Este trabalho tem a finalidade de apresentar e comparar técnicas de redução de ruído utilizando como critérios de avaliação a mínima distorção espectral e a redução de ruído, na reconstrução dos sinais de voz degradados por ruído. Para tanto, utilizou-se os filtros de Kalman de tempo discreto e de freqüência discreta em conjunto com a técnica de subtração espectral de potência. Os sinais utilizados foram contaminados por ruídos branco e colorido, e a avaliação do desempenho dos algoritmos foi realizada tendo-se como parâmetros a relação sinal/ruído segmentada (SNRseg) e a distância de Itakura-Saito (d(a,b)). Após o processamento, verificou-se que a técnica, proposta neste trabalho, de filtragem de Kalman no tempo em conjunto com a subtração espectral de potência, apresentou resultados um pouco melhores em relação à filtragem de Kalman na freqüência em conjunto com a subtração espectral de potência. / This work has as main objective to present and to compare techniques of noise reduction using as evaluation criterion the low spectral distortion and the noise reduction in the reconstruction of corrupted speech signals. For so much, it was used the Kalman\'s filters in the time and frequency domain together with the technique of power spectral subtraction. The used signals were corrupted by white and colored noises and the evaluation of effectiveness of the algorithms was accomplished using the segmental signal-to-noise ratio (SNRseg) and the Itakura-Saito distance (d(a,b)). After the processing, it was noticed that the Kalman filtering in the time together with power spectral subtraction presented better results than the Kalman filtering in the frequency together with power spectral subtraction.

Hearing aid devices are used to help people with hearing impairment. The number of people that requires hearingaid devices are possibly constant over the years, however the number of people that now have access to hearing aiddevices increasing rapidly. The hearing aid devices must be small, consume very little power, and be fairly accurate.Even though it is normally more important for the user that hearing impairment look good (are discrete). Once thehearing aid device prescribed to the user, she/he needs to train and adjust the device to compensate for the individualimpairment.We are within the framework of this project researching on hearing aid devices that can be trained by the hearingimpaired person her-/himself. This project is about finding suitable noise cancellation algorithm for the hearing-aiddevice. We consider several types of algorithms like, microphone array signal processing, Independent ComponentAnalysis (ICA) based on double microphone called Blind Source Separation (BSS) and DRNPE algorithm.We run this current and most sophisticated and robust algorithms in certain noise backgrounds like Cocktail noise,street, public places, train, babble situations to test the efficiency. The BSS algorithm was well in some situation andgave average results in some situations. Where one microphone gave steady results in all situations. The output isgood enough to listen targeted audio.The functionality and performance of the proposed algorithm is evaluated with different non-stationary noisebackgrounds. From the performance results it can be concluded that, by using the proposed algorithm we are able toreduce the noise to certain level. SNR, system delay, minimum error and audio perception are the vital parametersconsidered to evaluate the performance of algorithms. Based on these parameters an algorithm is suggested forheairng-aid. / Hearing-Aid

Στην παρούσα διπλωματική αναπτύχθηκε μια μέθοδος ψηφιακής επεξεργασίας σημάτων για ακουστικά σήματα συμβατή με πραγματικού χρόνου επεξεργασία. Αρχικά έγινε περίληψη των λειτουργιών των επεξεργαστών ειδικού σκοπου. Έγινε μελέτη της ανάλυσης στο πεδίο της συχνότητας καθώς και της συνάρτησης συνεκτικότητας. Για τους σκοπούς της διπλωματικής υλοποιήθηκε αλγόριθμος αφαίρεσης θορύβου από σήματα ομιλίας που αξιοποιεί την συνάρτηση συνεκτικότητας και χρησιμοποιεί είσοδο από δύο μικρόφωνα. Ο αλγόριθμος αυτός υλοποιήθηκε και δοκιμάστικε σε μη-πραγματικό χρόνο σε μαθηματικό λογισμικό , καθώς και σε πραγματικό χρόνο σε επεξεργαστή ειδικού σκοπού. / In this thesis, a method of digital signal processing for acoustic signals was developed, compatible with real-time processing. At first, a review of the operations that special purpose digital signal processors feature. We also studied the frequency domain analysis and the coherence function in depth. For the purposes of this thesis an algorithm of noise reduction from speech signals was implemented, that exploits the coherence function and takes two microphone signals as inputs. The algorithm was implemented offline in a mathematical software, as well as real time in a special purpose digital signal processor.

The use of personal hearing protection devices (HPDs) is often recommended to protect workers' hearing from noise-induced damage when no other means of reducing noise levels at the source is effective. The effects of HPDs on speech communication cannot be neglected in spite of their benefit in reducing the risk of hearing loss. While much research has been directed at speech perception, much less is known on how HPDs affect speech production. The tendency of talkers to raise their vocal effort in noise, known as the Lombard effect, is often disrupted by HPDs due to their occlusion effect and the lower noise at the ears as well as the attenuated feedback from one’s own voice. Three main knowledge gaps are addressed in this thesis. The first gap is to characterize speech produced by talkers with or without HPDs under realistic acoustic conditions while immersed in an external noise field. The second gap is to evaluate more comprehensively speech production under protected and unprotected talker and listener ear conditions in different types of fluctuating and continous noises. The third gap is to assess the alterations in the characteristics of speech produced by talkers wearing level-dependent HPDs set at different transmission gain settings and in comparison with passive HPDs. This thesis extends methods used to recover Lombard speech elicited in an external noise field. For this purpose, two noise suppression methods, direct waveform subtraction (DWS) and adaptive noise cancellation (ANC), were found to adequately remove noise from speech recorded for SNRs as low as −10 dB. Moreover, this work contributes new knowledge on the effects of conventional passive HPDs on speech production. When talker wears HPD in noise then speech level were found to decrease by up to 9 dB in continuous noises and by 7 dB in fluctuating noises compared to open ears, while speech levels were found to increase by about 5 dB in all noises when the listener wears HPD. Furthermore, changes in pitch and spectral levels were consistent with changes in speech levels. The effects of level-dependent HPD on speech production, depending on the chosen transmission gain setting, revealed that it led to smaller decrease in talkers’ speech levels in noise compared to conventional passive HPD. These findings indicate that the level-dependent HPDs may impede communication less than conventional passive HPDs, while providing protection against high levels of noise.

This research is mainly concerned with a robust method for improving the performance of a real-time speech enhancement and noise cancellation for Automatic Speech Recognition (ASR) in a real-time environment. Therefore, the thesis titled, “Real-time adaptive beamformer for Automatic speech Recognition in a car environment” presents an application technique of a beamforming method and Automatic Speech Recognition (ASR) method. In this thesis, a novel solution is presented to the question as below, namely: How can the driver’s voice control the car using ASR? The solution in this thesis is an ASR using a hybrid system with acoustic beamforming Voice Activity Detector (VAD) and an Adaptive Wiener Filter. The beamforming approach is based on a fundamental theory of normalized least-mean squares (NLMS) to improve Signal to Noise Ratio (SNR). The microphone has been implemented with a Voice Activity Detector (VAD) which uses time-delay estimation together with magnitude-squared coherence (MSC). An experiment clearly shows the ability of the composite system to reduce noise outside of a defined active zone. In real-time environments a speech recognition system in a car has to receive the driver’s voice only whilst suppressing background noise e.g. voice from radio. Therefore, this research presents a hybrid real-time adaptive filter which operates within a geometrical zone defined around the head of the desired speaker. Any sound outside of this zone is considered to be noise and suppressed. As this defined geometrical zone is small, it is assumed that only driver's speech is incoming from this zone. The technique uses three microphones to define a geometric based voice-activity detector (VAD) to cancel the unwanted speech coming from outside of the zone. In the case of a sole unwanted speech incoming from outside of a desired zone, this speech is muted at the output of the hybrid noise canceller. In case of an unwanted speech and a desired speech are incoming at the same time, the proposed VAD fails to identify the unwanted speech or desired speech. In such a situation an adaptive Wiener filter is switched on for noise reduction, where the SNR is improved by as much as 28dB. In order to identify the signal quality of the filtered signal from Wiener filter, a template matching speech recognition system that uses a Wiener filter is designed for testing. In this thesis, a commercial speech recognition system is also applied to test the proposed beamforming based noise cancellation and the adaptive Wiener filter.

In this thesis, we are interested in processing noisy speech signals that are meant to be heard by humans, and hence we approach the noise-suppression problem from a perceptual perspective. We develop a noise-suppression paradigm that is based on a model of the human auditory system, where we process signals in a way that is natural to the human ear. Under this paradigm, we transform an audio signal in to a perceptual domain, and processes the signal in this perceptual domain. This approach allows us to reduce the background noise and the audible artifacts that are seen in traditional noise-suppression algorithms, while preserving the quality of the processed speech. We develop a single- and dual-microphone algorithm based on this perceptual paradigm, and conduct subjecting tests to show that this approach outperforms traditional noise-suppression techniques. Moreover, we investigate the cause of audible artifacts that are generated as a result of suppressing the noise in noisy signals, and introduce constraints on the noise-suppression gain such that these artifacts are reduced.

Disertacijoje nagrinėjamas triukšmo šaltinių keliamo triukšmo sklidimas į aplinką bei ieškoma mažinimo priemonių jam slopinti. Disertacijos tikslas – įver-tinti stacionarių ir mobilių šaltinių keliamo triukšmo sklaidą aplinkoje ir numatyti efektyvias priemones jai mažinti, naudojant triukšmo slopinimo kamerą, o prie-monių efektyvumą įvertinti modeliuojant. Šiame darbe sprendžiami keli pagrindiniai uždaviniai: sukurti ir įrengti triukšmo slopinimo kamerą; nustatyti triukšmo sklaidą, esant skirtingiems mobi-liems ir stacionariems triukšmo šaltiniams; numatyti bei modeliavimo būdu įver-tinti priemones triukšmui slopinti. Disertaciją sudaro įvadas, penki skyriai, rezultatų apibendrinimas, naudotos literatūros ir autoriaus publikacijų disertacijos tema sąrašai. Įvadiniame skyriuje aptariama tiriamoji problema, darbo aktualumas, apra-šomas tyrimų objektas, formuluojamas darbo tikslas bei uždaviniai, aprašoma tyrimų metodika, darbo mokslinis naujumas, darbo rezultatų praktinė reikšmė, ginamieji teiginiai. Įvado pabaigoje pristatomos disertacijos tema autoriaus pa-skelbtos publikacijos ir pranešimai konferencijose bei disertacijos struktūra. Pirmasis skyrius skirtas literatūros apžvalgai. Jame išnagrinėti skirtingi triukšmo parametrai, jo šaltiniai, triukšmo mažinimo būdai, poveikis aplinkai, triukšmo sklaidos modeliavimo programos bei medžiagų akustinių savybių tyri-mai. Skyriaus pabaigoje formuluojamos išvados. Antrajame skyriuje pateiktos eksperimentinių tyrimų, skirtingų šaltinių... [toliau žr. visą tekstą] / The dissertation analyses the problem of environment noise pollution and mitigation. The dissertation is aimed to evaluate the dispersion of noise from sta-tionary and mobile sources, provide for effective techniques and measures to re-duce it by using a noise suppression chamber for research into acoustic properties and to estimate the efficiency of the noise reduction measures through modelling. The work deals with main tasks: to design and fit out a noise suppression chamber and use it for research into the noise absorption and insulation proper-ties of different materials; to determine the levels and dispersion of noise from different mobile and stationary sources of noise; to determine the longitudinal sound wave attenuation coefficient for individual building materials and for buil-ding structures - the airborne sound attenuation index; to estimate the efficiency of the measures reducing noise from diferent sources with software “CadnaA”. The scientific work consist of the general characteristic of the dissertation, 5 chapters, conclusions and recommendations, list of literature, list of publications. The introduction discusses the problem addressed, topicality of the work, describes the object of research, and formulates the aim and tasks of the work, scientific novelty of the work, practical value of the research results and defen-ded propositions. It presents the author’s publications and conference reports on the topic of the dissertation and the structure of the... [to full text]

Power line communication has been around in past decades and gained renewed attention thanks to the demand of high-speed Internet access. With the significant advantages of existing infrastructure and accessibility to even remote areas, power grid has become one of the promising competitors for multi-media transmission in household. However, the power line was not oriented for data transmission providing a rather hash environment. To overcome the difficulties, advanced modulation and channel coding schemes should be employed. In the thesis low density parity check code (LDPC) is employed to reduce the loss caused by various kinds of effects in the channel especially the noise since its performance approaches to Shannon capacity limit. Moreover, OFDM multi-carrier transmission technique is involved which could decrease the inter-symbol interference and frequency selective fading. Nevertheless, LDPC decoding process was designed specifically for the common Gaussian white noise condition, combined with OFDM modulation the system still could not provide satisfying and practicable performance so improvements are needed for the system. The main works of the thesis are as follows. Set up an environment of power line transmission investigating and simulating the channel characteristics; employ multi-path channel model and Class‐A noise model for further developing the improvement algorithms to deal with the selective fading and impulse noise. Two algorithms proposed here are from different perspectives: the first one is modifying initial posterior information for LDPC decoding and the second one aims at suppressing the impulse noise after demodulation. Finally, a few simulations are performed to reveal the effectiveness of proposed methods. As a result, the improved scheme shows a great superiority improving the performance by no less than 5dB compared to traditional system.

No description available.

The demand for low power wireless transceiver implementations has been fueled by multiple applications in the recent decades, including cellular systems, wireless local area networks, personal area networks, biotelemetry and sensor networks. Dynamic range, which is set by linearity and sensitivity performance, is a critical design metric in many of these systems. Both linearity and sensitivity requirements continue to become progressively challenging in many systems due to greater spectrum usage and the need for high data rates respectively. The objective of this research is to investigate power-efficient circuit techniques for reducing the power requirement in receiver front-ends without compromising the dynamic range performance. In the first part of the dissertation, a low power receiver down-converter topology for enhancing dynamic range performance is presented. Current mode down-converters with passive mixer cores have been shown to provide excellent dynamic range performance. However, in contrast to a current commutating Gilbert cell, these down-converters require separate bias current paths for the RF transconductor and the baseband transimpedance amplifier. The proposed topology reduces the power requirement of conventional current mode passive down-converter by sharing the bias current between the transconductance and transimpedance stages. This is achieved without compromising the available voltage headroom for either stage, which is a limitation of bias-sharing based on the use of stacked stages. The dynamic range of the basic bias-current-shared topology is further enhanced through suppression of low frequency noise and IM3 products. Two variants of the down-converter, employing a broadband common-gate and a narrowband common-source input stage, are implemented in a 0.18-μm CMOS technology. The dynamic range performance of the architecture is analyzed. Finally, a prototype of a full direct-conversion receiver implementation with quadrature outputs and integrated LO synthesis is demonstrated. A power-efficient oscillator design for phase noise minimization is presented in the second part of this dissertation. This design is targeted towards multi-radio platforms where several communication links operate simultaneously over multiple frequency bands. Blockers from concurrently operating radios present a major design challenge. The blockers not only make the frontend linearity requirement more stringent but also degrade receiver sensitivity through reciprocal mixing with the phase noise sidebands of LO. Phase noise minimization is thus critical for ensuring high sensitivity in frequency bands where large blockers are present and not sufficiently attenuated by pre-select filters. A capacitive power combining technique in oscillators is introduced to improve phase noise performance. By combining this approach with current reuse, the phase noise is reduced at lower power, compared to conventional LC oscillators. This leads to improved power efficiency. Moreover, the technique mitigates modeling uncertainty arising from phase noise reduction through simultaneous impedance and current scaling. The mode selection in this oscillator, which employs multiple coupled resonators, is analyzed and the impact of coupling on far-out phase noise performance is discussed. Multi-mode oscillation can potentially arise in other oscillator topologies too, e.g., in multiphase oscillators. Mode selection in a widely used transistor-coupled quadrature oscillator is analyzed in detail in the final part of the dissertation. The analysis shows how cross-compression among multiple competing modes can lead to suppression of non-dominant modes in the steady state. / text