A5081116719- graph theory and CDMA systems
- Optical Network Technologies
- Coding theory and cryptography
- Photonic and Optical Devices
- Advanced Photonic Communication Systems
- Terahertz technology and applications
- Graph Labeling and Dimension Problems
- Photonic Crystal and Fiber Optics
- Cloud Computing and Resource Management
- Advanced Optical Network Technologies
- Optical Wireless Communication Technologies
- Data Stream Mining Techniques
- Advanced Fiber Optic Sensors
- IoT and Edge/Fog Computing
- Photonic Crystals and Applications
- Motor Control and Adaptation
- Water Quality Monitoring Technologies
- Direction-of-Arrival Estimation Techniques
- Muscle activation and electromyography studies
- Advanced Malware Detection Techniques
- Speech and Audio Processing
- Network Security and Intrusion Detection
- Artificial Intelligence in Healthcare
- Wireless Communication Networks Research
- Smart Grid Security and Resilience
The University of Sydney
2024-2025
Universiti Malaysia Perlis
2009-2022
Massey University
2016-2022
Rajshahi University of Engineering and Technology
2014-2021
Tata Consultancy Services (India)
2017
University of Southern California
2010-2016
Jet Propulsion Laboratory
2013
Orbital angular momentum (OAM), which describes the "phase twist" (helical phase pattern) of light beams, has recently gained interest due to its potential applications in many diverse areas. Particularly promising is use OAM for optical communications since: (i) coaxially propagating beams with different azimuthal states are mutually orthogonal, (ii) inter-beam crosstalk can be minimized, and (iii) efficiently multiplexed demultiplexed. As a result, multiple could used as carriers...
We simulate and analyze the mode properties propagation effects of orbital angular momentum (OAM) modes in a ring fiber. A fiber with 0.05 up-doping is designed simulation to support up 10 OAM while maintaining single-mode condition radially. With multiple-ring fiber, tens can be potentially multiplexed greatly enhance system capacity spectral efficiency. The index difference maintained above <sup xmlns:mml="http://www.w3.org/1998/Math/MathML"...
Abstract Big Data analytics for storing, processing, and analyzing large-scale datasets has become an essential tool the industry. The advent of distributed computing frameworks such as Hadoop Spark offers efficient solutions to analyze vast amounts data. Due application programming interface (API) availability its performance, becomes very popular, even more popular than MapReduce framework. Both these have 150 parameters, combination parameters a massive impact on cluster performance....
Software fault prediction and proneness has long been considered as a critical issue for the tech industry software professionals. In traditional techniques, it requires previous experience of faults or faulty module while detecting inside an application. An automated recovery models enable to significantly predict recover using machine learning techniques. Such ability feature makes run more effectively reduce faults, time cost. this paper, we proposed defect predictive development...
Optical code division multiple access (OCDMA) provides another dimension to systems, in which each user is assigned a unique code. This allows subscriber simultaneously the medium without any contention. However, simultaneous of users introduce interference (MAI) primarily deteriorates performance OCDMA systems. paper proposes new called diagonal double weight (DDW) elevate and cardinality spectral amplitude coding (SAC) Performance our proposed evaluated using comprehensive analytical...
Abstract Due to the rapid growth of available data, various platforms offer parallel infrastructure that efficiently processes big data. One critical issues is how use these optimise resources, and for this reason, performance prediction has been an important topic in last few years. There are two main approaches problem predicting performance. fit data into equation based on a analytical models. The other machine learning (ML) form regression algorithms. In paper, we have investigated...
This study proposes a dielectric terahertz (THz) porous core fibre with ultra‐high birefringence and near zero dispersion‐flattened properties. The finite element method is used to design analyse properties of the proposed THz fibre. crystal has triangular lattice microstructured circular air holes in outer cladding elliptical core. exhibits high 7 × 10 −2 low effective material loss 0.1 cm −1 at operating frequency f = 1 THz. It also shows nearly flattened dispersion an absolute variation...
A near zero dispersion flattened porous core photonic crystal fibre having ultra‐flat low material loss suitable for terahertz (THz) transmission is represented. novel type of geometrical structure air hole position has been proposed in this design. Numerical analysis exhibits effective (EML) 0.03746 cm −1 at 1.0 THz operating frequency and a flat EML 0.0398 from 1.5 to 5.0 with absolute coefficient ±0.000416 . The waveguide shows 0.4 ± 0.042 ps/THz/cm the range 1.25 THz.
Abstract This article proposes a new parallel performance model for different workloads of Spark Big Data applications running on Hadoop clusters. The proposed can predict the runtime generic as function number executors, without necessarily knowing how algorithms were implemented. For certain problem size, it is shown that based serial boundaries 2D arrangement executors fit empirical data various workloads. was obtained from real cluster, using and HiBench. used in this work included...
Breast Cancer is the second most important cause of death among women. As per clinical expert, breast cancer one prominent cancers after lung cancer. However, early detection this type in its initial stage helps to save lifes and increases lifespan. The survival chance a patient can increase if there classifier that with quick prediction Therefore, smart framework required effectively detect predict high accuracy In article, six machine learning classification algorithms, namely Logistic...
Currently, solar energy has turned into a popular alternative source to meet certain demands around the world due instability of oil and coal prices with global warming issues. The aim this paper is develop simple cost-effective system for rural areas where grid electricity not available. To fulfil objective, 5-Watt PV (photovoltaic) stand-alone module was used as power common type lead acid battery (12V, 7AH) applied backup system. panel connected via charge controller which responsible...
Water is an essential resource in the world and currently household drinking water important asset to save life due shortage of earth. This paper describes design development process automatic control system for tap using floatless level sensor which can wastage without presence any operator physically. sensing utilizes a combination solenoid valve, electromagnetic relay (a type electrical switch), controller (61F) electrodes (used sense watery electrically). The developed automatically...
This paper presents the design and development process of Wireless Data Acquisition System (WiDAS) which is a multi-sensor system for water level monitoring. It also consists microcontroller (ATMega8L), data display device an ultrasonic distance sensor (Parallax Ping). wireless based acquisition can communicate through RF module (Tx-Rx) from measurement sources, such as sensors devices digital or analog values over period time. The developed has option to store in computer memory. was tested...
A hybrid optical code division multiple access (OCDMA) over wavelength multiplexing (WDM) system is proposed for supporting large numbers of subscriber fiber-to-the-home (FTTH) network. The combines two different techniques, namely OCDMA and WDM in order to support many subscribers simultaneously. In this study, the modified double weight (MDW) used as signature address designing because can accommodate more number simultaneous active users under standard acceptable bit-error-rate (e.g. ≤ 10...
In this paper we propose a system for unobtrusive automated indoor surveillance of subjects in environment using the Kinect sensor. We demonstrate that features identity, location and activity person can be detected with considerable accuracy system. Further, show how existing design patterns used to create data parallel scalable architecture such real-time.
In this paper a new detection technique known as NAND subtraction scheme based on Modified Double Weight (MDW) code for spectral-amplitude coding OCDMA (SAC-OCDMA) is proposed. The MDW the modified version of double weight (DW) family. theoretical and simulation results show that can have more better BER performance than other conventional techniques such AND Complimentary techniques. system was characterized by signal to noise ratio (SNR), bit error rate (BER), different Psr (-10 dBm). This...
We demonstrate crosstalk degradation mitigation using 4×4 MIMO equalization on an orbital-angular-momentum-multiplexed free-space data link with heterodyne detection. 20-Gbit/s QPSK signal up to -4.8 dB is recovered and the BER reduced by two orders of magnitude.
In this paper, we have extensively analyzed the SAC-OCDMA system performance using Modified Double Weight (MDW) code with NAND detection technique as newly developed by authors. theoretical analysis, various noises and Multiple-Access Interference (MAI) effect were taken into account. The was characterized signal-to-noise ratio (SNR) Bit-Error-Rate (BER). analysis results obtained compared those complementary subtraction techniques. comparison revealed that MDW can support more number of...
We propose a practical Heart Rate Estimation algorithm utilizing wrist-based photoplethysmography (PPG) signals for continuous health monitoring of crane workers who spend long hours in an isolated cabin the harsh factory environment. Our novelty lies devising low footprint that can reliably estimate presence motion artefact as well offers feasibility deploying on wearable platform. More particularly, our solution addresses two fundamental issues: a) correcting weak wrist PPG signal from...