VEU - Grup de Tractament de la Parla
http://hdl.handle.net/2117/3746
2017-02-28T05:52:07ZAnalytic performance evaluation of cumulant-based arma system identification methods
http://hdl.handle.net/2117/101469
Analytic performance evaluation of cumulant-based arma system identification methods
Rodríguez Fonollosa, José Adrián; Vidal Manzano, José
The authors perform an analytic study of some cumulant-based methods for estimating the AR parameters of ARMA processes. The analysis includes new AR identifiability results for pure AR process and the analytic performance evaluation of system identification methods based on cumulants. The authors present examples of pure AR processes that are not identifiable via the normal equations based on the diagonal third-order cumulant slice. The results of the performance evaluation are illustrated graphically with plots of the variance of the estimates as a function of the parameters of the process.
2017-02-23T13:28:32ZRodríguez Fonollosa, José AdriánVidal Manzano, JoséThe authors perform an analytic study of some cumulant-based methods for estimating the AR parameters of ARMA processes. The analysis includes new AR identifiability results for pure AR process and the analytic performance evaluation of system identification methods based on cumulants. The authors present examples of pure AR processes that are not identifiable via the normal equations based on the diagonal third-order cumulant slice. The results of the performance evaluation are illustrated graphically with plots of the variance of the estimates as a function of the parameters of the process.Blind multiuser detection with probabilistic algorithms - application to underwater communications
http://hdl.handle.net/2117/101382
Blind multiuser detection with probabilistic algorithms - application to underwater communications
Antón Haro, Carles; Rodríguez Fonollosa, José Adrián; Zvonar, Zaran; Rodríguez Fonollosa, Javier
In this paper, two probabilistic adaptive algorithms for jointly detecting active users in a DS-CDMA system are compared. The first one, which is based on the theory of hidden Markov models (HMM) is proposed within the CDMA scenario and compared with the previously developed Viterbi-based algorithm. Both techniques are completely blind in the sense that no knowledge of the signature sequences, channel state information or training sequences is required for any user. After convergence, an estimate of the signature of each user convolved with its physical channel impulse response (CIR), and estimated data sequences are provided. This CIR estimate can then be used to switch to any decision-directed (DD) adaptation scheme. Performance of the algorithms is verified with simulations as well as with experimental data from an underwater acoustics (UWA) environment. In both cases, performance is found to be highly satisfactory, showing the near-far resistance of the analyzed algorithms.
2017-02-22T15:10:27ZAntón Haro, CarlesRodríguez Fonollosa, José AdriánZvonar, ZaranRodríguez Fonollosa, JavierIn this paper, two probabilistic adaptive algorithms for jointly detecting active users in a DS-CDMA system are compared. The first one, which is based on the theory of hidden Markov models (HMM) is proposed within the CDMA scenario and compared with the previously developed Viterbi-based algorithm. Both techniques are completely blind in the sense that no knowledge of the signature sequences, channel state information or training sequences is required for any user. After convergence, an estimate of the signature of each user convolved with its physical channel impulse response (CIR), and estimated data sequences are provided. This CIR estimate can then be used to switch to any decision-directed (DD) adaptation scheme. Performance of the algorithms is verified with simulations as well as with experimental data from an underwater acoustics (UWA) environment. In both cases, performance is found to be highly satisfactory, showing the near-far resistance of the analyzed algorithms.Frequency averaging: a useful multiwindow spectral analysis approach
http://hdl.handle.net/2117/101335
Frequency averaging: a useful multiwindow spectral analysis approach
Nadeu Camprubí, Climent; Padrell, Jaume; Esquerra Llucià, Ignasi
The multiwindow approach is a meaningful framework for nonparametric spectral estimation. It also encompasses several conventional methods as WOSA and frequency-averaged periodogram. Recently, some authors claimed that the Slepian windows of Thomson's method and other related optimal sets of windows show a better performance in terms of resolution, variance and leakage. In this paper, that claim is discussed by means of some simulation examples and by applying the various methods to speech recognition. In conclusion, frequency averaging of the periodogram is a computationally simple method that has a great flexibility for band specification and comparatively shows good performance. In fact, it is the spectral analysis technique most extensively employed for speech recognition.
2017-02-21T15:30:35ZNadeu Camprubí, ClimentPadrell, JaumeEsquerra Llucià, IgnasiThe multiwindow approach is a meaningful framework for nonparametric spectral estimation. It also encompasses several conventional methods as WOSA and frequency-averaged periodogram. Recently, some authors claimed that the Slepian windows of Thomson's method and other related optimal sets of windows show a better performance in terms of resolution, variance and leakage. In this paper, that claim is discussed by means of some simulation examples and by applying the various methods to speech recognition. In conclusion, frequency averaging of the periodogram is a computationally simple method that has a great flexibility for band specification and comparatively shows good performance. In fact, it is the spectral analysis technique most extensively employed for speech recognition.Blind multiuser identification and detection in cdma systems
http://hdl.handle.net/2117/101325
Blind multiuser identification and detection in cdma systems
Rodríguez Fonollosa, Javier; Rodríguez Fonollosa, José Adrián; Zvonar, Z; Vidal Manzano, José
Multiuser detection in code division multiple access systems usually requires either knowledge of the transmitted signature sequences and channel state information or use of a known training sequence for adaptation. We develop a scheme that can be employed for the joint adaptive blind multiuser identification and detection in asynchronous CDMA systems. This scheme relies on a multiuser Viterbi algorithm that incorporates an adaptive estimation of the overall channel impulse responses, given by the convolution of the signature sequences of the users and corresponding physical channels impulse responses. Once the overall channel responses are estimated, the blind multiuser detection algorithm performs like the maximum-likelihood sequence estimator. Results are provided to illustrate the convergence of the blind multiuser approach, near-far resistance and sensitivity to the algorithm initialization.
2017-02-21T14:37:37ZRodríguez Fonollosa, JavierRodríguez Fonollosa, José AdriánZvonar, ZVidal Manzano, JoséMultiuser detection in code division multiple access systems usually requires either knowledge of the transmitted signature sequences and channel state information or use of a known training sequence for adaptation. We develop a scheme that can be employed for the joint adaptive blind multiuser identification and detection in asynchronous CDMA systems. This scheme relies on a multiuser Viterbi algorithm that incorporates an adaptive estimation of the overall channel impulse responses, given by the convolution of the signature sequences of the users and corresponding physical channels impulse responses. Once the overall channel responses are estimated, the blind multiuser detection algorithm performs like the maximum-likelihood sequence estimator. Results are provided to illustrate the convergence of the blind multiuser approach, near-far resistance and sensitivity to the algorithm initialization.New hos-based parameter estimation methods for speech recognition in noisy environments
http://hdl.handle.net/2117/101323
New hos-based parameter estimation methods for speech recognition in noisy environments
Moreno Bilbao, M. Asunción; Tortola, S; Vidal Manzano, José; Rodríguez Fonollosa, José Adrián
The problem of recognition in noisy environments is addressed. Often, a recognition system is used in a noisy environment and there is no possibility of training it with noisy samples. Classical speech analysis techniques are based on second-order statistics and their performance dramatically decreases when noise is present in the signal under analysis. New methods based on higher order statistics (HOS) are applied in a recognition system and compared against the autocorrelation method. Cumulant-based methods show better performance than autocorrelation-based methods for low SNR
2017-02-21T14:09:57ZMoreno Bilbao, M. AsunciónTortola, SVidal Manzano, JoséRodríguez Fonollosa, José AdriánThe problem of recognition in noisy environments is addressed. Often, a recognition system is used in a noisy environment and there is no possibility of training it with noisy samples. Classical speech analysis techniques are based on second-order statistics and their performance dramatically decreases when noise is present in the signal under analysis. New methods based on higher order statistics (HOS) are applied in a recognition system and compared against the autocorrelation method. Cumulant-based methods show better performance than autocorrelation-based methods for low SNRAdaptive blind equalization using weighted cumulant slices
http://hdl.handle.net/2117/101322
Adaptive blind equalization using weighted cumulant slices
Vidal Manzano, José; Rodríguez Fonollosa, José Adrián
Many linear methods have been proposed in the literature to blindly estimate the ARMA parameters of a time series using HOS. Nevertheless, they are mainly off-line and not much has been done in the adaptive case. The method proposed in this contribution is the adaptive version of the w-slice method. The recursion is based on the inversion lemma when attempting the solution of an undetermined matrix equation. The system impulse response can be recovered regardless of the ARMA or MA character of the system. The number of operations depends on the square of the system order and it is considerably reduced with respect to previous approaches. Application to channel deconvolution is shown.
2017-02-21T14:00:01ZVidal Manzano, JoséRodríguez Fonollosa, José AdriánMany linear methods have been proposed in the literature to blindly estimate the ARMA parameters of a time series using HOS. Nevertheless, they are mainly off-line and not much has been done in the adaptive case. The method proposed in this contribution is the adaptive version of the w-slice method. The recursion is based on the inversion lemma when attempting the solution of an undetermined matrix equation. The system impulse response can be recovered regardless of the ARMA or MA character of the system. The number of operations depends on the square of the system order and it is considerably reduced with respect to previous approaches. Application to channel deconvolution is shown.Selection of correction candidates for the normalization of Spanish user generated content
http://hdl.handle.net/2117/101281
Selection of correction candidates for the normalization of Spanish user generated content
Melero, Maite; Ruiz Costa-Jussà, Marta; Lambert, Patrik; Quixal, Martí
We present research aiming to build tools for the normalization of User-Generated Content (UGC). We argue that processing this type of text requires the revisiting of the initial steps of Natural Language Processing, since UGC (micro-blog, blog, and, generally, Web 2.0 user-generated texts) presents a number of nonstandard communicative and linguistic characteristics – often closer to oral and colloquial language than to edited text. We present a corpus of UGC text in Spanish from three different sources: Twitter, consumer reviews, and blogs, and describe its main characteristics. We motivate the need for UGC text normalization by analyzing the problems found when processing this type of text through a conventional language processing pipeline, particularly in the tasks of lemmatization and morphosyntactic tagging. Our aim with this paper is to seize the power of already existing spell and grammar correction engines and endow them with automatic normalization capabilities in order to pave the way for the application of standard Natural Language Processing tools to typical UGC text. Particularly, we propose a strategy for automatically normalizing UGC by adding a module on top of a pre-existing spell-checker that selects the most plausible correction from an unranked list of candidates provided by the spell-checker. To build this selector module we train four language models, each one containing a different type of linguistic information in a trade-off with its generalization capabilities. Our experiments show that the models trained on truecase and lowercase word forms are more discriminative than the others at selecting the best candidate. We have also experimented with a parametrized combination of the models by both optimizing directly on the selection task and doing a linear interpolation of the models. The resulting parametrized combinations obtain results close to the best performing model but do not improve on those results, as measured on the test set. The precision of the selector module in ranking number one the expected correction proposal on the test corpora reaches 82.5% for Twitter text (baseline 57%) and 88% for non-Twitter text (baseline 64%).
2017-02-21T10:09:17ZMelero, MaiteRuiz Costa-Jussà, MartaLambert, PatrikQuixal, MartíWe present research aiming to build tools for the normalization of User-Generated Content (UGC). We argue that processing this type of text requires the revisiting of the initial steps of Natural Language Processing, since UGC (micro-blog, blog, and, generally, Web 2.0 user-generated texts) presents a number of nonstandard communicative and linguistic characteristics – often closer to oral and colloquial language than to edited text. We present a corpus of UGC text in Spanish from three different sources: Twitter, consumer reviews, and blogs, and describe its main characteristics. We motivate the need for UGC text normalization by analyzing the problems found when processing this type of text through a conventional language processing pipeline, particularly in the tasks of lemmatization and morphosyntactic tagging. Our aim with this paper is to seize the power of already existing spell and grammar correction engines and endow them with automatic normalization capabilities in order to pave the way for the application of standard Natural Language Processing tools to typical UGC text. Particularly, we propose a strategy for automatically normalizing UGC by adding a module on top of a pre-existing spell-checker that selects the most plausible correction from an unranked list of candidates provided by the spell-checker. To build this selector module we train four language models, each one containing a different type of linguistic information in a trade-off with its generalization capabilities. Our experiments show that the models trained on truecase and lowercase word forms are more discriminative than the others at selecting the best candidate. We have also experimented with a parametrized combination of the models by both optimizing directly on the selection task and doing a linear interpolation of the models. The resulting parametrized combinations obtain results close to the best performing model but do not improve on those results, as measured on the test set. The precision of the selector module in ranking number one the expected correction proposal on the test corpora reaches 82.5% for Twitter text (baseline 57%) and 88% for non-Twitter text (baseline 64%).Estimation of the modulation index of cpm signals using hos
http://hdl.handle.net/2117/101254
Estimation of the modulation index of cpm signals using hos
Rodríguez Fonollosa, Javier; Rodríguez Fonollosa, José Adrián
Three simple methods are proposed for the estimation of the modulation index of continuous phase modulated signals in noise. These methods employ the estimated autocorrelation and fourth-order cumulant sequences of the received signal after sampling at the symbol rate. Analytic expressions are derived for the asymptotic mean and variance of the estimated parameters which are corroborated by means of Monte Carlo simulations. The performance of the methods is illustrated graphically and numerically. It is concluded that, under significant noise degradation, only the scheme based on the fourth-order cumulant sequence can be used to estimate consistently the modulation index h in the range 0(h(1.
2017-02-20T16:54:59ZRodríguez Fonollosa, JavierRodríguez Fonollosa, José AdriánThree simple methods are proposed for the estimation of the modulation index of continuous phase modulated signals in noise. These methods employ the estimated autocorrelation and fourth-order cumulant sequences of the received signal after sampling at the symbol rate. Analytic expressions are derived for the asymptotic mean and variance of the estimated parameters which are corroborated by means of Monte Carlo simulations. The performance of the methods is illustrated graphically and numerically. It is concluded that, under significant noise degradation, only the scheme based on the fourth-order cumulant sequence can be used to estimate consistently the modulation index h in the range 0(h(1.Performance evaluation of interference cancellation techniques using adaptive antennas
http://hdl.handle.net/2117/101104
Performance evaluation of interference cancellation techniques using adaptive antennas
Antón Haro, Carles; Rodríguez Fonollosa, José Adrián; Rodríguez Fonollosa, Javier
Two array-based algorithms, which jointly exploit or compensate for the spatial and temporal characteristics of the propagation channel, are proposed for intercell interference suppression in UMTS scenarios. The first one is the array extension of the Viterbi algorithm and is referred to as Vector Viterbi algorithm (VVA). The second algorithm, known as filtered training sequence multisensor receiver (FTS-MR), belongs to a class of algorithms in which a narrowband beamformer is placed prior to the MLSE detector. In order to assess performance of the proposed schemes, a set of link-level computer simulations adopting FRAMES' proposal for UMTS air-interface as well as realistic channel models for third generation communication systems is provided, Simulation results reveal gains, in terms of C/I, of 7-10 dB for the VVA with respect to the conventional VA and even higher for the FTS-MR.
2017-02-15T15:05:51ZAntón Haro, CarlesRodríguez Fonollosa, José AdriánRodríguez Fonollosa, JavierTwo array-based algorithms, which jointly exploit or compensate for the spatial and temporal characteristics of the propagation channel, are proposed for intercell interference suppression in UMTS scenarios. The first one is the array extension of the Viterbi algorithm and is referred to as Vector Viterbi algorithm (VVA). The second algorithm, known as filtered training sequence multisensor receiver (FTS-MR), belongs to a class of algorithms in which a narrowband beamformer is placed prior to the MLSE detector. In order to assess performance of the proposed schemes, a set of link-level computer simulations adopting FRAMES' proposal for UMTS air-interface as well as realistic channel models for third generation communication systems is provided, Simulation results reveal gains, in terms of C/I, of 7-10 dB for the VVA with respect to the conventional VA and even higher for the FTS-MR.Assessment of the effect of the financial crisis on agents’ expectations through symbolic regression
http://hdl.handle.net/2117/101084
Assessment of the effect of the financial crisis on agents’ expectations through symbolic regression
Claveria, Oscar; Monte Moreno, Enrique; Torra Porras, Salvador
Agents’ perceptions on the state of the economy can be affected during economic crises.
Tendency surveys are the main source of agents’ expectations. The main objective of this study
is to assess the impact of the 2008 financial crisis on agents’ expectations. With this aim, we
evaluate the capacity of survey-based expectations to anticipate economic growth in the United
States, Japan, Germany and the United Kingdom. We propose a symbolic regression (SR) via
genetic programming approach to derive mathematical functional forms that link survey-based
expectations to GDP growth. By combining the main SR-generated indicators, we generate
estimates of the evolution of GDP. Finally, we analyse the effect of the crisis on the formation
of expectations, and we find an improvement in the capacity of agents’ expectations to anticipate
economic growth after the crisis in all countries except Germany.
2017-02-15T14:06:00ZClaveria, OscarMonte Moreno, EnriqueTorra Porras, SalvadorAgents’ perceptions on the state of the economy can be affected during economic crises.
Tendency surveys are the main source of agents’ expectations. The main objective of this study
is to assess the impact of the 2008 financial crisis on agents’ expectations. With this aim, we
evaluate the capacity of survey-based expectations to anticipate economic growth in the United
States, Japan, Germany and the United Kingdom. We propose a symbolic regression (SR) via
genetic programming approach to derive mathematical functional forms that link survey-based
expectations to GDP growth. By combining the main SR-generated indicators, we generate
estimates of the evolution of GDP. Finally, we analyse the effect of the crisis on the formation
of expectations, and we find an improvement in the capacity of agents’ expectations to anticipate
economic growth after the crisis in all countries except Germany.