizer and it is basically employed in every high-speed non-coherent optical transmission. It can compensate high bandwidth limitations. - Maximum likelihood sequence estimation (MLSE): the MLSE is the optimum detector and thus provides the best performance when it comes to dealing with CD and bandwidth limitations. - Geometrical constellation shaping: in multilevel optical intensity modulation schemes the distance between amplitude levels can be adjusted (such that they are no longer equidistant) in order to increase the signal's tolerance to noise. - Probabilistic shaping: another technique designed specifically for multilevel modulation schemes; it adjusts the probability of each amplitude level such that the tolerance to optical noise is increased. - Partial response signaling (PRS): this is a DSP-based approach where a controlled inter-symbol interference (ISI) is intentionally introduced in such a way that the resulting signal requires less bandwidth. PRS can be customized to also mitigate CD impairment, effectively increasing transmission distances up to three times. - Digital pre-emphasis (DPE): this technique consists in applying the inverse of the transfer function of the system to the signal at the transmitter side which reduces the impact of bandwidth limitations on the signal at the receiver side. - Trellis-coded modulation (TCM): a modulation scheme that combines forward error correction (FEC) elements with set-partitioning techniques and multidimensional modulation to generate a signal that is more resistant to noise. - Multidimensional set-partitioned modulation: very similar with TCM but without any FEC elements. It has lower gains than TCM in terms of noise tolerance but is not so sensitive to ISI. By using the techniques enumerated above, this thesis demonstrates that is possible to achieve 100 Gbps/¿ optical transmission bitrate employing cost-effective components. Even more, bitrates higher than 200 Gbps are also demonstrated, indicating that non-coherent PAM is a viable cost-effective solution for next-generation 800 Gbps (4¿×200 Gbps) WDM transceivers. / Prodaniuc, C. (2019). Advanced Signal Processing for Pulse-Amplitude Modulation Optical Transmission Systems [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/117315

Várias técnicas de cifragem óptica de imagens surgem a todo o momento devido à necessidade que se tem de transmitir informações de uma maneira segura e confiável. Nessas técnicas a informação é criptografada de tal modo que para conseguir recuperar a informação criptografada é necessário ter um conhecimento prévio da mesma chave que a criptografou, sendo esta chave única. Isto faz com que as técnicas sejam robustas. Estes processos de cifragem de imagem trabalham com a distribuição de fase obtida diretamente da imagem que se deseja criptografar e conseqüentemente a imagem decodificada também está em modulação de fase. Porém imagens em distribuição de fase não são visíveis ao olho humano. Para transformá-Ias em modulação de amplitude, a qual é visível ao olho humano, é utilizado o método de contraste de fase de Zernike. Propôs-se então um método para cifrar imagem binária apenas em modulação de amplitude. Após o processo de decodificação, a imagem obtida é reconhecida graças à capacidade do olho humano em discernir regiões contínuas de regiões aleatórias. Este método é de fácil implementação, podendo ser obtido utilizando apenas materiais de baixo custo como papéis, filmes plásticos, etc. Vários ataques foram realizados contra o processo de cifragem óptica de imagens utilizando máscara binária aleatória para analisar a robustez de tal processo. Este processo de cifragem não é robusto ao ataque do texto conhecido e ao ataque das cifras conhecidas. / Several optical encryption techniques have been developed due to the need to transmit information in a secure and reliable way. In these techniques the information is encrypted in a way that to recover the encrypted information is necessary to have a previous knowledge of the same mask that encrypted it, being this mask unique. Because of this, the techniques are robust. These image encryption methods work with the phase distribution obtained directly from the image that is intended to encrypt and consequently the decrypted image will also be in the phase modulation. But, images in a phase distribution are not visible for the human eyes. To transform them in an amplitude modulation, that is visible for the human eyes, it is used the Zernike phase contrast method. Then, it is proposed a method for encryption and decryption of information processed in a binary form in an amplitude modulation. After the decryption process, the image obtained is recognized because the human eyes are able to differentiate continuous, periodical regions to random pattern. This method has an easy implementation and it can be obtained using regular and inexpensive materials, like paper, plastic films, etc. Many attacks were realized against the proposed method to analyse if this method is robust. This method can not resist the known-plaintext attack and the known-ciphertext attack.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / Neste trabalho, apresentamos uma anÃlise numÃrica para a obtenÃÃo de portas lÃgicas totalmente Ãptica baseada em um acoplador direcional nÃo-linear simÃtrico (NLDC) em fibras de cristal fotÃnico (PCF). O projeto mais comumente usado para o NLDC-PCF à uma fibra holey, utilizada neste trabalho, em que a seÃÃo transversal à uma matriz periÃdica de buracos de ar que se prolonga por todo o comprimento da fibra, operando com dois pulsos de luz ultracurtos na forma de sÃlitons, de duraÃÃo mÃnima de 100fs (femtosegundos). Investigamos dois formatos de modulaÃÃo por pulsos, a modulaÃÃo por amplitude de pulso (PAM) na modalidade de chaveamento de mudanÃa de amplitude (ASK) e a modulaÃÃo por largura de pulso (PWM) para obter portas lÃgicas no NLDC-PCF duplo simÃtrico. Avaliamos o efeito resultante de um incremento no parÃmetro codificaÃÃo PAM (&#949;) e PWM (w), considerando a dispersÃo de segunda ordem (&#946;2), a dispersÃo de terceira ordem (&#946;3) e efeitos nÃo-lineares, tais como: SPM (Self Phase Modulation), SS (Self-Steepening) e IRS (lntrapulse Raman Scattering) em uma configuraÃÃo sem perdas. Os nossos resultados indicam que à possÃvel obter operaÃÃes lÃgicas utilizando um controle de fase entre os pulsos de entrada. / We present a numerical analysis for obtaining all-optical logic gates based on a nonlinear directional coupler symmetric (NLDC) based on photonic crystal fibers (PCF). The most commonly used to project the NLDC-PCF is a holey fiber, used here in cross section which is a periodic array of air holes extending through the length of the fiber, using two ultrashort light pulses in form of solitons, the minimum duration of 100fs (femtoseconds). We investigated two forms of modulation pulse, pulse amplitude modulation (PAM) in the form of amplitude shift keying (ASK) modulation and pulse width modulation (PWM) for logic gate NLDC-PCF symmetrical double. We evaluated the effect resulting from an increase in the offset parameter encoding PAM (&#949;) and PWM (w), considering the second order dispersion (&#946;2), the third order dispersion (&#946;3) and non-linear modulation effects SPM (Self Phase Modulation), SS (Self-Steepening) and IRS (lntrapulse Raman Scattering) in a configuration without loss. Our results indicate that logical operations can be obtained using a phase control between the input pulses.