The Neural Network Model

2. Radius; increasing the radius will increase the inductance value and the Q of the device; somehow it will resonate at high frequency.

The neural network is composed of an input of 12 neurons presenting in the X values, an output layer with 2 neurons presenting in the Y values and hidden layer with 10

This antenna is designed on FR-4 substrate. It has dielectric constant 4.4 and thickness 1.6 mm. copper material is used for ground plane and patch. We provided microstrip line feeding to this antenna.

Among numerous choices, it is the most simple and result oriented approach. After selecting the resonant mode, it is the resonator’s dielectric constant value and its aspect ratio (a/h) that define the obtainable bandwidth. In simple geometric shapes such as cylindrical DRAs the smallest possible εr value makes Q-factor low which as a consequence results in wider bandwidth [34].

Based on Chapter 2, Neural Network Method (NN) will be chosen for voice-based command recognition method because it can handle bigger databased. For Neural Network to implement pattern recognition is quite common, and beneficial to use is backpropagation. Supervised learning that starts by inputting the training data through the network is a form of this method. When the data is put in the network, it will generate propagation output activations and then propagated backwards through the neural network, and generating a delta value for all hidden and output neuron. The weights of the network are then update by calculated delta values that generate by neural network, which increase the speech and quality of the learning process.

Table 2 illustrates that the second, third and fourth kernel parameter gives the highest accuracy. The performance of the classifier can be evaluated by using Accuracy. The equation (1) can be used to calculate the accuracy. Any of the kernel parameters can be used for further classification to

In the fig1 represents the number of nodes varying with respect to the delay as compared with MILP optimal formulation. It explained our proposed algorithm is better than the MILP formulation.

This algorithm was simulated with Matlab. These datasets and the mentioned characteristics are considered and the algorithm of each dataset with different slopes for the activation fumcion of interest were evaluated so that the best slope can be obtained. After running the program for several times and computing the average to obtain the best result, the optimum slope was evaluated for each dataset and the best slopes for Breast Cancer, Diabetes, Bupa, and

Window embedded wire antennas have been introduced at FM frequencies for low-power or receive-only applications. The power transmitted by the army radios at VHF and UHF frequencies can exceed tens of watts. At these power levels, it is of crucial importance to isolate the interior of the vehicle from the antenna radiation. In addition, this isolation eliminates the affects of the cabin interior and passengers on the antenna input impedance and radiation pattern. Hence a transparent ground plane will be needed behind the radiating element to reduce the radiation inside the vehicle and at the same time increase the directivity of the antenna. A wire-mesh or a more sophisticated band-selective ground plane can be used as the reflecting element. A meshed patch antenna on a mesh ground plane can provide the required isolation with the cabin interior. However because of the sub-wavelength dimensions of the armored windows at VHF and UHF frequencies a patch antenna, if feasible, cannot provide the one octave bandwidth required by the army radios. In this paper a monopole bow-tie antenna backed by a cavity (armored window) is proposed. Because of the limited thickness of the window glass, the radiation element and the transparent ground plane need to be placed very close to each other. Therefore the gain and/or the

RBF is relatively adaptive in nature, which is used for classification and pattern reorganization. The performance of the RBF neural network is compared to the commonly used multilayer network model and the classical logistic regression.RBF neural network is based on the supervised learning. This method works independently and based on many researches and is popular alternative to the traditional MLP (Multi Layer Preceptor).RBF neural networks works perfectly in modeling the non linear data and it can be trained in one stage rather than the iterative process as like

For the second part of the experiment, we created a capacitor by using two sheet of aluminum foil with the same area and a wax paper. We inserted the wax paper, which is the dialectic, between the aluminum foils to create a capacitor. Then we connected the capacitor to the RC circuit. In this part, we need to find the half time, the time constant, and capacitance in order to calculate for the thickness of the wax paper, or distance between the plates. In order to solve for the distance, we used the formula C=, which k is the dialectic constant, is the permittivity of free space, A is the area, and d is the distance between the plates. We rearranged the formula do solve for distance, so we used the formula, d=We measured the length and width of the plate, so the area is the multiplication of the length and width of the area plate. After the calculations, we obtain 580 micrometers for the thickness of the paper while the theoretical thickness is 35 micrometers. The percent difference between the theoretical and calculated thickness of the wax paper is 180%, which is significant to indicate that this method does not validate the

ANDREA ALU et al(2) The matching and radiation properties of subwavelength resonant patch antennas filled with double negative,double-positive and/or single-negative metamaterials blocks.These configurations may exhibit in principle an arbitrarily low resonant frequency for a fixed dimension,but they may not necessarily radiate efficiently when their size is electrically small.Realistic numerical simulation considering material dispersion,losesand presence of antenna feed are presented.The demand for compact radiators with sufficiently high gain is rapidly increasing in many application areas.Even through such antenna are very thin compared to operating

To construct an optimal hyperplane, SVM employs an iterative training algorithm, which is used to minimize an error function. According to the form of the error function, SVM models can be classified into four distinct groups:

For fulfilling all the requirement introduced above fractal microstrip patch antennas are designed.As the number of iteration increases the time consume for solving matrix generated in simulator based on method of moment(IE3D) increases. Due to this reason we are designing an artificial neural network for microstrip fractal antenna.

Fundamental characteristics of this antenna such as the radiation directivity and the frequency response are measured. Figure shows a picture of the proposed antenna that has a shape of ―tears drop‖. Using Novel Ultra Wideband Planar Reflector Antenna a reflector antenna relies on reflection of energy from the reflector. Reflector antennas tend to be good directional and high gain antennas [11]. There are some classes of ultra wideband reflector antennas, such as planar, corner, and parabolic cylinder reflector antennas.