Sheikh Md. Rabiul Islam

Design of a programmable digital IIR filter based on FPGA

FPGAs are increasingly being promoted in signal processing applications with their tractability, parallelism, high speed, and fast time-to-market. Digital filter is one of the important contents of digital signal process. The characteristic of frequency selection in lower order in comparison with FIR, IIR digital filter is widely applied in modern signal processing systems. Hardware description languages such as Verilog differ from software programming languages because their include ways of describing the propagation of time and signal dependencies (sensitivity). In this paper, architecture of a programmable digital IIR filter has been proposed based XILINX FPGA board. In this architecture gate level design has been used to analyze the impulse response of the IIR filter.

A wireless video streaming system based on OFDMA with multi-layer H.264 coding and adaptive radio resource allocation

With the advancement of video-compression technology and the wide deployment of wireless networks, there is an increasing demand for wireless video communication services, and many design challenges remain to be overcome. In this work, an adaptive framework is proposed and simulated for multi-user video transmission over OFDMA-based communication system for the efficient streaming of multi-layered H.264 video over wireless channels. The aim of the proposed method is to combine adaptive sub-carrier allocation and bit loading with the transmission of the H.264 SVC(Scalable Video Coding ) encoded video sequences in order to increase the number of supported users in the system and provide the best quality of service(QoS) to the sub-carriers. The scalable extension H.264/MPEG4-AVC video codec is used to obtain two different bitstreams: a base-layer (carrying the most critical information content at low bitrate), and an enhancement layer, characterized by increasing bitrate and decreasing importance of video content. A prior study of the video quality is performed with a theoretical model, as the quality of the received video will depend on the errors introduced by the wireless channel. The required BER is depends on the video quality, which is the most crucial parameter to determine. The bandwidth is distributed among the orthogonal sub-carriers and the best suited sub-carriers is given to the base layers to achieve the required BER. The paper proposes an overview of the system and tests its performance through extensive simulations. The proposed framework significantly improves the video transmission quality with the maximum use of radio resources.

Biomedic Signal Processing and Analysis of N euroimaging from fNIRS for Human pain

One of major biomedical signals, pain, and its diagnosis has been critical but hard in clinical practice, in particularly for nonverbal patients. However, as we know that neuroimaging methods, such as functional near-infrared spectroscopy (fNIRS), have shown some great encouraging assessing neuronal function corresponding to nociception and pain. Specially some research results strongly suggest that neuroimaging, together with supports from machine learning, may be practically used to not only facilitate but also can predict different cognitive tasks over this challenge. The aim of this current research is to expand our previous studies by exploring the classification of fNIRS signals (oxyhaemoglobin) according to temperature level (we define cold and hot) and corresponding pain intensity (say low and high) by means of machine learning models. In order to find out the relations between temperatures and pain intensity, we defined and used the quantitative sensory testing to determine pain threshold and pain tolerance for the cold and heat in all eighteen-healthy people. The classification algorithm is based on a bag-of-words approach, a histogram representation was used in document classification based on the frequencies of extracted words and adapted for time series. Two machine learning algorithms were used separately, namely, K-nearest neighbor (K-NN) and support vector machines (SVM). A comparison between two sets of fNIRS channels was made in our classification task. The results showed that K-NN obtained slightly better results (92.1%) than SVM (91.3%) with all the 24 channels; however, the performances slightly dropped if using only channels from the region of interest with K-NN (91.5%) and SVM (90.8%). These research results encourage potential applications of fNIRS in the development of a physiologically based diagnosis of human pain, including in clinical parties.

A Computational Investigation of an Active Region in Brain Network Based on Stimulations with Near-Infrared Spectroscopy

Near-infrared spectroscopy (NIRS) has been widely used in medical imaging to observe oxygenation and hemodynamic responses in the cerebral cortex. In this paper, the major target is reporting our current study about the computational investigation of functional near infrared spectroscopy (fNIRS) in the somatosensory region with noxious stimulations. Based on signal processing technologies within communication network, the related technologies are applied, including cross correlation analysis, optic flow, and wavelet. The visual analysis exposed pain-related activations in the primary somatosensory cortex (S1) after stimulation which is consistent with similar studies, but the cross correlation results strongly evidenced dominant channels on both cerebral hemispheres. Our investigation also demonstrated that the spatial distribution of the cortical activity origin can be described by the hemodynamic responses in the cerebral cortex after evoked stimulation using near infrared spectroscopy. The current outcomes of this computational investigation explore that it is good potential to be employed to deal with pain assessment in human subjects.

Design and implementation of low cost ECG monitoring system for the patient using smart device

Cardiovascular disease (CSD) has become the leading cause of death worldwide in recent years. This CSD is the most challenging problem for detection or identification in early stages of patients. This research work approaches to develop an ECG signal generator at very low cost for the patients who can receive his/her ECG signal and detect the probability of cardiovascular diseases instantly. This ECG signal is transmitted via Bluetooth/Wi-Fi/Zigbee module to smart device with support software simulation where feature extraction and detection algorithm is setup for cardiovascular disease. This network can be connected with the doctors and hospitals to get the fastest treatment. In this paper, we have also proposed extraction and detection algorithm for detecting of CSD. This proposed idea is to contribute to bring under control heart diseases and also act as an expected results in health care service to patients in remote area.

ECG signal for artrial fibrillation detection

Atrial Fibrillation (AF) is one of the most common cardiac arrhythmias. The number of patients related to heart failure due to AF is increasing day by day. Early detection of AF may reduce the risk of death due to heart failure. So, it has become more important to detect AF. There are various methods to detect AF. In this paper, we use ECG signal for AF detection based on RR interval. The MIT-BIH Atrial Fibrillation database is used to import ECG data for analysis. We use the algorithm that mainly follows statistical method for detection of AF. Parametric statistic RMSSD and SE, and non-parametric statistic, TPR are used for this purpose. The threshold value of RMSSD divided by Mean RR is 0.1, SE is 0.7 and TPR is greater than 0.54 and less than 0.77 is considered for AF detection. The resultant value of RMSSD, SE and TPR of every beat is checked weather it crosses the threshold level or not. If all the three parameters cross the threshold level then the beat flagged as AF. The result is compared with the annotations of the database and the sensitivity, specificity and accuracy is determined. The algorithm has the sensitivity of 98%, specificity of 98% and accuracy of 95%. The result obtained in this study is appreciable compared to the other study found in the literature.

Feature extraction and classification of EEG signal for different brain control machine

Brain computer interface is used for human and machine learning analysis. This paper represents the EEG datasets that are built with different cognitive task such as left, right, back and front imaginary movement with eye open. We have used different feature extraction method to classify these EEG signal using Support Vector Machine (SVM), k-Nearest Neighbor (k-NN) and Artificial Neural Network (ANN). All these methods are compared with other work that have done with other datasets. The proposed work is obtained 95.21% accuracy 98.95% sensitivity for SVM and k-NN is 90.88% and ANN is 94.31%. The performance results have shown higher enough than all others.