Hamming Window Research Articles

Trainable windows for SincNet architecture

SincNet architecture has shown significant benefits over traditional Convolutional Neural Networks (CNN), especially for speaker recognition applications. SincNet comprises parameterized Sinc functions as filters in the first layer followed by convolutional layers. Although SincNet is compact in nature and offers top-level understanding of the features extracted, the effect of window function used in SincNet is not thoroughly addressed yet. Hamming and Hann are popularly used as the default time-localized windows to reduce spectral leakage. Hence, a comprehensive investigation of 28 different windowing functions on SincNet architecture towards speaker recognition task using TIMIT dataset was performed in this work. Additionally, “trainable” window functions were configured with tunable parameters to characterize the performance. The paper benchmarks the effect of the time-localized windowing function in terms of the bandwidth, side-lobe suppression, and spectral leakage for the filter banks employed in the first layer of the SincNet architecture. Trainable Gaussian and Cosine-Sum functions exhibited relative improvement of 41.46% and 82.11% in the sentence level classification error rate over Hamming window when employed on SincNet architecture.

Blind modulation classification in multiple input and output‐orthogonal frequency division multiplexing using time‐frequency analysis and customized convolutional neural network architecture

AbstractMIMO with OFDM is a powerful communication technology in both army and civilian environments. Because there is no prior knowledge of carrier state information or signal overlapping in MIMO‐OFDM systems, outdated probability and feature‐based approaches can be applied. However, these methods cannot easily be used to provide the channel characteristics. Blind modulation categorization is a critical stage in deploying 5G networks. The issue of blind modulation classification for a multiple input and output (MIMO) antenna system using orthogonal frequency division multiplexing (OFDM) is investigated in this research. The independent component analysis (ICA) method has been used to separate the MIMO modulated signal, and the smoothed pseudo Wigner‐Ville distribution (SPWVD) with a Hamming window of 0.2 s has been used to analyze the time‐frequency characteristics of time‐domain modulated signals to get around the challenge of blind attenuation categorization in MIMO‐OFDM arrangements. The recovered time‐frequency characteristics are then transformed into red‐green‐blue (RGB) spectrogram pictures and a perfect‐tweaked customized CNN in which CNN (convolutional neural network) is integrated with LSTM (long short term memory) is used to categorize modulation kinds based on the illustrations of RGB spectrograms. Finally, a module for merging decisions based on majority voting was employed to combine the categorization findings of all receiving antennas. As a result, a customized CNN‐based blind modulation technique outperforms existing works to find the exact modulation type in the MIMO‐OFDM communication.

Range-angle-dependent beamforming for FDA radar with Hamming interelement spacing and sinusoidal multicarrier approach

By employing a tiny frequency offset across the array elements, a frequency diverse array (FDA) produces a beampattern with the property of being range-angle-dependent. However, the beampattern of a traditional FDA has many maxima at undesired range-angle couples, which is unfavorable for target localization and jamming suppression. In this paper, a novel FDA scheme with Hamming interelement spacing, namely, the HIS-FDA, is proposed. By adopting Hamming window weighted interelement spacing and frequency offsets optimized by the genetic algorithm, the HIS-FDA produces a dot-shaped well-performed beampattern, which has superior performance over the existing FDA systems, especially in the angle dimension. The multiple input multiple output technique and multiple matched filters are employed at the receiver to solve the time-varying problem. A new multicarrier (MC) technique termed the sinusoidal MC approach is also presented to improve the beampattern performance, especially in the range dimension. Our proposed scheme outperforms the existing FDA systems in terms of the mainlobe width in the range and angle profiles, the sidelobe level, and the range-angle spatial region of the mainlobe. Comparison results are provided to validate the superiority of the proposed scheme.

Performance Analysis of Linearly Arranged Concentric Circular Antenna Array with Low Sidelobe Level and Beamwidth Using Robust Tapering Technique.

In this research, a novel antenna array named Linearly arranged Concentric Circular Antenna Array (LCCAA) is proposed, concerning lower beamwidth, lower sidelobe level, sharp ability to detect false signals, and impressive SINR performance. The performance of the proposed LCCAA beamformer is compared with geometrically identical existing beamformers using the conventional technique where the LCCAA beamformer shows the lowest beamwidth and sidelobe level (SLL) of 12.50° and -15.17 dB with equal elements accordingly. However, the performance is degraded due to look direction error, for which robust techniques, fixed diagonal loading (FDL), optimal diagonal loading (ODL), and variable diagonal loading (VDL), are applied to all the potential arrays to minimize this problem. Furthermore, the LCCAA beamformer is further simulated to reduce the sidelobe applying tapering techniques where the Hamming window shows the best performance having 17.097 dB less sidelobe level compared to the uniform window. The proposed structure is also analyzed under a robust tapered (VDL-Hamming) method which reduces around 69.92 dB and 48.39 dB more sidelobe level compared to conventional and robust techniques. Analyzing all the performances, it is clear that the proposed LCCAA beamformer is superior and provides the best performance with the proposed robust tapered (VDL-Hamming) technique.

Four window functions for simulation of vibration transmission effects in a planetary gear set

Analysis of vibration spectrum is one of the main investigation methods in the research filed for condition monitoring on planetary gear sets. However, such a gear set always exhibits complicated spectral structures in the vibration spectrum because of the planet carrier’s rotation, and the path distance from the fixed transducer to the moving meshing point will always be changed, which brings a modulation effect on the vibration signals received by a fixed transducer. Aiming to investigate the vibration transmission effects of planetary gear sets, four different window functions are introduced in this paper to simulate the transmission path changing with time i.e. the path between moving meshing position and a fixed transducer. Comparison results show that the Hamming window function may be a more suitable choice to model the transmission path changing with time in the planetary gear set.

Characterization of Parkinson's Disease using spectral features of kinetic tremor: correlation of on-line digitized handwriting and classical motor scales

Tremor is one of the motor impairments of Parkinson’s disease (PD). To date, kinetic tremor in PD is barely examined and there is lack of information on the relation between its severity and scales related to specific motor deficits [1]. In this study, we aimed at investigating the correlation between handwriting-related kinetic tremor and motor score measures by motor part of Unified Parkinson's Disease Rating Scale (UPDRS-III) [2], using a digitizing tablet. In this preliminary study, eight PD patients (7 M/1 F; age 74.5±8 years) draw an accurate Archimedes’ Spiral (AS) and fast, overlapped Circles (C) for a duration of 15 seconds. All patients underwent motor deficit assessment using UPDRS-III. Power Spectral Density of both velocity and acceleration profiles in their horizontal, vertical, and curvilinear components was estimated by using Welch’s method, with a Hamming window on intervals of 5 s and a 50% overlap. To analyze the power distribution related to different movement-associated phenomena, the ratio between two frequency bands (BME/BT) and the BT bandwidth (BW) were calculated for each subject. BME is the band of voluntary Movement Execution required by the task, ranging from 0.2 to 4 Hz, and BT is the band associated with involuntary Tremor, ranging from 4.0 to 12 Hz [3]. Normalized Jerk, a classic kinematic feature representing handwriting fluidity, was also estimated. The correlation between the parameters and UPDRS-III scores were assessed using Spearman’s rank correlation coefficients. All the evaluations were conducted in the pharmacological on state of PD patients. A positive correlation was found between Jerk and UPDRS-III scores (Table 1). On the contrary BME /B T correlates negatively with motor scale scores in horizontal and vertical velocities (Vx, Vy) for both tasks, vertical and curvilinear acceleration (Ay, Ac) for AS task, horizontal and vertical acceleration (Ax, Ay) for C task. Only Ay of C task shows correlation with motor scores. The results highlight that the severity of motor deficits in PD patients, as assessed by a widely employed motor scale, correlates with the outcomes of spectral and kinematic analysis of handwriting that indicate a loss of fluency, an increased power at BT level and a thinning of the spectral peak of B T . This suggests that handwriting assessment of parkinsonian dysgraphia can be used to implement clinical evaluation and represents a non-invasive, low-cost method for the identification of objective and reproducible biomarkers of kinetic tremor.

The Perspective of Binary and Tertiary Artificial Neurons Mutual Application for White Noise Quality Analysis within Hamming Window Convolution Calculated by Different Modules

The work is devoted to simplification of binary sequence quality testing, obtained either from physical noise generators, or from instable part of human biometric data. The problem originated from exponential computational complexity of a frequently used test “pressing keys by mistake”. As a result, such a test can be realized only on a powerful computer. The work shows that the simplest testing of white noise quality is performed by counting of units. It can be considered as a special case of Hamming window convolution, calculated by module two in comparison with unit code «111...11». A generalization of this test with its analogues using Hamming window convolutions calculated by modules 2 and 256 was presented. Programs based on MathCAD for performing numerical tests are given. Hamming spectral line probability amplitude distribution was presented. It was suggested to enhance verification of white noise with additional analysis using tertiary artificial neurons. The assessment of benefit from transition to mutual application of binary and tertiary artificial neurons when verifying the white noise hypothesis for one or another binary sequence was given. The calculational procedures considered in the article do not require significant computational resources. They are oriented on low power and consumption reliable microcontrollers of SIM-cards and SD-cards. This is stipulated by linear computational complexity of all tests built on Hamming window convolution irrespectively of module values, they are calculated by. The number of output quantizer levels of one or another artificial neurons has no effect on computational complexity.

Non-Invasive Microwave-Based Imaging System for Early Detection of Breast Tumours.

This work introduces a microwave-based system able to detect tumours in breast phantoms in a non-invasive way. The data acquisition system is composed of a hardware system which involves high-frequency components (antennas, switches and cables), a microcontroller, a vector network analyser used as measurement instrument and a computer devoted to the control and automation of the operation of the system. Concerning the software system, the computer runs a Python script which is in charge of mastering and automatising all the required stages for the data acquisition, from initialisation of the hardware system to performing and saving the measurements. We also report on the design of the high-performance broadband antenna used to carry out the measurements, as well as on the algorithm employed to build the final medical images, based on an adapted version of the so-called Improved Delay-and-Sum (IDAS) algorithm improved by a Hamming window filter and averaging preprocessing. The calibration and start-up of the system are also described. The experimental validation includes the use of different tumour models with different dielectric properties inside the breast phantom. The results show promising tumour detection capabilities, even when there is low dielectric contrast between the tumoural and healthy tissues, as is the usual case for dense breasts in young women.

Credibility evaluation system for The Internet of Things: Impact and Implications for Health Care Delivery

Due to the imperfection of the technology of speech recognition embedded in the Internet of things devices, the evaluation results of the credibility evaluation system for speech recognition control of the Internet of things devices have errors, so a credibility evaluation system for the Internet of things devices based on speech recognition technology is designed. First of all, the hardware structure of the system is designed, the MCU main control chip is developed, and the system hardware is optimized with the help of MCU. In the software design, the digital filter is used for speech signal preprocessing, the rectangular window and Hamming window are used to frame and add windows to the speech, the function related gate is introduced to divide the fault events and basic events, and the credibility evaluation system of Internet of things equipment based on speech recognition technology is completed. The performance test results show that compared with the traditional system, the average recognition rate of the designed system for different lengths of instructions is improved.

Assessment of underlying topography and forest height inversion based on TomoSAR methods

ABSTRACT Due to the strong penetrability, long-wavelength synthetic aperture radar (SAR) can provide an opportunity to reconstruct the three-dimensional structure of the penetrable media. SAR tomography (TomoSAR) technology can resynthesize aperture perpendicular to the slant-range direction and then obtain the tomographic profile consisting of power distribution of different heights, providing a powerful technical tool for reconstructing the three-dimensional structure of the penetrable ground objects. As an emerging technology, it is different from the traditional interferometric SAR (InSAR) technology and has advantages in reconstructing the three-dimensional structure of the illuminated media. Over the past two decades, many TomoSAR methods have been proposed to improve the vertical resolution, aiming to distinguish the locations of different scatters in the unit pixel. In order to cope with the forest mission of European Space Agency (ESA) that is designed to provide P-band SAR measurements to determine the amount of biomass and carbon stored in forests, it is necessary to systematically evaluate the performance of forest height and underlying topography inversion using TomoSAR technology. In this paper, we adopt three typical algorithms, namely, Capon, Multiple Signal Classification (MUSIC), and Compressed Sensing (CS), to evaluate the performance in forest height and underlying topography inversion. The P-band airborne full-polarization (FP) SAR data of Lopè National Park in the AfriSAR campaign implemented by ESA in 2016 is adopted to verify the experiment. Furthermore, we explore the effects of different baseline designs and filter methods on the reconstruction of the tomographic profile. The results show that a better tomographic profile can be obtained by using Hamming window filter and Capon algorithm in uniform baseline distribution and a certain number of acquisitions. Compared with LiDAR results, the root-mean-square error (RMSE) of forest height and underlying topography obtained by Capon algorithm is 2.17 m and 1.58 m, which performs the best among the three algorithms.

Implementation of a System for Assessing the Quality of Spoken English Pronunciation Based on Cognitive Heuristic Computing.

This paper analyzes and investigates the quality assessment of spoken English pronunciation using a cognitive heuristic computing approach and designs a corresponding spoken pronunciation quality assessment system for practical training. Using the general Goodness of Pronunciation assessment algorithm as a benchmark, the shortcomings of the traditional Goodness of Pronunciation method are explored through statistical experiments, and the validity of the overall posterior probability output from the speech model for pronunciation quality assessment is verified. For the analysis of rhythm, there is no common algorithm framework, but in this paper, the F0 similarity algorithm based on dynamic time regularization and the stop similarity algorithm based on forced alignment is proposed for the two main factors of rhythm, intonation, and pause, respectively. After framing, the Hamming window processing is used to make the signal smoother, reduce the side lobe size after fast Fourier transform processing, and solve the problem of spectrum leakage. Compared with the ordinary rectangular window function, the Hamming window can obtain a higher quality spectrum. And combined with CTC for speech recognition modeling, the recognition rates are comparable in the case of using BLSTM and bidirectional threshold cyclic unit BGRU as the hidden layer unit, respectively, and the training time is 23% less than BLSTM using BGRU; in addition, the BGRU-CTC model is improved by using a 2-BGRU-CTC model with 256 hidden layer nodes, so that the error rate of phoneme recognition is reduced to 33%. The effectiveness of the algorithm framework is also verified through experiments, which further proves the effectiveness of our proposed phoneme segment feature and rhyme similarity algorithm.

Optimum Extrapolation Techniques for Two-Dimensional Antenna Array Tapered Beamforming

Optimizing antenna arrays is essential for achieving efficient beamforming with very low sidelobe level (SLL) where adopting tapered window functions is one of the straightforward efficient techniques for achieving this goal. Recently, two-dimensional (2D) beamforming has been extensively required for many applications; therefore, this paper proposes two extrapolation techniques applied to one-dimensional (1D) tapered functions to efficiently feed 2D antenna arrays using cross-linear and adaptive radial tapering techniques. The first proposed 2D cross-linear tapering technique determines the 2D tapering coefficients by Hadamard multiplication of two right-angled grids of repeated 1D functions, while the second proposed adaptive radial tapering technique locates the antenna element in the 2D array in terms of its radial distance with respect to the array center, then converts this distance to an element index in a virtual 1D tapering window to determine the element weighting value. The adaptive radial tapering technique is optimized for achieving the minimum SLLs. The two proposed techniques are analyzed and discussed, where it is found that the adaptive radial tapering provides deeper SLLs compared to the cross-linear tapering technique. The two extrapolation techniques are examined for four window functions including triangular (Bartlett), Hamming, cosine-square, and Blackman windows, and the simulation results show that for extrapolating the Blackman window using adaptive radial tapering, a –50 dB SLL can be achieved which is independent on the array size, while cross-linear tapering provides –35 dB and –41 dB SLLs for and antenna arrays, respectively.

A Comparison Study for Cardiac Arrhythmias Noise Cancellation Techniques

The electrocardiogram (ECG), which represents the electrical activity of the heart, is always chosen as the basic signal for diagnosing cardiac abnormalities and detecting the patient's states. Generally, the various filters are applied (preprocessing) to denoise the artifacts from ECG signal. This step makes the decision making and diagnosis simpler and faster. The most common noise sources are power line interference and baseline wandering. This article discusses different filtering techniques used to denoise the ECG signals. Electrocardiogram signals were downloaded from the MIT-BIH Arrhythmia Database. The MIT-BIH Database was the first set of standard test materials that is generally available to evaluate arrhythmia detection. To remove baseline wander, 4 filters were designed. The filters are a fourth-order high-pass Butterworth filter, a second-order finite impulse response Hamming window, an finite impulse response rectangular window with length L, and an finite impulse response Hanning window. To extract the power line interference (60 Hz), 4 filters were designed: a second-order digital bandpass Butterworth filter, a second-order finite impulse response filter, a Chebyshev filter, and an finite impulse response notch filter with Hanning window. Finally, 5 parameters were used to evaluate filter performance; the parameters are frequency response, phase response, average filter delay (group delay), phase delay, and signal-to-noise ratio. The results show that, for baseline wander filters, a fourth-order high-pass Butterworth filter removes more noise and has no time delay. Power line interference filters have the same signal-to-noise ratio approximately.

Toward a cylinder pressure signals-based active synthesis algorithm for engine sound

The active sound design (ASD) technology is one of the most flexible methods to achieve the diverse design of automobile sound, in which the active sound synthesis algorithm is essential. However, in previous works, the discontinuity phenomena always appear in the synthesized sound. Thus, a new method of active engine sound synthesis based on cylinder pressure signals is proposed in this paper for the sound synthesis algorithm of ASD technique. Firstly, the index matrix of sound grains is constructed based on the periodicity of engine cylinder pressure signal. Then the frame length and frame shift are defined by tracing the time step of Revolutions Per Minute (RPM) of engine, and the hamming window function is constructed to extract sound grains. Subsequently, the overlap-and-add (OLA) method is used to realize the synthesis of engine sounds with different durations. In addition, the code program of proposed algorithm is written and the ASD prototype system is built respectively to synthesize the engine sounds. Finally, a similarity evaluation metric is defined to verify the effectiveness of proposed algorithm. The results show that our algorithm not only restitutes the main features of original sound, but also satisfies both the real-time feedback on the driving status and the continuity of synthesized sound. It contributes to the development of real-time ASD engineering for automobiles.

Multibit Coded Passive Hybrid Resonator Based RFID Transponder With Windowing Analysis

This study proposes novel design options for a compact passive transponder that incorporates both stepped impedance resonators (SIR) and rectangular stub resonators in a hybrid approach. The analysis aims to support the progress of chipless radio frequency identification (RFID) technology by overcoming barriers like cost, size, and coding capacity. A step-by-step analysis of the design ensures a 10-bit coding capacity with a physical footprint of 32 mm × 15 mm having -25.66 dBsm of maximum radar cross-section (RCS) response. The same has been analysed at the receiver with a rectangular windowing approach which enhances the coding capacity and security features of the RFID system. Further, this has been analysed with Hamming, Hanning, and Bartlett windows to show the effectiveness of the windowing techniques in RFID coding. The performance of the tag is analysed for different tilt-angles and read-distances to strengthen the detection accuracy. The hybrid resonator (HR) with windowing analysis is a novel approach to enhance the usage of the transponder by improving the coding capacity, information storage capability, and detection performance in various retail and healthcare domains in the RFID industry.

Multimodal biomicroscopic system for the characterization of cells with high spatial phase sensitivity and sub-pixel accuracy.

Multimodal analysis is highly advantageous for various biomedical applications including cancer and brain studies. Simultaneous measurement of quantitative phase with sub-pixel accuracy and fluorescence image is difficult to achieve in single measurement. Conventionally, off-axis interferograms are analyzed using the Fourier-transform method which limits the accuracy of the phase maps by pixel size, and usually the location of the carrier peak is in sub-pixel. We report a multimodal microscopic system consisting of high-resolution (HR) quantitative phase interferometer to retrieve sub-pixel accuracy in phase imaging and an oblique-illumination-based fluorescence imaging system which decouples the excited light from emitted signal light to avoid saturation of the camera, both integrated into a single unit. Here, highly resolved phase maps are obtained using a two-step process. First, using a speckle-free illumination which offers high spatial phase sensitivity. Second, using a hamming window for accurate estimation of original signal frequency information and HR discrete Fourier transform (DFT) which offers sub-pixel accuracy in phase measurements. HR-DFT has computational load of , where is the size of the interferogram and is the upsampling factor, making system computationally more robust and efficient compared to zero-padded FFT. The experiment is conducted on MG63 osteosarcoma and human mesenchymal stem cells (hMSCs) and their quantitative parameters are extracted with significantly improved accuracy. The average phase for MG63 cells and hMSCs, for nucleus is obtained to be 8.02 rad ± 0.80 rad and 4.29 rad ± 0.43 rad, respectively, and for cytoplasm is obtained to be 2.63 rad ± 0.96 rad and 1.73 rad ± 0.57 rad, respectively.

An Improved PMF-FFT Algorithm Based on Hamming Window

When the rocket uses low-orbit satellite space-based measurement and control, due to the relatively large Doppler shift in relative motion, a spread spectrum system is needed to improve anti-interference; the spread spectrum signal under high dynamic conditions often uses partial matched filtering based on FFT (PMF-FFT) algorithm, but due to the introduction of FFT, this algorithm has serious scallop loss, which affects the system’s capture performance. Through the improved method of adding a Hamming window to the system, theoretical analysis and simulation show that it effectively suppresses scallops. Loss has a certain meaning for the acquisition of spread spectrum signals.

Design of FIR digital filters using Semi-ellipse window

A fixed window function which is similar in shape to a semi-ellipse is proposed. The semi–ellipse which has its major axis to be equal to the window length and the minor axis at unity produced about 4.2 dB lower ripple ratio than the rectangular window. The proposed window function is derived from the equation of an ellipse in the explicit and parametric forms. First of all, the spectral characteristic of the proposed window is studied in terms of spectral parameters and compared with other fixed windows like Rectangular, Bartlett, Hann, Hamming and Blackman windows. The window simulation results reveal that the proposed window produced comparable spectral characteristic with existing standard fixed windows. Secondly, the paper presents the application of the proposed window in a digital filter design. The filter analysis comparison results with other fixed windows namely Bartlett, Von Hann, Hamming, and Kaiser window, an adjustable window, confirm that filter design with the proposed window exhibits good spectral characteristic, and can be used to design better filter than the Bartlett window using less than half the Bartlett’s filter length for a fixed transition width. The similicity of its coefficients formulation and design algorithm makes it a good choice for digital filter design applications.

Evaluation of One-Class Classifiers for Fault Detection: Mahalanobis Classifiers and the Mahalanobis–Taguchi System

Today, real-time fault detection and predictive maintenance based on sensor data are actively introduced in various areas such as manufacturing, aircraft, and power system monitoring. Many faults in motors or rotating machinery like industrial robots, aircraft engines, and wind turbines can be diagnosed by analyzing signal data such as vibration and noise. In this study, to detect failures based on vibration data, preprocessing was performed using signal processing techniques such as the Hamming window and the cepstrum transform. After that, 10 statistical condition indicators were extracted to train the machine learning models. Specifically, two types of Mahalanobis distance (MD)-based one-class classification methods, the MD classifier and the Mahalanobis–Taguchi system, were evaluated in detecting the faults of rotating machinery. Their performance for fault detection on rotating machinery was evaluated with different imbalanced ratios of data by comparing with binary classification models, which included classical versions and imbalanced classification versions of support vector machine and random forest algorithms. The experimental results showed the MD-based classifiers became more effective than binary classifiers in cases in which there were much fewer defect data than normal data, which is often common in the real-world industrial field.

Precise inland surface altimetry (PISA) with nadir specular echoes from Sentinel-3: Algorithm and performance assessment

In the recent years satellite radar altimetry has evolved from pulse-limited low resolution mode (LRM) to a synthetic aperture radar (SAR) high resolution mode. The SAR mode focusses and coherently sums all radar return echoes within the 2-s time window the target surface area is in the antenna beam. In principle the SAR processing improves along-track resolution. Land contamination has been a major concern for inland waters altimetry and SAR can reduce land interference. This paper shows that the physics of specular echoes from smooth inland waters leads to a very different approach which we call precise inland surface altimetry (PISA). PISA uses only echoes within the specular “flash” period, which is approximately the time the satellite nadir crosses over the water body. The processing is four orders of magnitude less than SAR. Land interference is negligible because specular water echoes are usually >50 dB greater than land. Sentinel-3 SRAL dataset on the salar de Uyuni (Bolivia) is used to evaluate PISA ranging precision. During inundation (wet months), echoes are at the theoretical maximum radar cross section (RCS), σ = 129 dBsm, and ranging precision is ~1 mm. In dry months the echoes are quasi-specular, with σ = 70–100 dBsm, and ranging precision is ~1 cm. The precision assessment is made with variate-differences, with pass-to-pass repeatability, and by comparison with GPS measurements. In addition to the salar de Uyuni analysis we gathered σ statistics on five millions Sentinel-3 SRAL Ku-band altimeter bursts (one burst = 64 contiguous echoes) from 52 passes of Sentinel-3A track 167 over South America. We illustrate specular and quasi-specular waters on lakes, a river, and a fjord. Ranging precision is similar to Uyuni, in the 1 mm-1 cm range. Water surface slopes of 1–3 cm/km are detected. We propose a simple rule-set to distinguish specular waveforms (σ >100 dBsm, sidelobes (with Hamming window) of −37 dB or lower) and quasi-specular (σ >70 dBsm, sidelobes lower than −20 dB), and non-specular (sidelobes> − 20 dB). PISA is appropriate to specular and quasi-specular echoes.