Variation Tolerant SRAM Cell for Low Power Applications
Static Random Access Memory (SRAM) is the important component across a wide range of microelectronics applications like high performance server processors, multimedia and System on Chip (SoC). The lower power consumption, static noise margin (SNM), read and write stability are the major design metrics for designing an SRAM cell. It is difficult to achieve that conventional 6 T SRAM cell in scaled technology, particularly in deep-subthreshold region. In this paper, impact of process parameters variations on various design metrics of the proposed cell are presented and compared with conventional 6T and 8T.The proposed 9T cell utilizes a scheme with separate read and write word lines it is shown that the 9T cell achieves improvements in power dissipation, performance and stability compared with previous designs (that require 6T and 8T) for low-power operation. The 9T scheme is amenable to small feature sizes as encountered in the deep sub-micron/nano ranges of CMOS technology. The proposed 9T SRAM cell designed in 130 nm Complementary Metal Oxide Semiconductor (CMOS) technology in the power supply voltage of 1.8 V. The simulation is done by using AARON 9.03 tool to achieve the power consumption of 91.55% lesser than existing method, Then the read and write delay of proposed 9T SRAM cell is reduced by 8.3% and 23.0% lesser than conventional 6T and 8T.
Published by: Dineshkumar .C, M. Parimala Devi
Author: Dineshkumar .C
Paper ID: V3I3-1257
Paper Status: published
Published: May 15, 2017
3D Face Reconstruction from 2D Images Using Data Mining Algorithm
Human faces are amazingly comparative in worldwide properties, including size, viewpoint proportion, and area of principle components, however can differ extensively in subtle elements crosswise over people, gender, race, or because of facial expression. We propose a novel technique for 3D face reconstruction of appearances that adventures the similitude of faces. Our technique use as info a solitary picture and uses an insignificant 3D reference model of an alternate individual's face. Traditional remaking techniques from single pictures, i.e. shape-from-shading, require learning of the reflectance properties and lighting and also profundity values for limit conditions. Late techniques evade these prerequisites by speaking to information confronts as mixes (of hundreds) of put away 3D models. We propose instead to use the input image as a guide to “mold” a single reference model to reach a reconstruction of the sought 3D shape.
Published by: Seema Chavan, Prajakta A. Satarkar
Author: Seema Chavan
Paper ID: V3I3-1255
Paper Status: published
Published: May 15, 2017
Automatic Irrigation System Based On Volumetric Soil Moisture Content Detection
In the field of agriculture, use of the proper method of irrigation is crucial. To have it done with reduced human intervention and still ensure proper irrigation is a challenge. We have developed an automatic irrigation system which switches the pump motor ON/OFF on sensing the moisture content of the soil. The soil Moisture Sensor module is used to detect the moisture of soil and infer to the volumetric presence of moisture present in the soil and if there is water on the sensor. We would use this module and insert it into the soil and then adjust the on-board potentiometer to modify the sensitivity. The sensor would give us the output logic HIGH when the moisture is higher and LOW when it is lower than the threshold value set by the potentiometer. The Soil Moisture sensor measure the volumetric water content in the soil, indirectly by using some other property of the soil, such as electrical resistance, dielectric constant or interaction with neutrons, as a substitute for the moisture content. We will preferably design soil moisture sensor using Arduino. An Arduino code will be used which will read the amount of moisture present in the soil and accordingly send information from the moisture sensor to the phone via the GSM module.
Published by: Anirudh Itagi, Nidhish Meherotra, Utkarsh Shukla, Dipak Singh, Prof. Deepika Rani Sona
Author: Anirudh Itagi
Paper ID: V3I3-1252
Paper Status: published
Published: May 15, 2017
Stochastic Economic Load Dispatch Using Pareto Distribution with Multiple Fuels Local Convex By Bacterial Foraging Optimization
In this paper, Stochastic Economic Load Dispatch problem with Multiple Fuels is solved using Pareto distribution with local convex optimization by bacterial foraging approach. Bacterial foraging optimization is swarmed intelligence technique used to solve the problem of the power system. The algorithm is based on the group foraging behavior of Escherichia coli (E-coli) bacteria present in the human intestine. This social behavior of E-coli bacteria has been used to solve optimization problems. This paper presents a BFO to solve Economic load dispatch (ELD) problems. The result is obtaining from test system with seven generating units. In this paper performance of the BFO is compared with Particle swarm optimization (PSO). The result clearly shows that the proposed method gives a better optimal solution as compared to the other methods.)
Published by: Akshay Rana, Shavet Sharma, Amandeep Thakur
Author: Akshay Rana
Paper ID: V3I3-1250
Paper Status: published
Published: May 15, 2017
Hybrid Recommendation Model With Nearest Neighbor Classification Based Collaborative Approach
The e-commerce product ranking models are designed to handle the very large volumes of the data involved in the database. Formally, the multifactor ranking models are incorporated over the online portals, which are capable of producing the multivariate lists. The product lists are prepared on the basis of various features, which include the popularity, accessibility and trust based factors which are associated with the e-commerce products for the realization of the content-based filtering over the e-commerce portals. In this paper, the multivariate and multifactor ranking algorithm has been proposed in order to solve the problems related to the low entropy, duplication and unbalanced feature analysis. The proposed model design is entirely based upon the popularity, visitor density, number of customers and security analysis based factors of the e-commerce pages containing the product lists. The multifactor values are organized in the different columns containing the different kinds of information, which are converted to the normalized and compatible values to create the data uniformity. The proposed model offers the collaborative index based product ranking model over the dense e-commerce databases. The proposed model has been designed to use the collaborative filtering based upon the k-nearest neighbor algorithm. The proposed model has been undergone the various experiments for the performance evaluation based upon the time complexity, resource utilization and other similar factors.
Published by: Ramandeep Sharma, Samarth Kapoor
Author: Ramandeep Sharma
Paper ID: V3I3-1238
Paper Status: published
Published: May 15, 2017
Design and Implementation of Fir Filter Using Retiming Technique
Finite Impulse Response (FIR) Filter can be designed by provision of specifications which are for a particular application requirement. An efficient FIR filter is designed using register reduction retiming technique. Also, an optimization environment is designed such that filter components of post retimed circuit such as adder and multiplier are upgraded depending on the Very Large Scale Integration (VLSI) design metrics such as area, speed and power. In conventional cut-set retiming of FIR filter, higher critical path delay and latency occurs due to unwanted pipelining. They require higher clock period. So overcome, FIR filter it’s designed using Novel node – splitting and Node merging techniques. These schemes reduce the critical path delay by 50% and latency by 60%. In this paper node merging method is simulated and its performance parameters such as power, delay, critical path, latency, clock period and number of registers are analysed. In Data Flow Graph (DFG) retiming of digital circuit is used to reduce the propagation delay and critical path. The reduced propagation delay in turn reduces pipeline overheads. Thus, cut-set retiming can be made more efficient without increasing the register complexity and latency. The FIR architecture for node merging has to be designed and its performance is to be analysed. The proposed FIR filter using retiming technique are simulated using MODELSIM 10.1b and power analysis of the proposed work is done using Xilinx ISE 9.2i. In this proposed work the layout will be developed using AARON 9.03 tool.
Published by: R. Dhivya, Dr. P. Brindha
Author: R. Dhivya
Paper ID: V3I3-1237
Paper Status: published
Published: May 15, 2017
