Research Journal of Engineering and Technology

Antim Bala Sharma ¹, T. Vigneswaran ², Ajay Sharma ³, Fanibhushan Sharma ⁴, Nirmala Sharma ⁵

Affiliations
1 Gurukul Institute of Engineering and Technology, Kota, IN
2 Department of Electronics and Communication Engineering, S.R.M. University Kattankuluthar, IN
3 Maharishi Arvind International Institute of Technology, Kota
4 St. Margaret Engineering College, Neemrana, IN
5 Rajasthan Technical University, Kota, IN

Abstract

The use of the proposed method in residue to binary conversion in Residue Number System (RNS) and Distributed Arithmetic Algorithm (DAA) in modern telecommunication and multimedia applications is becoming more and more important because it allows interesting advantages in terms of area, power consumption and speed.. This paper presents new method for conversion procedure (residue to binary) based on a {2ⁿ -1,2ⁿ , 2ⁿ+1, 2^{n + 1}+1, 2^{n - 1} -1} moduli set. Based on the proposed method for conversion in RNS and DAA algorithm, an architecture which efficiently implements the digital fir filter is synthesized using Xilinx Virtex2. Proposed method simplify the computing procedure by maximizing the utilization of the modulo-mi adders and multipliers present in the RNS functional units. For an n-digit RNS number X = (x₁, x₂, x₃, …., x_n) the proposed method takes at most n iterations. Each iteration requires one parallel subtractions and 2 multiplications except the first one.

Keywords

Residue Arithmetic, Distributed Arithmetic, Fir Filters, High Speed, VLSI.

Full Text

     

Ajay Sharma ¹, Annapurna Bhargava ², R. D. Rathor ³, Fanibhushan Sharma ⁴, Nirmala Sharma ⁵

Affiliations
1 Maharishi Arvind International Institute of Technology, Kota (RAJ), IN
2 Dept. of Electrical Engg., Rajasthan Technical University, Kota (RAJ), IN
3 Govt. Polytechnic, Kota (RAJ), IN
4 St. Margaret Engineering College, Neemrana (RAJ), IN
5 Rajasthan Technical University, Kota (RAJ), IN

Abstract

Short-term load forecasting is an important component in the power system load forecast, it is very important to unit optimum combination, economic scheduling, optimum current of dispatching department. Classical load forecasting methods include time sequence, regression method, and so on, but many of them have defects, for example, numerical value is instability and the factor which influences load can't be considered. Artificial intelligence method is main method now; neural network BP algorithm is representative among of them. When using neural network to predict electric power load, front neural network can predict with more precision fitting high linking and non-linear relation of shining upon between inputting and outputting from complicated sample data through studying. However some new problems have appeared while predicting electric power load using this method, it can't distinguish the impact on load data of the influence factor clearly, network structure can't be optimized and fixed automatically and need to confirm network structure artificially, the result is easy to fall into local optimum. So General Regression Neural Network- GRNN is proposed in this paper, it achieves global optimizing and can sample or calculate the data obtained to revise the network directly under the same structure, it need not calculate the parameter again, but only need a simple smooth parameter, it needn't carry on the training course of circulation.

Keywords

Short Term Load Forecasting, Artificial Neural Network, Generalized Regression Neural Network, Radial Basis Function.

Full Text