Recognition of Bangla Handwritten Number Using Combination of PCA and FIS with the Aid of DWT

The structure of any Bangla numerical character is more complex compared to English numerical character. Two pairs of numerical character in Bangla resembles to be closed and they are: “one and nine” and “five and six”. We found that, handwritten Bangla numerical character cannot be recognized using single machine learning algorithm or discrete wavelet transform (DWT). Above phenomenon motivated us to use combination of DWT, Fuzzy Inference System (FIS) and Principal Component Analysis (PCA) to recognize numerical characters of Bangla in handwritten format. The four lowest spectral components of a preprocessed image are taken using DWT, which is considered as the feature vector to recognize the digits in first phase. The feature vector is then applied to FIS and PCA separately. The combined method provides recognition accuracy of 95.8% whereas application of individual method gives less rate of accuracy. Instead of storing the images itself in a folder, if we can store the feature vector of images achieved from DWT in tabular form. The records of table can be applied in FIS, PCA or other object detection algorithm. Although the technique used in the paper can detect objects with moderate rate of accuracy but can save huge storage against a benchmark database of images. If a tradeoff is made between storage requirements and accuracy of recognition, the model of the paper is preferable compared to other present state-of-art. Another finding of the paper is that, the spectral components of images acquired by DWT only matched with FIS and PCA for classification but do not match properly with unsupervised (K-mean clustering) and supervised (support vector machine) learning.

Traffic Modelling of Low Dense Femtocellular Network for Long Term Evolution

One of the features of Long Term Evolution (LTE) is the deployment of femtocells as the underlain cell of a macrocell without any intervention of frequency planning to offload the traffic. In this paper we used Markov chain to derive the expression of blocking probability for both macro and femtocell in terms of traffic parameters of the network. We developed an analytical model to find the expression of probability of forced termination (FT) using combination of mobility model and probability tree considering low dense femtocellular network. Two different trees were designed: a newly originating call which starts its session in a femtocell and that of in a macro cell. The link parameters of small scale fading of wireless network under Multiple Input Multiple Output (MIMO) are combined with the proposed traffic model to get the probability of FT of a real-life network. A new state transition chain was also developed including its solution for LTE traffic of variable bandwidth (BW) and a comparison was made with Erlang’s traffic model.

Electrical Load Forecasting Using Fuzzy System

To cope with the demand and supply of electrical load of an interconnected power system of a country, we need to forecast its demand in advance. In this paper, we use a fuzzy system to forecast electrical load on short-term basis. Here, we consider temperature, humidity, seasons of a year and time segments of a day as the parameters, which govern the demand of electrical load. For each of the parameter, we use several membership functions (MFs) and then apply the Mamdani rule on MFs and the output is determined by using the centroid method. Finally, the surface plot reveals the real scenario of the load demand. The difference between actual load and the output of the fuzzy system is found as +1.65% to -13.76%. The concept of the paper can be applied in interconnected power system of Bangladesh to reduce power loss, especially when generation is higher than the demand.

Determination of Array Gain of Single Hop to Achieve the Performance of a 2-Hop Wireless Link

In adaptive beamforming system adaptive algorithm of digital filter is applied to update the weighting vector of the antenna elements to get antenna gain along the desired direction and attenuation along the jammer. The objective of the paper is to evaluate the threshold gain of the adaptive beam former along the line of sight (LOS) between the transmitter and the receiver (including jammer suppression) to make the single hop link comparable with 2-hop link. The single hop and 2-hop communication systems are compared in context of symbol error rate (SER) under fading condition theoretically and verified by simulation. Finally we evaluate the numerical value of threshold gain of adaptive beamformer of two antenna elements under Rayleigh and Nakagami-m fading conditions.

Performance Analysis of Optical Wireless Communication System Employing Neuro-Fuzzy Based Spot-Diffusing Techniques

The spot-diffusing technique provides better performance compared to conventional diffuse system for indoor optical-wireless communication (OWC) system. In this paper, the performance of an OW spot-diffusing communication system using Neuro-Fuzzy (NF) adaptive multi-beam transmitter configuration has been proposed. The multi-beam transmitter generates multiple spots pointed in different directions, hence, forming a matrix of diffusing spots based on position of the receiver and receiver mobility. Regardless of the position of the transmitter and receiver, NF controller target the spots adaptively at the best locations and allocates optimal power to the spots and beam angle are adapted in order to achieve better signal-to-noise plus interference ratio (SNIR). Maximum ratio combining (MRC) is used in the imaging receiver. The proposed OW spot-diffusing communication system is compared with other spot-beam diffusion methods proposed in literature. Performance evaluation revels that the proposed NF based OW spot-diffusing communication system outperforms other spot-beam diffusion methods.

Performance Analysis of Multi-Hop Wireless Link under Maximum Flow Algorithm

To enhance link capacity of a wireless link one or more repeater is used between the sender and the receiver. Recent literature deals with multiple parallel links to enhance throughput instead of conventional single path. In case of a multidirectional and multi-hop wireless network, the selection of link of maximum signal to noise ratio (SNR) does not guarantee the maximum throughput. In this paper, we use augmenting path of Ford-Fulkerson algorithm in detection of maximum flow from sender to receiver. To reduce the process time at the sending node, minimum-cut theorem is used to determine maximum flow like power flow of previous work. Using the maximum flow algorithm, we obtain the capacity of multi-hop wireless link higher than the conventional theorem. The concept of the paper is applicable in MANET (Mobile Ad-hoc Network), WSN (Wireless Sensor Network) and CRN (Cognitive Radio Network).

Bangla Handwritten Character Recognition Using Extended Convolutional Neural Network

The necessity of recognizing handwritten characters is increasing day by day because of its various applications. The objective of this paper is to provide a sophisticated, effective and efficient way to recognize and classify Bangla handwritten characters. Here an extended convolutional neural network (CNN) model has been proposed to recognize Bangla handwritten characters. Our CNN model is tested on “BanglalLekha-Isolated” dataset where there are 10 classes for digits, 11 classes for vowels and 39 classes for consonants. Our model shows accuracy of recognition as: 99.50% for Bangla digits, 93.18% for vowels, 90.00% for consonants and 92.25% for combined classes.

Data Prediction Model Using Combination of Clustering and Fuzzy Technique

The analysis of environmental daily evaporation plays a vital role in the field of agriculture. It is very essential to know the daily evaporation rate of a particular area for proper cultivation. So, we need a standard prediction model which can predict the daily evaporation. In this paper, we use subtractive clustering and Fuzzy logic to predict daily evaporation of a particular area. The input data used in the paper are: maximum soil temperature, average soil temperature, average air temperature, minimum relative humidity, average relative humidity and total wind, which are related to the daily evaporation of a particular area as the output. The accuracy of output of the paper is compared with the previous model of Artificial Neural Network (ANN) and we get better result towards the target value. The finding of the paper is applicable in environmental science, geological science and agriculture.

Data Classification Using Combination of Five Machine Learning Techniques

Data clustering plays a vital role in object identification. In real life we mainly use the concept in biometric identification and object detection. In this paper we use Fuzzy Weighted Rules, Fuzzy Inference System (FIS), Fuzzy C-Mean clustering (FCM), Support Vector Machine (SVM) and Artificial Neural Network (ANN) to distinguish three types of Iris data called Iris-Setosa, Iris-Versicolor and Iris-Virginica. Each class in the data table is identified by four-dimensional vector, where vectors are used as the input variable called: Sepal Length (SL), Sepal Width (SW), Petal Length (PL) and Petal Width (PW). The combination of five machine learning methods provides above 98% accuracy of class identification.

Recovery of Image through Alamouti Channel with Incorporation of RSA Algorithm

In many applications, it is necessary to transmit images at a remote station, where wired Internet service is not available. In this case, wireless local loop (WLL) can help in making wireless link between one end node of the internet and remote service center. In such link, the communication is heavily affected by large and small scale fading;hence the received signal experiences huge distortion in case of forward error correction. Otherwise, huge service delay arises due to frequent negative acknowledgements. To combat the situation, we can choose Alamouti channel of full rate and fully orthogonal space-time block code (OSTBC). Our aim is to transmit images through Alamouti channel and to observe the quality of the recovered image, in context of bit error rate (BER). We have also observed the impact of fading and additive white Gaussian noise (AWGN) on the image without application of error correction or detection technique of channel coding. To ensure security, we apply the RSA algorithm on each pixel prior transmitting and decrypt them at the receiving end, where we found no impairment from the algorithm. Finally, we observe that the relative performance of the system changes digital modulation schemes.