Low Signal-to-noise Ratio Values Research Articles

Robust acoustic reflector localization using a modified EM algorithm

In robotics, echolocation has been used to detect acoustic reflectors, e.g., walls, as it aids the robotic platform to navigate in darkness and also helps detect transparent surfaces. However, the transfer function or response of an acoustic system, e.g., loudspeakers/emitters, contributes to non-ideal behavior within the acoustic systems that can contribute to a phase lag due to propagation delay. This non-ideal response can hinder the performance of a time-of-arrival (TOA) estimator intended for acoustic reflector localization especially when the estimation of multiple reflections is required. In this paper, we, therefore, propose a robust expectation-maximization (EM) algorithm that takes into account the response of acoustic systems to enhance the TOA estimation accuracy when estimating multiple reflections when the robot is placed in a corner of a room. A non-ideal transfer function is built with two parameters, which are estimated recursively within the estimator. To test the proposed method, a hardware proof-of-concept setup was built with two different designs. The experimental results show that the proposed method could detect an acoustic reflector up to a distance of 1.6 m with 60%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$60\\%$$\\end{document} accuracy under the signal-to-noise ratio (SNR) of 0 dB. Compared to the state-of-the-art EM algorithm, our proposed method provides improved performance when estimating TOA by 10%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$10\\%$$\\end{document} under a low SNR value.

Novel Results on SNR Estimation for Bandlimited Optical Intensity Channels.

In a previous work of the author about non-data-aided estimation of the signal-to-noise ratio (SNR) for bandlimited optical intensity channels, a couple of limitations have been identified in terms of error performance and computational complexity. In the current paper, these deficiencies are avoided by the introduction of a second receiver filter with specific properties that is operated in parallel to the receiver filter normally used in this respect. Although not initially intended, the concept is also applied to data-aided SNR estimation by deriving a maximum likelihood algorithm and the Cramer-Rao lower bound (CRLB) as the theoretical limit of the error performance. In the next step, the dual-filter framework is used in the context of SNR estimation without knowledge about data symbols. The most significant benefit of this method is that the number of payload data employed for the estimation procedure might be selected arbitrarily long without impacting the spectral efficiency of the link. Since the computation of the true CRLB was out of scope due to complexity reasons, an asymptotic variant for very low SNR values is analyzed, which ends up in a closed-form solution. Furthermore, an algorithm based on first- and second-order moments of the samples at the dual-filter output is investigated, which turned out to be very attractive in terms of error performance and computational complexity.

Performance Improvement of CDMA Wireless Sensor Networks in Low SNR Channels Based on Raptor Codes

Low signal to noise ratio (SNR) channels require careful design of communication systems in order to operate reliably in such extreme channels. Special measures and techniques must be considered when designing communication systems for such environments to counteract the effect of high noise power. Channel coding is a powerful tool to aid against adverse channel conditions including low SNR. Coupling channel coding with a spreading technique can further strengthen the performance of communication systems. Hence, we studied the performance of a system that combines Raptor codes with direct sequence code division multiple access (DS-CDMA) in low SNR scenarios. The studied design can operate in a wide range of low SNR values. Simulation results show that the system can offer satisfactory BER performance at different code rates. Also, combining Raptor codes with DS-CDMA showed an improvement of approximately 6 dB when compared with other schemes using DS-CDMA in low SNR conditions.

Assessment of speech intelligibility during different teaching activities in classrooms with and without acoustic treatment

There is limited data for assessing speech intelligibility in real classrooms with realistic occupied noise levels and teacher’s and students’ speech levels for different teaching activities in Turkish secondary school classrooms. This study investigates the effect of reverberation time (RT), occupied noise levels and signal-to-noise ratio (SNR) on speech transmission index (STI) in real classrooms for instruction and group work. Noise levels were measured and STI values were calculated in two secondary school classrooms with RT of 0.88 s and 2.73 s. During instruction situation, calculated mean STI values were 0.63 (classroom with acoustic treatment) and 0.29 (classroom without acoustic treatment) for normal voice effort and 0.65 (classroom with acoustic treatment) and 0.39 (classroom without acoustic treatment) for raised voice effort. During group work, STI values ranged from 0.50 to 0.74 in classroom with acoustic treatment and from 0.01 to 0.34 in classroom without any acoustic treatment. SNR of 15 dBA is only approached in classroom with acoustic treatment during instruction situation. The effect of increasing SNR on STI is prominent in the classroom without acoustic treatment which already has low SNR values. The present results show that the classroom with acoustic treatment will have good speech intelligibility for instruction with normal vocal effort and for group work with raised vocal effort at 3 m distance. However, the classroom with no acoustic treatment will have STI values in the range of bad and poor intelligibility. The present results confirm the impact of SNR and RT on speech intelligibility and shows that RT value of minimum 0.8 s is optimal to have good speech intelligibility in a secondary school classroom of approximately 250 m3 with a capacity of 18 students.

Vehicle-to-roadside-unit-to-vehicle communication system under different amplify-and-forward relaying schemes

A vehicular Rayleigh fading communication system in which a vehicular transmitter (VT) is communicating with a vehicular receiver (VR) only through a dedicated stationary amplify-and-forward (AF) relaying roadside unit (RSU) is proposed in this paper (i.e., V2RSU2V system). To capture the effect of VT and VR motion, system's fading channels from VT to RSU and from RSU to VR are modeled to be time-varying in accordance with the first order autoregressive model. For the sake of performance comparison, the three common AF relaying sub-schemes are adopted at the RSU relay in separate scenarios; namely, the optimal channel-state-information-and-noise-assisted variable-gain (CSIN-VG) sub-scheme, the blind fixed-gain (B-FG) sub-scheme, and the semi-blind fixed-gain (SB-FG) sub-scheme. Under these AF relaying sub-schemes, closed-form expressions for outage probabilities are derived. The derived expressions are given in terms of variant vehicular system and channel parameters and can be directly considered for the special scenario of stationary VT and VR. Numerical and simulation results are reported to validate the analytical work and to reveal new insightful guidelines into vehicular systems outage performance. For example, results reveal that, for low average signal-to-noise-ratio (SNR) per symbol values, the SB-FG sub-scheme is superior in terms of outage performance for all vehicular scenarios and all threshold SNR values. As well, at high per-symbol average SNR values, results illustrate that the CSIN-VG sub-scheme is superior for high threshold SNR values, while the SB-FG is superior for low threshold SNR values. The study in this paper is timely due to the rapid advancements towards autonomous vehicles and smart transportation networks.

Low Complexity Deep Learning-Assisted Golden Code Sphere-Decoding with Sorted Detection Subsets

Golden code is a space-time block coding (STBC) scheme that has spatial multiplexing gain over the Alamouti STBC which is widely used in modern wireless communication standards. Golden code has not been widely adopted in modern wireless standards because of its inherent high detection complexity. However, detection algorithms like the sphere-decoding with sorted detection subsets (SD-SDS) have been developed to lower this detection complexity. Literature indicates that the SD-SDS algorithm has lower detection complexity relative to the traditional sphere-decoding (SD) algorithm, for all signal-to-noise ratio (SNR) values. The SD-SDS algorithm exhibits low detection complexity at high SNR; however, at low SNR the detection complexity is higher. We propose a deep neural network (DNN) aided SD-SDS algorithm (SD-SDS-DNN) that will lower the Golden code's SD-SDS low SNR detection complexity, whilst maintaining the bit-error-rate (BER) performance. The proposed SD-SDS-DNN is shown to achieve a 75% reduction in detection complexity relative to SD-SDS at low SNR values for 16-QAM, whilst maintaining the BER performance. For 64-QAM, the SD-SDS-DNN achieves 99% reduction in detection complexity relative to the SD-SDS at low SNR, whilst maintaining the BER performance. The SD-SDS-DNN has also shown to achieve low detection complexity comparable to that of the Alamouti linear maximum likelihood (ML) detector for a spectral efficiency of 8 bits/s/Hz. For a spectral efficiency of 12 bits/s/Hz, the SD-SDS-DNN achieves a detection complexity that is 90% lower than the Alamouti linear ML detector.

Deep Learning-Based Next-Generation Waveform for Multiuser VLC Systems.

Due to the growing number of users, power, and spectral effectiveness, most communication systems are complex and difficult to implement on a large scale. Artificial Intelligence (AI) has played an outstanding role in the implementation of theoretical systems in the real world, with less complexity achieving better results. In this direction, we compare the Non-Orthogonal Multiple Access (NOMA) technique for a multiuser Visible Light Communication (VLC) system with Successive Interference Cancellation (SIC) for two types of detectors: (1) the deep learning-based system and (2) the traditional maximum likelihood (ML) decoder-based system. For multiplexing, we compare the variations of novel Orbital Angular Momentum (OAM) multiplexing and Orthogonal Frequency Division Multiplexing (OFDM) with Index Modulation (IM). In this article, we implement OFDM-IM and OAM-IM for four users for the Gaussian fading MIMO Line-of-Sight (LoS) and Non-Line-of-Sight (NLoS) VLC channels. The suggested systems’ bit error rate (BER) performances are compared in simulations for a wide range of Signal-to-Noise Ratios (SNRs), which shows that deep learning-based systems outperform the ML-based system for both users to ensure better decoding at the receiver end, especially at higher SNR values. The detection error is lower in a deep learning-based system at around 20% and around 30% for low SNR and high SNR values, respectively.

Direction of Arrival Estimation Based on Received Signal Strength Using Two-Row Electronically Steerable Parasitic Array Radiator Antenna.

In this paper, we present a novel approach to direction-of-arrival (DoA) estimation using two-row electronically steerable parasitic array radiator (ESPAR) antenna which has 12 passive elements and allows for elevation and azimuth beam switching using a simple microcontroller, relying solely on received signal strength (RSS) values measured at the antenna output port. To this end, we thoroughly investigate all 18 available 3D antenna radiation patterns of the antenna measured in an anechoic chamber with respect to radiation coverage in the horizontal and vertical direction and propose a generalization of the power-pattern cross-correlation (PPCC) algorithm involving a high number of multiple calibration planes (MCP) as well as specific combinations of radiation pattern sets. Additionally, a new way of RSS-based DoA estimation accuracy assessment, which involves thorough testing conducted along the elevation direction when RF signals impinging on the antenna arrive from arbitrary θ angles, has been reported in this paper to verify the overall algorithm’s performance. The results obtained for different signal-to-noise ratio (SNR) levels indicate that two-row ESPAR antenna can produce, even for low SNR values, accurate DoA estimation in the horizontal plane without prior knowledge about the elevation direction of the unknown RF signals by using appropriate combinations of only 12 3D antenna radiation patterns.

MmWave V2V Localization in MU-MIMO Hybrid Beamforming

Recent trends for vehicular localization in millimetre-wave (mmWave) channels include employing a combination of parameters such as angle of arrival (AOA), angle of departure (AOD), and time of arrival (TOA) of the transmitted/received signals. These parameters are challenging to estimate, which along with the scattering and random nature of mmWave channels, and vehicle mobility lead to errors in localization. To circumvent these challenges, this paper proposes mmWave vehicular localization employing difference of arrival for time and frequency, with multiuser (MU) multiple-input-multiple-output (MIMO) hybrid beamforming; rather than relying on AOD/AOA/TOA estimates. The vehicular localization can exploit the number of vehicles present, as an increase in the number of vehicles reduces the Cramér-Rao bound (CRB) of error estimation. At 10 dB signal-to-noise ratio (SNR) both spatial multiplexing and beamforming result in comparable localization errors. At lower SNR values, spatial multiplexing leads to larger errors compared to beamforming due to formation of spurious peaks in the cross ambiguity function. Accuracy of the estimated parameters is improved by employing an extended Kalman filter leading to a root mean square (RMS) localization error of approximately 6.3 meters.

Increasing the Radar Equivalent Energy Potential by “Track Before Detect” Method

The problem of radar detection of small-sized targets using the traditional methods of selection of signals embedded in background noise is considered. It is shown that for a false alarm rate of 10–5, which provides for 1–2 false alarms within the entire coverage of the modern 3D radar, the probability of detection of a small-sized target is getting unacceptably low. Repeatedly decreasing the threshold can provide an acceptable level of the detection probability at ultra-low signal-tonoise ratio (SNR) values. At the same time, decreasing the threshold will result in an unacceptable increase of the false alarm rate. A new target detection procedure using the “track before detect” method (TBD) is proposed. In the TBD procedure, the target is considered detected when two conditions are met: the signal exceeds once a definite threshold; the target is detected within a strictly defined observation area (acquisition or tracking gate). For low SNR values in the range of 3–8 dB and equal false alarm rate, the detection probability increases by 20–50 % compared to the traditional detection method. The simulation results showed a strong dependence of efficacy of the TBD algorithm on the threshold value and the decision rule. The possibility is noted of adaptive control over the threshold due to the use the detection results in the preceding scanning cycles, as well as the introduction of matrix radar surveillance not only by the target coordinates and parameters, but also by the detection threshold, decision rules, etc. Examination of these issues is the subject of further research.

SNR‐dependent drone classification using convolutional neural networks

Radar sensing offers a method of achieving 24-h all-weather drone surveillance, but in order to be maximally effective, systems need to be able to discriminate between birds and drones. This work examines drone-bird classification performance as a function of signal to noise ratio (SNR). Classification at low SNR values is necessary in order to classify drones with a small radar cross-section (RCS), as well as to facilitate reliable classification at longer ranges. To investigate the relationship between classification performance and SNR, Gaussian noise is added to an experimentally obtained dataset of radar spectrograms. Classification is performed by convolutional neural networks (CNNs). It is shown that for the data available classification accuracy drops with falling SNR, as might be expected for any given CNN. The degree to which performance degrades with reduced SNR is presented. It is further shown that simpler network architectures are more robust to noise. Finally, it is demonstrated that data augmentation can be used as a means of enhancing classification accuracy at lower SNR values. Bayesian optimisation is used to find the optimal augmentation hyperparameters and overall, classification accuracies of 92% are achieved at low SNR.

Tiny golden angle ultrashort echo-time lung imaging in mice.

Imaging the lung parenchyma with MRI is particularly difficult in small animals due to the high respiratory and heart rates, and ultrashort T2* at high magnetic field strength caused by the high susceptibilities induced by the air-tissue interfaces. In this study, a 2D ultrashort echo-time (UTE) technique was combined with tiny golden angle (tyGA) ordering. Data were acquired continuously at 11.7T and retrospective center-of-k-space gating was applied to reconstruct respiratory multistage images. Lung (proton) density (fP ), T2*, signal-to-noise ratio (SNR), fractional ventilation (FV) and perfusion (f) were quantified, and the application to dynamic contrast agent (CA)-enhanced (DCE) qualitative perfusion assessment tested. The interobserver and intraobserver and interstudy reproducibility of the quantitative parameters were investigated. High-quality images of the lung parenchyma could be acquired in all animals. Over all lung regions a mean T2* of 0.20 ±0.05 ms was observed. FV resulted as 0.31 ±0.13, and a trend towards lower SNR values during inspiration (EX: SNR = 12.48 ±6.68, IN: SNR = 11.79 ±5.86) and a significant (P< 0.001) decrease in lung density (EX: fP =0.69 ±0.13, IN: fP =0.62 ±0.13) were observed. Quantitative perfusion results as 34.63 ±9.05 mL/cm3 /min (systole) and 32.77 ±8.55 mL/cm3 /min (diastole) on average. The CA dynamics could be assessed and, because of the continuous nature of the data acquisition, reconstructed at different temporal resolutions. Where a good to excellent interobserver reproducibility and an excellent intraobserver reproducibility resulted, the interstudy reproducibility was only fair to good. In conclusion, the combination of tiny golden angles with UTE (2D tyGA UTE) resulted in a reliable imaging technique for lung morphology and function in mice, providing uniform k-space coverage and thus low-artefact images of the lung parenchyma after gating.

Performance of Modulating Retro-Reflector FSO Communication Systems With Nonzero Boresight Pointing Error

A comprehensive investigation of the impact of nonzero boresight pointing error on the performance of modulating retro-reflector (MRR) free space optical (FSO) communication systems is the goal of this work. In particular, the end-to-end signal-to-noise ratio (SNR) statistical characteristics including probability density function (PDF), cumulative distribution function (CDF), moments as well as moment generating function (MGF) are formulated. Capitalizing on these results, the system average BER and outage probability are derived. Moreover, the asymptotic performances of the system ergodic capacity at low and high SNR values are deduced as well. The obtained expressions are verified via Monte-Carlo simulation.

Optimal 4-Ary Imbalanced-Phase-Amplitude Modulation in Uncorrelated and Correlated Receive Diversity PLC Systems Under Nakagami-$m$ Noise Environment

A receive diversity power line communication (PLC) system subject to correlated multipath channel fading and corrupted by additive Nakagami- <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$m$</tex-math></inline-formula> background noise is considered. The optimal receiver employing a 4-ary imbalanced-phase-amplitude modulation (I-PAM) scheme for data transmission is obtained, using which closed form expressions for the symbol error probability (SEP) in the union bound sense and the SEP in the asymptotic sense at high signal-to-noise ratio (SNR) are derived. The diversity order of the system is found to be independent of the shape parameter of the additive noise. Furthermore, the optimal 4-ary I-PAM constellation minimizing the SEP at lower SNR values is computed numerically. The optimal 4-ary I-PAM constellation is found to be elliptical in shape as opposed to the traditional circular QPSK constellation generally employed in wireless communication systems. The effects of the correlation among the multipath channel gains and the shape parameter of the noise on the optimality of the considered constellation are also presented via numerical studies.

The PRISMA imaging spectroscopy mission: overview and first performance analysis

The PRISMA satellite mission launched on March 22nd, 2019 is one of the latest spaceborne imaging spectroscopy mission for Earth Observation. The PRISMA satellite comprises a high-spectral resolution VNIR-SWIR imaging spectrometer and a panchromatic camera. In summer 2019, first operations during the commissioning phase were mainly devoted to acquisitions in specific areas for evaluating instrument functioning, in-flight performance, and mission data product accuracy. A field and airborne campaign was carried out over an agriculture area in Italy to collect in-situ multi-source spectroscopy measurements at different scales simultaneously with PRISMA. The spectral, radiometric and spatial performance of PRISMA Level 1 Top-Of-Atmosphere radiance (LTOA) product were analyzed. The in-situ surface reflectance measurements over different landcovers were propagated to LTOA using MODTRAN5 radiative transfer simulations and compared with satellite observations. Overall, this work offers a first quantitative evaluation about the PRISMA mission performance and imaging spectroscopy LTOA data product consistency. Our results show that the spectral smile is less than 5 nm, the average spectral resolution is 13 nm and 11 nm (VNIR and SWIR respectively) and it varies ±2 nm across track. The radiometric comparison between PRISMA and field/airborne spectroscopy shows a difference lower than 5% for NIR and SWIR, whereas it is included in the 2–7% range in the VIS. The estimated instrument signal to noise ratio (SNR) is ≈400–500 in the NIR and part of the SWIR (<1300 nm), lower SNR values were found at shorter (<700 nm) and longer wavelengths (>1600 nm). The VNIR-to-SWIR spatial co-registration error is below 8 m and the spatial resolution is 37.11 m and 38.38 m for VNIR and SWIR respectively. The results are in-line with the expectations and mission requirements and indicate that acquired images are suitable for further scientific applications. However, this first assessment is based on data from a rural area and this cannot be fully exhaustive. Further studies are needed to confirm the performance for other land cover types like snow, inland and coastal waters, deserts or urban areas.

Fast and accurate primary user detection with machine learning techniques for cognitive radio networks

&lt;p class="IJASEITAbtract"&gt;&lt;span&gt;Spectrum decision is an important and crucial task for the secondary user to avail the unlicensed spectrum for transmission. Managing the spectrum is an efficient one for spectrum sensing. Determining the primary user presence in the spectrum is an essential work for using the licensed spectrum of primary user. The information which lacks in managing the spectrum are the information about the primary user presence, accuracy in determining the existence of user in the spectrum, the cost for computation and difficult in finding the user in low signal-to noise ratio (SNR) values. The proposed system overcomes the above limitations. In the proposed system, the various techniques of machine learning like decision tree, support vector machines, naive bayes, ensemble based trees, nearest neighbour’s and logistic regression are used for testing the algorithm. As a first step, the spectrum sensing is done in two stages with Orthogonal Frequency Division Multiplexing and Energy Detection algorithm at the various values of SNR. The results generated from the above algorithm is used for database generation. Next, the different machine learning techniques are trained and compared for the results produced by different algorithms with the characteristics like speed, time taken for training and accuracy in prediction. The accuracy and finding the presence of the user in the spectrum at low SNR values are achieved by all the algorithms. The computation cost of the algorithm differs from each other. Among the tested techniques, k-nearest neighbour (KNN) algorithm produces the better performance in a minimized time.&lt;/span&gt;&lt;/p&gt;

Gravitational-Wave Burst Signals Denoising Based on the Adaptive Modification of the Intersection of Confidence Intervals Rule.

Gravitational-wave data (discovered first in 2015 by the Advanced LIGO interferometers and awarded by the Nobel Prize in 2017) are characterized by non-Gaussian and non-stationary noise. The ever-increasing amount of acquired data requires the development of efficient denoising algorithms that will enable the detection of gravitational-wave events embedded in low signal-to-noise-ratio (SNR) environments. In this paper, an algorithm based on the local polynomial approximation (LPA) combined with the relative intersection of confidence intervals (RICI) rule for the filter support selection is proposed to denoise the gravitational-wave burst signals from core collapse supernovae. The LPA-RICI denoising method’s performance is tested on three different burst signals, numerically generated and injected into the real-life noise data collected by the Advanced LIGO detector. The analysis of the experimental results obtained by several case studies (conducted at different signal source distances corresponding to the different SNR values) indicates that the LPA-RICI method efficiently removes the noise and simultaneously preserves the morphology of the gravitational-wave burst signals. The technique offers reliable denoising performance even at the very low SNR values. Moreover, the analysis shows that the LPA-RICI method outperforms the approach combining LPA and the original intersection of confidence intervals (ICI) rule, total-variation (TV) based method, the method based on the neighboring thresholding in the short-time Fourier transform (STFT) domain, and three wavelet-based denoising techniques by increasing the improvement in the SNR by up to 118.94% and the peak SNR by up to 138.52%, as well as by reducing the root mean squared error by up to 64.59%, the mean absolute error by up to 55.60%, and the maximum absolute error by up to 84.79%.

Inter-annual variability and skill of tropical rainfall and SST in APCC seasonal forecast models

The present study explored the performance of the current coupled models obtained from the Asia Pacific Economic Cooperation (APEC) Climate Centre (APCC) in representing the tropical Indo-Pacific sea surface temperature (SST) and rainfall during boreal summer season (June through September; JJAS). We have used the retrospective/hindcast runs for 28 years from 1983 to 2010 initialized in May. The mean SST bias in the tropical Indo-Pacific Oceans showed large diversity among the models in JJAS. In the case of the rainfall, most of the models displayed a strong dry bias over the major continental regions and wet biases over the tropical oceans. The majority of the models simulated the Inter-annual variability (IAV) of JJAS rainfall and SST reasonably well over the equatorial Pacific region, where the models are close to observed IAV and maximum signal to noise ratio (SNR). It is found that the models display, low IAV of rainfall and SST over the Indian Ocean with low SNR values, resulting in less predictive skill as compared to the tropical Pacific region. Similarly, all models showed a higher skill in summer rainfall prediction over the oceanic regions compared to the Asian land region, where SNR is very low. Further analysis suggested that the models have greater skill in predicting El Nino-Southern Oscillation (ENSO). The category wise analysis showed that models could predict 60–70% of the extreme ENSO events, but the normal events are represented only by 50%. It is noted that the models predict many false alarms for El Nino resulting in a higher frequency of El Nino occurrence. This is mainly responsible for stronger ENSO and the Asian Monsoon teleconnections in the models than in the observations. Meanwhile, the category wise rainfall skill for extended Indian monsoon region (EMR) displayed 50–60% accuracy for the extreme monsoon years and is around 50% for normal years. However, models such as CCSM3, CFSV2, and CANCM3 have displayed higher rainfall skills over EMR as compared to the other models possibly due to better representation of teleconnections spatial patterns between EMR rainfall and SST anomalies over Indo-Pacific Oceans.

Binary Time-Frequency Mask for Improved Malay Speech Intelligibility at Low SNR Condition

A binary masking is often seen as a medium to enhance speech signal especially in hearing aid and speech recognition applications due to its simplicity and efficient approach for supervised source separation. High intelligibility could be obtained by applying the binary time-frequency masking to noisy speech signal. Since the issue of linear filtering algorithms might affect the intelligibility of noisy speech signal, so, this paper presents the binary time-frequency mask for improved Malay speech intelligibility at low SNR condition. The clean Malay speech signals that were contaminated by car and train noise with different signal to noise ratio (SNR) and local criterion (LC) level in forming the binary mask are analysed. The performance of intelligibility improvement was evaluated using a normalized sub band enveloped correlation (nSec). Overall results showed that the proposed approach produces slightly improved speech intelligibility at low SNR value.

A variable step-size incremental LMS solution for low SNR applications

Many practical applications built around adaptive algorithms require solutions that work at low signal-to-noise ratio (SNR) values. In this situation, a variable step-size algorithm is a must to accommodate this scenario. A plethora of variable step algorithms exist in the literature for high SNR values. This paper presents a different variable step-size incremental least mean square strategy that provides excellent performance at low SNR values. A detailed theoretical analysis has been carried out for the proposed algorithm. Simulations have been performed for different scenarios to demonstrate the validity of the proposed strategy.