Ricardo Kehrle Miranda

ESPRIT-Hilbert-Based Audio Tampering Detection With SVM Classifier for Forensic Analysis via Electrical Network Frequency

Audio authentication is a critical task in multimedia forensics demanding robust methods to detect and identify tampered audio recordings. In this paper, a new technique to detect adulterations in audio recordings is proposed by exploiting abnormal variations in the electrical network frequency (ENF) signal eventually embedded in a questioned audio recording. These abnormal variations are caused by abrupt phase discontinuities due to insertions and suppressions of audio snippets during the tampering task. First, we propose an ESPRIT-Hilbert ENF estimator in conjunction with an outlier detector based on the sample kurtosis of the estimated ENF. Next, we use the computed kurtosis as an input for a support vector machine classifier to indicate the presence of tampering. The proposed scheme, herein designated as SPHINS, significantly outperforms related previous tampering detection approaches in the conducted tests. We validate our results using the Carioca 1 corpus with 100 unedited authorized audio recordings of phone calls.

High accuracy and low complexity adaptive Generalized Sidelobe Cancelers for colored noise scenarios

Abstract The Generalized Sidelobe Canceler (GSC) is a beamforming scheme which is applied in many fields such as audio, RADAR, SONAR and telecommunications. Recently, the adaptive Reduced Rank GSC (RR-GSC) has been proposed for applications with a large number of sensors. Due to its dimensionality reduction step, the adaptive RR-GSC achieves an enhanced performance in comparison with the standard GSC. However, both standard GSC and RR-GSC have their performance drastically degraded in the presence of colored noise. In this paper, we propose to extend further the GSC and the RR-GSC for colored noise scenarios. As shown in this paper, such improvement in colored noise scenarios can be obtained by incorporating a stochastic or a deterministic prewhitening step in the GSC and RR-GSC algorithms. Since the prewhitening increases the computational complexity, a block-wise reduced rank stochastic gradient GSC beamformer is also proposed. The block-wise step allows only one prewhitening step per block while in the previous schemes one per sample was needed. Another proposed advance in colored noise scenarios is the incorporation of the Vandermonde Invariance Transform (VIT). The VIT works as a pre-beamformer which reduces the interferent power of the undesired sources and the colored noise effect. We show by means of simulations the improved results even for highly correlated scenarios.

Generalized sidelobe cancellers for multidimensional separable arrays

The usage of antenna arrays brought innumerable benefits to radio systems in the last decades. Arrays can have multidimensional structures that can be exploited to achieve superior performance and lower complexity. However, the literature has not explored yet all the advantages arising from these features. This paper uses tensors to provide a method to design efficient beamformers for multidimensional antenna arrays. In this work, the generalized sidelobe canceller (GSC) is extended to a multidimensional array to create the proposed R-Dimensional GSC (R-D GSC). The proposed scheme has a lower computational complexity and, under certain conditions, exhibits an improved signal to interference and noise ratio (SINR).

Evaluation of Space-Time-Frequency (STF)-Coded MIMO-OFDM Systems in Realistic Channel Models

By taking into account several dimension of the transmitted signal, such as space, frequency, period and time, MIMO-OFDM systems allow an increased spectral efficiency and an improved identifiability in comparison to matrix solutions. In this paper, we evaluate MIMO-OFDM systems for geometric scenarios where the narrow band approximation is violated. To this end a new data model is proposed to better represent the behavior of the system in the presence of wide band signals. Moreover, we also relax the assumption that the amount of transmitted antennas is equal to the number of transmitted symbols.

Low-Complexity and High-Accuracy Semi-blind Joint Channel and Symbol Estimation for Massive MIMO-OFDM

In order to fully exploit the scarce spectrum, antenna arrays are incorporated into wireless communication devices in 4G and 5G communication networks to deploy MIMO-OFDM systems. Recently, the least squares Khatri–Rao factorization has been applied to MIMO-OFDM systems for semi-blind joint channel and symbol estimation. Its cubic computational complexity is prohibitive when the number of transmit and receive antennas is very large. Therefore, the average vector and Hadamard ratio rank one approximation has been proposed for MIMO-OFDM systems, showing a linear complexity, but being limited to channels and transmitted symbols with offsets. In this paper, we present four novel MIMO-OFDM algorithms for massive antenna array systems that outperform the state-of-the-art approaches in terms of complexity and/or accuracy. The four proposed schemes are the alternating least squares with vector selection initialization, the vector projection rank one approximation including vector selection rank one initialization, the factorization based on eigenvalue decomposition with eigenvector projection and the factorization based on sectional truncated singular value decomposition and vector projection. Our analytical complexity analysis and numerical results corroborate the trade-offs offered by the different receiver algorithms in terms of complexity, parallelism and performance.

Tensor based Blind Source Separation for Current Source Density Analysis of evoked potentials from somatosensory cortex of mice

In order to understand brain mechanisms and functionalities, neural probes with electrode arrays are incorporated into mice and Local Field Potentials (LFP) are recorded indicating the activities of groups of neurons. Next, the brain activity can be analyzed in terms of Current Source Density (CSD), which are computed via the LFP. In this paper, we propose the analysis of the somatosensory cortex signals of a mouse applying Blind Source Separation (BSS) schemes. In contrast to the standard CSD, we show that signal separation using BSS schemes can be useful to identify groups of neurons of different layers of the somatosensory cortex that are associated. Another contribution of this work is to propose the use of the PARAFAC model on the analysis of somatosensory cortex signals, whose results are consistent with results obtained via Spatiotemporal Independent Component Analysis (stICA).

Low complexity performance assessment of a sensor array via unscented transformation

Due to the advances on electronics, applications of antenna array signal processing are becoming more frequent. When employing antenna arrays for beamforming, the signal to interference and noise ratio (SINR) should be assessed. Many factors can affect the SINR such as the array element positioning error and the direction of arrival (DOA) estimation error. In these cases, the assessment is traditionally performed via the SINR average obtained using Monte Carlo (MC) simulations. However, this approach requires a great amount of realizations that demand a high computational effort and processing time due to its slow convergence. In this paper, we propose a low complexity performance assessment of the average SINR via unscented transformation. Compared to MC simulations, our proposed method requires only a few trials and has a negligible computational complexity, yet giving a comparable SINR when the DOA estimation is perturbed. When the antenna elements positioning is perturbed, a multivariate scenario arises. For multivariate scenario the proposed scheme has an exponential increase in complexity, therefore, still being advantageous for a small number of antennas.

Efficient and low cost MIMO communication architecture for smartbands applied to postoperative patient care

Frequently, postoperative patient care requires a long period of observation in hospitals by the medical board. During this observation period, the medical board manually inserts time to time vital information of the patients into the medical information system. Such manual procedure can be improved and become more reliable if the patients are equipped with wearable devices that allow the real-time data acquisition and the data processing by machine learning systems. In this work, we propose an efficient and low cost monitoring system for postoperative patient care based on commercial smartbands. Since the current smartbands are restricted to single input single output (SISO) communication, i.e. only one smartband can connect to one smartphone in a short distance range, we propose to expand the commercial smartband capability to a multiple input multiple output (MIMO) communication by proposing a new architecture based on a signal concentrator. According to our experimental results, our proposed architecture with a single concentrator allows a 248: N communication, i.e. the simultaneous usage of 248 smartbands in a same room and the data collected data is transmitted to any number N of physicians without distance restrictions. By considering a system with multiple concentrators, we also propose an M:N architecture communication for smartbands.

Radiofrequency energy harvesting system based on a rectenna array in urban environments

Energy harvesting has become very attractive due to the extended usage time of devices. Among several forms of recycling energy, radiofrequency (RF) harvesting has been suggested due to its wide availability mainly in urban areas. Its applications range from sensor nodes to charging low power consumption portable devices and depend on the amount of antennas. In this paper, we evaluate the feasible application of RF harvesting for charging a cell phone. To validate our analysis, we conduct an RF measurement campaign at four important locations in Brasilia, Brazil. Considering the average incidence of 11 dBm, we achieve the final value of 2.5 mW/m2. With an incident power of +10 dBm, only 2 rectennas per hour are needed to charge a cell phone whose battery is approximately 3.72 mWh. We perform a comparison between rectenna arrays and simple antennas directly connected to one external matching circuit, dismissing adaptive beamforming circuits as a way of avoiding intermediate energy losses. In order to apply RF energy harvesting in higher power consumption devices, we propose a rectenna array system which increases considerably the amount of recycled power. For both Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) paths, harvesting systems based on rectenna arrays outperform standard antenna array based solutions.

PCA-Kalman based load forecasting of electric power demand

Electricity demand time series are stochastic processes related to climate, social and economic variables. By predicting the evolution of such time series, electrical load forecasting can be performed in order to support the electrical grid planning. In this paper, we propose a Kalman based load forecasting system for daily demand forecasting. Our proposed approach incorporates a Principal Component Analysis (PCA) of the input variables obtained from linear and nonlinear transformations of the candidate time series. In order to validate our predicting scheme, data collected from Brasília distribution company has been used. Our proposed approach outperforms state-of-the-art approaches based on state space and artificial neural networks.