Energy Efficient Memory Design using Low Voltage Complementary Metal Oxide Semiconductor on 28nm FPGA

Similar documents
Webpage: Volume 3, Issue V, May 2015 ISSN

Power Efficient Optimized Arithmetic and Logic Unit Design on FPGA

Energy Efficient and High Performance 64-bit Arithmetic Logic Unit using 28nm Technology

Course Outcome of M.Tech (VLSI Design)

Aalborg Universitet. Published in: Indian Journal of Science and Technology. DOI (link to publication from Publisher): /ijst/2016/v9i25/96633

(VE2: Verilog HDL) Software Development & Education Center

Design and Implementation of FPGA Based Digital Base Band Processor for RFID Reader

Technology Timeline. Transistors ICs (General) SRAMs & DRAMs Microprocessors SPLDs CPLDs ASICs. FPGAs. The Design Warrior s Guide to.

PE713 FPGA Based System Design

A Case Study of Nanoscale FPGA Programmable Switches with Low Power

A Novel Latch design for Low Power Applications

Comparative Analysis of different Algorithm for Design of High-Speed Multiplier Accumulator Unit (MAC)

<Explanation of Improved the Quality of ALU And Ten Different Types of Designs for Decreasing Power Dissipation>

Low Power Techniques for Digital System Design

Self-Checking Carry-Select Adder Design Based on Two-Pair Two-Rail Checker

A COMPARATIVE ANALYSIS OF AN ULTRA-LOW VOLTAGE 1-BIT FULL SUBTRACTOR DESIGNED IN BOTH DIGITAL AND ANALOG ENVIRONMENTS

FPGA Implementation of Digital Modulation Techniques BPSK and QPSK using HDL Verilog

NOVEL DESIGN OF 10T FULL ADDER WITH 180NM CMOS TECHNOLOGY

International Journal of Modern Engineering and Research Technology

PV SYSTEM BASED FPGA: ANALYSIS OF POWER CONSUMPTION IN XILINX XPOWER TOOL

Performance Enhancement of the RSA Algorithm by Optimize Partial Product of Booth Multiplier

Implementation of 32-Bit Carry Select Adder using Brent-Kung Adder

A10-Gb/slow-power adaptive continuous-time linear equalizer using asynchronous under-sampling histogram

Chapter 1 Introduction

Implementation of a FFT using High Speed and Power Efficient Multiplier

DESIGN OF EXTENDED 4-BIT FULL ADDER CIRCUIT USING HYBRID-CMOS LOGIC

PROMINENT SPEED ARITHMETIC UNIT ARCHITECTURE FOR PROFICIENT ALU

VLSI IMPLEMENTATION OF MODIFIED DISTRIBUTED ARITHMETIC BASED LOW POWER AND HIGH PERFORMANCE DIGITAL FIR FILTER Dr. S.Satheeskumaran 1 K.

2014, IJARCSSE All Rights Reserved Page 459

POWER OPTIMIZED DATAPATH UNITS OF HYBRID EMBEDDED CORE ARCHITECTURE USING CLOCK GATING TECHNIQUE

Low Power Design of Successive Approximation Registers

Design Analysis of 1-bit Comparator using 45nm Technology

AN EFFICIENT APPROACH TO MINIMIZE POWER AND AREA IN CARRY SELECT ADDER USING BINARY TO EXCESS ONE CONVERTER

CMOS VLSI IC Design. A decent understanding of all tasks required to design and fabricate a chip takes years of experience

DESIGN OF LOW POWER MULTIPLIER USING COMPOUND CONSTANT DELAY LOGIC STYLE

EC 1354-Principles of VLSI Design

SIGNED PIPELINED MULTIPLIER USING HIGH SPEED COMPRESSORS

Design of Signed Multiplier Using T-Flip Flop

FPGA Based System Design

Analysis of Low Power-High Speed Sense Amplifier in Submicron Technology

DESIGN OF LOW POWER / HIGH SPEED MULTIPLIER USING SPURIOUS POWER SUPPRESSION TECHNIQUE (SPST)

Design and Implementation of High Speed Carry Select Adder Korrapatti Mohammed Ghouse 1 K.Bala. 2

Available online at ScienceDirect. International Conference On DESIGN AND MANUFACTURING, IConDM 2013

II. Previous Work. III. New 8T Adder Design

Design and Implementation of Complex Multiplier Using Compressors

EE19D Digital Electronics. Lecture 1: General Introduction

Reduced Area Carry Select Adder with Low Power Consumptions

Fpga Implementation of Truncated Multiplier Using Reversible Logic Gates

[Singh*, 5(3): March, 2016] ISSN: (I2OR), Publication Impact Factor: 3.785

FPGA Implementation of Wallace Tree Multiplier using CSLA / CLA

ECE6332 VLSI Eric Zhang & Xinfei Guo Design Review

Design of Delay-Power Efficient Carry Select Adder using 3-T XOR Gate

Lecture 1. Tinoosh Mohsenin

Layout Design of LC VCO with Current Mirror Using 0.18 µm Technology

Implementation of a Block Interleaver Structure for use in Wireless Channels

Design and Implementation of High Speed Carry Select Adder

Analysis of shift register using GDI AND gate and SSASPL using Multi Threshold CMOS technique in 22nm technology

VCO Based Injection-Locked Clock Multiplier with a Continuous Frequency Tracking Loop

32-Bit CMOS Comparator Using a Zero Detector

COMPARISION OF LOW POWER AND DELAY USING BAUGH WOOLEY AND WALLACE TREE MULTIPLIERS

Heterogeneous Concurrent Error Detection (hced) Based on Output Anticipation

Implementation of Parallel Multiplier-Accumulator using Radix- 2 Modified Booth Algorithm and SPST

Power Efficiency of Half Adder Design using MTCMOS Technique in 35 Nanometre Regime

A Low Power VLSI Design of an All Digital Phase Locked Loop

Reference. Wayne Wolf, FPGA-Based System Design Pearson Education, N Krishna Prakash,, Amrita School of Engineering

CMOS Process Variations: A Critical Operation Point Hypothesis

The Metrics and Designs of an Arithmetic Logic Function over

MODELLING AND TESTING OF GATE OXIDE SHORTS IN SRAM AND DRAM

Design and Analysis of Sram Cell for Reducing Leakage in Submicron Technologies Using Cadence Tool

International Journal of Scientific & Engineering Research Volume 3, Issue 12, December ISSN

Separate Dual-Transistor Registers - A Circuit Solution for On-line Testing of Transient Error in UDSM-IC

High Speed and Reduced Power Radix-2 Booth Multiplier

Comparitvie Analysis and Proposed Schmitt Trigger Design using Different CMOS Foundries

Design and Implementation of Reversible Multiplier using optimum TG Full Adder

Low Power &High Speed Domino XOR Cell

VLSI Implementation of Digital Down Converter (DDC)

FPGA Implementation of Viterbi Algorithm for Decoding of Convolution Codes

Digital design & Embedded systems

Implementation of dual stack technique for reducing leakage and dynamic power

DESIGN OF LOW POWER HIGH PERFORMANCE 4-16 MIXED LOGIC LINE DECODER P.Ramakrishna 1, T Shivashankar 2, S Sai Vaishnavi 3, V Gowthami 4 1

An Optimized Design for Parallel MAC based on Radix-4 MBA

Low Power Adiabatic Logic Design

Design of High Speed Power Efficient Combinational and Sequential Circuits Using Reversible Logic

Design of Static Segment Adder for Approximating Computing Applications

Implementation of a High Speed and Power Efficient Reliable Multiplier Using Adaptive Hold Technique

2 Assoc Prof, Dept of ECE, George Institute of Engineering & Technology, Markapur, AP, India,

IJMIE Volume 2, Issue 3 ISSN:

IMPLEMANTATION OF D FLIP FLOP BASED ON DIFFERENT XOR /XNOR GATE DESIGNS

A HIGH SPEED FIFO DESIGN USING ERROR REDUCED DATA COMPRESSION TECHNIQUE FOR IMAGE/VIDEO APPLICATIONS

Implementation and Performance Analysis of different Multipliers

Analysis of Parallel Prefix Adders

EECS150 - Digital Design Lecture 28 Course Wrap Up. Recap 1

Design and Implementation of Truncated Multipliers for Precision Improvement and Its Application to a Filter Structure

A Fixed-Width Modified Baugh-Wooley Multiplier Using Verilog

Integration of Optimized GDI Logic based NOR Gate and Half Adder into PASTA for Low Power & Low Area Applications

Anitha R 1, Alekhya Nelapati 2, Lincy Jesima W 3, V. Bagyaveereswaran 4, IEEE member, VIT University, Vellore

Low Power Realization of Subthreshold Digital Logic Circuits using Body Bias Technique

Digital Systems Design

An Efficent Real Time Analysis of Carry Select Adder

Design and Performance Analysis of High Speed Low Power 1 bit Full Adder

Transcription:

Indian Journal of Science and Technology, Vol 8(17), DOI: 10.17485/ijst/20/v8i17/76237, August 20 ISSN (Print) : 0974-6846 ISSN (Online) : 0974-5645 Energy Efficient Memory Design using Low Voltage Complementary Metal Oxide Semiconductor on 28nm FPGA Sweety 1*, Minal Dhankar 1, Ravinder S. Kajal 1, Kartik Kalia 2, Kushagra Vashishta 2 and Amit Kumar 3 1 Department of Computer Science, Maharaja Surajmal Institute, Janakpuri, Delhi-110058, India; sweety.dabas@gmail.com 2 Department of ECE, Chitkara University, Punjab - 140 401, India 3 Department of IT, IIIT Gwalior, Gwalior, India Abstract In this work, we are designing a energy efficient memory circuit on 28nm FPGA. Four different are used to validate the energy efficient design. There is 40.67% power reduction when is used in place of. is better than IO Standard according to our experiment. With there is 75.70% total power reduction in compare with the. is most energy efficient IO Standard and is most power consuming IO Standard. To design a power efficient memory we are using Verilog as HDL, Xilinx ISE 14.6 simulator with kintex-7 FPGA. Keywords: Energy Efficient, FPGA, IO Standard, Low Power,, Memory 1. Introduction Role of IO standard is very crucial to make a power efficient digital circuit. IO standards are the input and output standards which are required to receive the input for the input receiver or to provide the result to output drive. In this work, we implemented our design on 28nm kintex-7 FPGA using Low Voltage Complementary Metal Oxide Semiconductor () IO standard. Kintex-7 FPGA also supports different family of IO standard other than like SSTL, HSTL, and LVDCI. The core reason of this work is to explore a power optimized design to save the energy. Kintex-7 FPGA requires the low power and produces the high performance circuit. Kintex-7 28nm FPGA is a new class of FPGA which provides the 50% less power consumption than the previous 40nm devices. is standard general purpose IO interface which provide a low voltage class of CMOS technology for unified impedance matching and to make a system to minimize the energy consumption in order to fulfill our goal of power optimized design of memory. We implemented our logic on a memory chip. Random Access Memory (RAM) is a form of computer memory which is used to store the data. It allows the stored data to be accessed quickly in random order rather than sequentially. It is primary or main memory which is directly accessed by the CPU. It is used to read/write the data very quickly. It stores the data in form of pages inside it. There are different page replacement algorithms which can be used to replace the data of RAM (like FIFO and LRU) to optimize the search process. It is volatile memory and needs the power to keep the data accessible, if power is lost or off all data contained in memory will lose. SRAM and DRAM are the categorization of RAM. It can hold very few data at a time inside it and it is not a permanent storage device like a CD or hard drive. In 1 IO Standards have been studied on 3.3V and 1.8V for verifying the frequency. Result shows that, for Input Output Buffer (IOB) characterization a simple pass/fail test is sufficient. During Fault Injection (FI) those bits will shown that affect the voltage oscilloscope. Mitigation techniques are used to find out the incorrect *Author for correspondence

Energy Efficient Memory Design using Low Voltage Complementary Metal Oxide Semiconductor on 28nm FPGA Figure 1. Family of IO Standard. behavior of the bits. Triple Module Redundancy (TMR) is a most efficient and useful mitigation technique in which missing or incorrect behavior are triplicates and to implement this mitigation technique mitigation logic should be fabricated in advance in circuit. Basically the motive of this work is to evaluate the performance of output block when non-mitigated is compared with mitigated block of output. In 2 0.5 μm SOI-CMOS is used with distributed RAM logic on LUT structure for designing and its verification. Here defined that pipeline design can be used to improve the hit ratio so that the output and address signal resisted synchronously. The design guarantees the improvement in frequency. In 3 Integrated automatic optimization flow technique solves the problems of memory partitioning which is the combination of four techniques. This technique is user for memory optimization which is combination of memory partitioning and pipelining merged with data reuse to improve behavior synthesis. In 4 Virtix-6 FPGA family of IO standards are used in ALU to reduce the power consumption and to design and implement a power efficient ALU. Power consumption will be increased when the frequency will increase because power is directly proportional with the frequency, so design of power efficient ALU directly depends upon the frequency passed in the circuit also. Family of is compared with three different families of IO standards like HSTL, SSTL, LVDCI and proved the most energy efficient IO standard for ALU among the rest. Figure 3. 16-bit RAM Implemented using Verilog HDL. In 5, Low Voltage Digitally Controlled Impedance (LVDCI) is the family of IO standard used to find the best DCI to reduce the power consumption. LVDCI is used at 1.2V, 1.5V, 1.8V, 2.5V output drive and LVDCI_ reduced 50percent of power consumption with comparison to other voltage in DCI. LVDCI_12 proved to most power efficient DCI and LVDCI_ consumed the maximum power among the rest. 2. Block Diagram of Memory This is 16 bit RAM used in the experiment to design a power efficient circuit. 2.1 Top Level Schematic of 16-bit Memory Read/Write capability is enabled in this RAM. The uniqueness of this design is that it is using a power efficient IO standard that makes it more energy efficient memory. 2.2 RTL Schematic of 16-bit Memory RTL stands for Register Transfer Language. RTL schematic show how the logic is implemented and it is saved with *.ngr file extension. The term NGR means Native Generic Register File. In Figure 3, there are 16 data flip-flops. Figure 2. Random Access Memory (RAM) Categorization. 2 Vol 8 (17) August 20 www.indjst.org Indian Journal of Science and Technology

Sweety, Minal Dhankar, Ravinder S. Kajal, Kartik Kalia, Kushagra Vashishta and Amit Kumar 3. Power Analysis 3.1 Power Consumption on 1 GHz The categorizations of power are Static and Dynamic. Leakage power is static power and Clock, Logic, Signal and I/O powers are Dynamic Powers. In Table 1, Clock power remains 0.022W, Logic power remains 0.000W, Signal Power remains 0.026W and BRAMs Power remains 0.753W for IO Standard at voltage 1.5V, 1.8V, 2.5V and 3.3V with device operating on 1 GHz frequency. IO power get changed frequently when we change the voltage over. Using Table 2 there is a 40.67percent power reduction when is used at place of, 65.53percent power reduction when is used at place of to design the memory and 75.70percent total power reduction when is used at place of over 1GHz frequency. Below in Figure 4, It shows the total power consumption between the family of IO standard at voltage 1.5V, 1.8V, 2.5V and 3.3V on device operated over 1 GHz frequency. This Figure 4 clearly shows that is most power saver IO standard and is consumes maximum power, So at voltage 1.5V is more Power efficient as compared to at voltage 1.8V, 2.5V and 3.3V on 1 GHz frequency. Figure 4. I/O Power Consumption on 1 GHz Device 3.2 Power Consumption on 10 GHz In Table 3, Clock power remains 0.209W, Logic power remains 0.003W, Signal Power remains 0.224W and BRAMs Power remains 7.753W for IO Standard at voltage 1.5V, 1.8V, 2.5V and 3.3V on device operating over 10 GHz frequency. Table 4 shows, there is a 40.31 percent power reduction when is used by replacing, 65.57 percent of power reduction when is used at place of and 75.93 percent total power reduction when is used at place of to design a power efficient memory over 10GHz frequency. In Figure 5, that shows the total power consumption of the between the family of IO standard at Table 1. Power Consumption on 1 GHz Device Clock 0.022 W 0.022 W 0.022 W 0.022 W Logic 0.000 W 0.000 W 0.000 W 0.000 W Signal 0.026 W 0.026 W 0.026 W 0.026 W BRAMs 0.753 W 0.753 W 0.753 W 0.753 W IOs 0.043 W 0.061 W 0.105 W 0.177 W Leakage 0.082 W 0.082 W 0.083 W 0.084 W Total 0.926 W 0.945 W 0.990 W 1.063 W Table 2. Power Efficiency on 1 GHz in Comparison With 40.67 % 65.53 % 75.70 % Table 3. Power Consumption on 10 GHz Device Clock 0.209 W 0.209 W 0.209 W 0.209 W Logic 0.003 W 0.003 W 0.003 W 0.003 W Signal 0.224 W 0.224 W 0.224 W 0.224 W BRAMs 7.535 W 7.535 W 7.535 W 7.535 W IOs 0.4 W 0.608 W 1.054 W 1.766 W Leakage 0.106 W 0.106 W 0.106 W 0.106 W Total 8.531 W 8.685 W 9.134 W 9.850 W Table 4. Power Efficiency on 10 GHz in comparison with 40.31 % 65.57 % 75.93 % Vol 8 (17) August 20 www.indjst.org Indian Journal of Science and Technology 3

Energy Efficient Memory Design using Low Voltage Complementary Metal Oxide Semiconductor on 28nm FPGA Figure 5. I/O Power Consumption on 10 GHz Device voltage 1.5V, 1.8V, 2.5V and 3.3V on device operated on 10 GHz frequency. It clearly shows that at voltage 1.5V is most power efficient and is least power efficient family of IO standard when implemented to design a power efficient memory circuit, So at voltage 1.5V is more Power efficient as compared to at voltage 1.8V, 2.5V and 3.3V on 10 GHz frequency. 3.3 Power Consumption on 100 GHz Table 5, shows the Clock power remains consistent at 2.110W, Logic power remains 0.010W, Signal Power remains 1.023W and BRAMs Power remains 75.346W for IO Standard at voltage 1.5V, 1.8V, 2.5V and 3.3V on device operating on 100 GHz frequency. Table 5. Power Consumption on 100 GHz Device Clock 2.110 W 2.110 W 2.110 W 2.110 W Logic 0.010 W 0.010 W 0.010 W 0.010 W Signal 1.023 W 1.023 W 1.023 W 1.023 W BRAMs 75.346 W 75.346 W 75.346 W 75.346 W IOs 4.3 W 6.076 W 10.536 W 17.655 W Leakage 1.050 W 1.050 W 1.051 W 1.052 W Total 84.091 W 84.616 W 90.076 W 97.196 W Table 6. Power Efficiency on 100 GHz in comparison with 40.32 % 65.58 % 75.91 % Figure 6. I/O Power Consumption on 100 GHz Device In Table 6, there is a 40.32 percent power reduction when is used by replacing, 65.58 percent of power reduction when is used at place of and 75.91 percent total power reduction when is used at place of to design a power efficient memory over 100 GHz frequency. Figure 6, shows the total power consumption of the between the family of IO standard at voltage 1.5V, 1.8V, 2.5V and 3.3V on device operated over 100 GHz frequency. It clearly shows that at voltage 1.5V is most power efficient and is least power saver family of IO standard, so at voltage 1.5V is more Power efficient as compared to at voltage 1.8V, 2.5V and 3.3V on 100 GHz frequency. 3.4 Power Consumption on 1 THz Table 7, shows the Clock power remains consistent at 1.8V, 2.5V and 3.3V but it is 22.310W at voltage 1.5. Logic power remains 0.079W, Signal Power remains 8.985W at voltage 1.8V, 2.5V, and 3.3V and it fluctuate on voltage 1.5V. BRAMs Power remains 753.463W for at voltage 1.5V, 1.8V, 2.5V and 3.3V on device operating on 1 THz frequency. Table 7. Power Consumption on 1 THz Device Clock 22.310 W 21.099 W 21.099 W 21.099 W Logic 0.079 W 0.079 W 0.079 W 0.079 W Signal 10.802 W 8.985 W 8.985 W 8.985 W BRAMs 753.463 W 753.463 W 753.463 W 753.463 W IOs 42.529 W 60.762 W 105.359 W 176.550 W Leakage 1.050 W 1.050 W 1.050 W 1.050 W Total 830.234 W 845.439 W 890.036 W 961.228 W 4 Vol 8 (17) August 20 www.indjst.org Indian Journal of Science and Technology

Sweety, Minal Dhankar, Ravinder S. Kajal, Kartik Kalia, Kushagra Vashishta and Amit Kumar In Table 8, there is a 40.32 percent power reduction when is used by replacing, This is same percentage of power reduction when the 100 GHz of frequency supplied to the circuit. It shows 65.60 percent of power reduction when is used at place of and 75.91 percent total power reduction when is used at place of to design a power efficient memory over 1 THz frequency. Figure 7, shows the total power consumption between the family of IO standard at voltage 1.5V, 1.8V, 2.5V and 3.3V on device operated on 1 THz frequency. It clearly shows that at voltage 1.5V is most power efficient and is least power saver family of IO standard according to the Figure 7, So at voltage 1.5V is more Power efficient as compared to at voltage 1.8V, 2.5V and 3.3V on 1 THz frequency. Our experiment to design a memory which consumes minimum power and gives the best result shows that, is the most and is least power efficient Low Voltage CMOS family of IO standard and there is slightly a minor change when we increase or decrease the frequency to the device. When we design and implement the memory circuit using IO standard at voltage 2.5V over 3.3V circuit the power reduction in percent is 40.67 percent at 1GHz, 40.31 percent at 10GHz, Table 8. Power Efficiency on 100 GHz in comparison with 40.32 % 65.60 % 75.91 % Figure 7. I/O Power Consumption on 1 THz Device Table 9. Power Efficiency of family of IO Standards POWER REDUCTION 40.32 percent at 100 GHz and 40.32 percent at 1THz. So if we design memory using, it will save maximum power at 1GHz frequency. Designing a circuit using over 65.60 percent is maximum possible power reduction which is on 1THz frequency and when we use to design memory over there is 75.93percent total power reduction possible at 10 GHz frequency. From the Table 9, it is clear that is the most power efficient when compared with and at voltage 3.3V is the least power efficient to design a power saver memory. 4. Conclusion at voltage 1.5V is proved most power efficient IO standard and at voltage 3.3V is least power efficient among the 4 different Low Voltage Complementary Metal Oxide Semiconductor () to design and implement a 16-bit power efficient memory circuit. reduces the power consumption of the circuit to 75.93% when it compared with the circuit which is designed using at Kentix-7 FPGA and the performance of is most power efficient at 10 GHz frequency. 5. Future Scopes 40.67% at 1 GHz 65.60% at 1THz 75.93% at 10 GHz This design is implemented on 28nm Kentix-7 FPGA. There is a scope to redesign this memory on 16nm ultra scale FPGA in order to reduce the area on FPGA and minimize the power consumption of memory. We can also implement this design on Vertix FPGA at place of Kentix FPGA. In this work, IO standards are an effective technique to reduce the power consumption of a circuit. Therefore it is important to choose the correct IO standard to make a design which save the power and make it a power efficient circuit. IO standards can use in other devices also like Arithmetic and Logic Unit (ALU), Control Unit (CU), Memory Unit (MU) to increase their power efficiency. Vol 8 (17) August 20 www.indjst.org Indian Journal of Science and Technology 5

Energy Efficient Memory Design using Low Voltage Complementary Metal Oxide Semiconductor on 28nm FPGA 6. References 1. Rezgui S, Swift GM, Lesea A. Characterization of Upsetinduced Degradation of Error-Mitigated High-Speed I/O s using Fault Injection on SRAM based FPGAs. IEEE Transactions on Nuclear Science. 2006 Aug; 53(4):2076 83. 2. Han X, Chen SL, Wu L, Yan Z, Li Y. Design and verification of Distributed RAM using Look-Up Tables in an SOI-based FPGA. 10th IEEE International Conference on Solid- State and Integrated Circuit Technology (ICSICT); 2012; p. 306 8. 3. Wang Y, Zhang P, Cheng X, Cong J. An Integrated and Automated Memory Optimization Flow for FPGA Behavioral Synthesis. 2012, 17th Asia and South Pacific Design Automation Conference (ASP-DAC); 2012; IEEE; p. 7 62. 4. Pandey B, Yadav J, Singh Y, Kumar R, Patel S. Energy Efficient Design and Implementation of ALU on 40-nm FPGA. IEEE International Conference on Energy Efficient Technologies for Sustainability-(ICEETs); 2013 Apr 10-12; Nagercoil, Tamilnadu. p. 45 50. 5. Pandey B, Kumar R. Low Voltage DCI based Low Power VLSI Circuit Implementation on FPGA. IEEE Conference on Information and Communication Technologies (ICT 2013); 2013 Apr 11-12; p. 128 31. 6 Vol 8 (17) August 20 www.indjst.org Indian Journal of Science and Technology