Darmawaty Mohd Ali

Diabetes mellitus forecast using artificial neural network (ANN)

The diabetes mellitus forecasting using ANN is presented in this paper. The result of this study will provide solutions to the medical staff in determining whether someone is the diabetes sufferer or not which is much easier rather than currently doing a blood test. The back-propagation algorithm has been chosen for learning and testing of 768 data whereby 268 of them are diagnosed with diabetes. Inputs to the network are number of times pregnant, plasma glucose concentration, blood pressure, triceps skin fold thickness, serum insulin, body mass index, diabetes pedigree function and age. The network with eight inputs and four inputs are then tested and results obtained are compared in terms of error. The outcome of this study is whether someone is the diabetes sufferer or not. Accurate results have been obtained which proves the effectiveness of the proposed ANN.

The Effect of Additive White Gaussian Noise and Multipath Rayleigh Fading On BER Statistic in Digital Cellular Network

This paper highlights the simulation using cyclic codes in order to identify the effect of two types of noise which are additive white Gaussian noise and multipath Rayleigh fading to a communication system. The main objective is to know which communication channel is better, a communication channel with AWGN or a communication channel with multipath Rayleigh fading. Modulation technique used in this simulation is quadrature phase shift keying (QPSK). Noise reduction and information message recovery are the important aspect in designing a communication system instead of speed and accuracy in transmitting the information. Matlab Version 7.1 is used to simulate the system. The model consists of transmitter, transmission channel and receiver. The messages are encoded and decoded using cyclic codes. The system also provides error detection and correction using forward error correction (FEC)

Performance measurement of VoIP over WiMAX 4G network

Worldwide Interoperability for Microwave Access or better known as WiMAX is a technology based on IEEE 802.16 standard with guaranteed QoS which provides high speed data rates. In this paper, we analyze the performance of VoIP over the commercial WiMAX networks through traffic measurements. The measurements focused on two QoS elements, jitter and latency that greatly affect the quality of the network. These parameters were measured using JPerf software and the values were compared with the Standard Quality Management scale by ITU-T P.862. A comparison between VoIP in WiMAX and broadband 3G are also included.

Performance study of the WiMAX uplink scheduler

The IEEE 802.16/802.16e provides powerful tools in order to achieve different QoS constraints [1]. The MAC layer defined five scheduling service types to provide QoS differentiation for different types of applications. In this paper, we investigate the performance of the scheduling schemes deployed in the OPNET modeler by considering VoIP, video conferencing and non-real-time services. We further investigate the scheduling algorithm with respect to the end-to-end delay (ETE) when the subframe ratio is set to 2∶1 and 1∶1. Simulation results revealed that higher downlink subframe ratio should be employed for a data-only (non-real time) network while 1∶1 subframe ratio is more appropriate in supporting real-time traffic network. A performance analysis and discussion are provided.

Reduction of interference with linear equalizer using quarter subcarrier mapping scheme

MIMO-OFDM is known for its novelty and capability for higher data rate transmission and has been an effective multicarrier transmission technique over frequency selective channels. However, its vulnerability to frequency offset (FO), has wiped out the orthogonality of the subcarriers and lead to inter-carrier interference (ICI) which degrades the system performance. The system performance will also be affected by the Inter-symbol interference (ISI) due to severe frequency selective fading (FSF) from a time delay at the receiver induced from multipath channel between transmitter and receiver. In order to overcome ICI and ISI a new subcarrier mapping (quarter) intercarrier interference self-cancellation (ICI-SC) scheme STFBC with an equalization technique is appended at the receiver to filter and compensate the ICI and ISI effects. The objective of this paper is to evaluate Pair-wise error probability (PEP) performance using a new quarter method subcarrier mapping ICI-SC scheme STFBC MIMO OFDM with added linear equalizer at the receiver for ICI and ISI reduction. The result indicates that the Pair-wise error probability (PEP) has improved better over STFBC MIMO-OFDM system by using quarter subcarrier mapping scheme (ICI-SC) with equalizer compared to existing available subcarrier mapping scheme, and efficient bandwidth with maximum diversity can be achieved.

A Queueing Analysis of Polling Based for Mobile Multihop Relay WiMAX Networks

By employing a relay station (RS) on the WiMAX networks, it offers an extended coverage and improved capacity over wireless broadband. However the mobile station (MS) such that connected to the RS has effect the system performance due to increase of the MS numbers. In WiMAX the base station (BS) adopts the centralized-based polling mode where the best way to guarantee the delay bound is by polling in every frame. Through the polling approach, the BS allocated dedicated or shared resources periodically to each MS, which can be used to request resources. Therefore in centralized-based polling MAC a few parameters become important factor to determine the performance of IEEE802.16j. This paper analyze the queue based polling for multi-hop relay in WIMAX networks in order to minimize the average polling delay while increase network throughput, based on tandem queue analysis. The product-form solution applied is valid since the models related to M/M/1. The numerical results showed that numbers of poll is affect by analyzed under various number of MS.

Mitigate the GPS ionospheric scintillation during solar flare

This paper to mitigate the Ionospheric Scintillation effect during solar flare happen on 7th June 2011. This scintillation data recorded by dual-frequency GPS Receiver, GPS ionospheric scintillation and TEC monitored (GISTM) at UKM, Malaysia (2.55°N, 101.46°E). Parameters on this study will focus on the amplitude scintillation in between 18:00 LT until 00:00 LT; 06:00 pm to 12:00 am. Observation showed that after the post sunset until midnight, the scintillation index in strong scintillation level, S4 ≥ 0.4.

Nighttime D-region ionosphere characteristics from tweek atmospherics observed in the North America Region

Tweek atmospherics are ELF/VLF pulse signals with frequency dispersion characteristics generate from lightning discharges and spread in the Earth-ionosphere waveguide mode. This study uses the electromagnetic wave approach describe as ELF (extremely low frequency) and VLF (very low frequency) which propagates sub-ionospherically among the earth and the lowest ionosphere. This paper presents the nighttime D-region ionosphere characteristics of tweeks at Gakona station (62.71° N, 143.99°W) in the North America Region. By using the fundamental of cut-off frequency, we estimated the nighttime variations of the tweek reflection height of the D-region ionosphere, equivalent electron density and the distance travel by tweek along its propagation. Our observation indicated that the equivalent electron densities for tweeks varies from 23-27 el/cm3 in the altitude range of 77 to 91 km and demonstrated that these signals is capable to travel up to 5600 km over a wide area in mid-latitude region. Comparison in term of the ionospheric parameters at different latitude shows that the reflection height and electron density in this study is almost consistent for all latitude tweek measurements.

Performance measurement of LTE along light rapid transit (LRT) railway track of Kelana Jaya line

This paper presents the results of measurements of LTE signal for commercial mobile devices on board of a train with the velocity of 80 to 100 km/h along the railway track. This study aims to measure the quality of the cellular LTE coverage from Kelana Jaya to Gombak considering the fact of the growing demand for better quality of service from the subscriber. The measurement is performed by conducting a drive test at the frequency range of 1800 MHz in a commercial network where the results are obtained using the G-Net Track. The measurement results present that the quality of the LTE network has not yet reached the standard specified by 3GPP standard in terms of RSRP, SINR, and data rate. The operators in Malaysia can reap the benefits from the measurement results provided in the study for future LTE deployments along the railway track.

Performance analysis of downlink scheduling algorithms in the rural and urban environments in LTE

Long Term Evolution (LTE) is the emerging technology that provides higher data rates and mobile Internet access. The higher demand from the users nowadays causes the performance of the three channel-aware scheduling algorithms to be explored. In this paper, the performance of Maximum-Largest Weighted Delay First (M-LWDF), Exponential/Proportional Fair (EX/PF) and Proportional Fair (PF) packet scheduling has been investigated for the downlink transmission. The performance analysis takes place in 2 different scenarios, rural, and urban in which propagation loss coexist. The evaluation is in single macro cell with interference. The simulation results shows variation in terms of performance metrics like packet loss ratio (PLR), throughput and delay for best effort, VoIP and video. In both rural and urban scenarios, M-LWDF outperform other algorithm in real-time services while PF algorithm is the best for non-real time services.