A person smiling for the camera

Description generated with very high confidence

Dr. Babu Devnarayan Choudhary

 

Techcelerate Fellow,

Imperial College London,

South Kensington, London SW7 2AZ

 

Email: b.choudhary@imperial.ac.uk, bdc@thinkclock.com

Phone: +44 (0)7517581192 ; +44 (0)7714051202

 

My CV: DOC ; PDF

 

My Profile URL: www.thinkclock.com/bdc

My LinkedIn Profile: https://www.linkedin.com/in/bdevnarayan/

My Profile at Imperial: www.imperial.ac.uk/people/b.choudhary

 

I am passionate about developing new technologies having transformative potential. Currently, I am involved in the further development of our CogniSenseTM technology which lets you monitor machines using RF signal non-invasively, and we believe this is set to revolutionize the Industrial IoT space.

 

So far, the predictive maintenance is not a reality. There is enough algorithmic intelligence but not enough data to feed. For collecting the data, machines need to be instrumented with sensors, which is a daunting task due to the hassles of wiring & packaging. CogniSense is a unique and promising technology based on two patents. This is the outcome of my PhD research, which has resulted in a Spin-Out with the support of Imperial Innovations Ltd. It is currently in development at TRL-4, and has received significant interest from a range of industries; which has helped us to raise £200,000 seed funding so far. This was covered by Digital-Catapult during Smart Factory Expo 2017 in Liverpool, UK (see here).

 

My PhD research was about how we can do Sensing without Sensors using RF signal. My expertise is around Signal Processing, Wireless Communication and Systems Modelling.

 

At heart, I am a curious being eager to understand everything with the first principle and, find it exciting where rational understanding falters.

 

 

Profile Summary:

v  In-depth understanding of communication systems especially Physical Layer (Transceiver Architecture, Antenna Theory, RF propagation, LTI systems etc.).

v  Strong implementation experience of Analog/Digital Signal Processing related to Noise Filtering, Spectral Analysis, Beamforming, Localization, RADAR, MIMO, Modulation/Demodulation etc. along with Electromagnetic Modelling.

v  Over 10-year experience of building complex embedded systems from prototyping to deployment involving sensors, actuators and wireless components (Cellular, LPWAN, VHF, SDR etc.) with various processor boards.

v  Wholistic understanding of system design involving trade-offs related to cost, performance, energy consumption etc. in-tune with the business logic.

v  My two patented inventions - CogniSenseTM and PhysioGram have been identified as transformative technologies by Imperial Innovations leading to a Spin-Out with over £200K funding already secured.

 

Education:

PhD

[Aug.’11-Sep.’17]

Thesis: “Sensing without Sensors: Non-Intrusive methods for Cognitive Sensing”

Department of Computing, Imperial College London, UK

Supervisor: Prof. Julie A. McCann

 

 Graduation 

[2003-07]       

B.Tech in Information Technology

CGPI– 7.98/10.0 (3.2/4.0)

Indian Institute of Information Technology, Allahabad, India

 

XII: [2001]

X   : [1999]

84.31 % (Rajasthan Board, India)

83.17 % (Rajasthan Board, India)

 

Work Experience:

[Aug’2011-Today]

 

Research Assistant,

AESE Group, Department of Computing,

Imperial College London, UK

[Sep’08-July’2011]

Founding Director & Research Engineer,

ThinkClock Innovation Labs Pvt. Ltd., IIIT Allahabad, India

[July’07-June’08]

Associate Software Engineer, (Information Security)

SafeNet India Pvt. Ltd., Noida, India

[Jan-July 2007]

Research Internship, (Information Security)

LESTER Laboratory, University of South Brittany, Lorient, France

 

 

Achievements:

v  Awarded £15,000 seed funding plus £150,000 follow-up funding by Imperial Innovations for developing my patented technology -CogniSenseTM for commercial trials.

v  Awarded £35,000 towards commercialization of my CogniSenseTM invention through the “Techcelerate Programme 2018” by Imperial.

v  Selected by Digital-Catapult to pitch CogniSenseTM to Industry Audience at Smart Factory Expo 2017 in Liverpool, which resulted in numerous industrial collaborations (see link).

v  Awarded 75,000 INR prize, as a winner of Next Big Idea Contest 2011 organized by Intel and Department of Science & Technology, Govt. of India in association with IIM Bangalore.

v  Awarded 525,000 INR (approx. $10,500) of initial funding from the TIDE scheme of Govt. of India for my ThinkClock Tactile project.

v  Awarded 100,000 INR prize by Texas Instruments, as a winner in All India DSP design Contest 2006 for project “Collision Avoidance System for Smart Automobiles”.

v  Awarded French International Scholarship (1000 euro/month plus airfare) from EGIDE for final semester internship at LESTER Lab, UBS France.  

v  Awarded Honorable Mention (including $300 prize and trip sponsorship to the award ceremony in San-Francisco) by 43rd Design Automation Conference ( DAC 2006 ) for the paper titled “Phase Delay based Collision Avoidance RADAR for Smart Automobiles”.

v  Selected by ISRO/ASI as one of 20 semi-finalists from across India for International Astronautical Congress (IAC 2006) held at Valencia, Spain.

v  One of the 6-finalists in Idea-to-Implementation 2008 -an All India Business Plan Competition held at IIM Calcutta.

v  One of the 8-finalists in NIRVAAN 2007 -an All India Business Plan Competition held at IIM Lucknow.

v  KVPY-2004 Fellowship. KVPY is a research program conducted by IISc Bangalore and funded by Department of Science and Technology, Government of India, in which about 20 students are selected from engineering stream every year.

 

Communication Skills:

v  Involved in teaching Signal Processing to Masters Students at Imperial College London.

v  Involved in marketing the research to industry audience through the commercialization activities of our CogniSenseTM technology.

v  Provided consultancy to Syngenta reflecting the role of IoT in precision agriculture.

 

Projects

 

S4: Science for Sensor Systems Software

Imperial College London

Sponsor: EPSRC

Time: Jan. 16-Dec. 17

Signal Processing: MATLAB-Simulink, Noise Filtering, Spectral Analysis, SNR Enhancement, MIMO Beamforming, WiFi-OFDM, Correlation, Classification and Localization based on basis functions.

Instruments: Software Defined Radio (USRP, RTL SDR) Handheld Spectrum Analyser, PC based Oscilloscope: Picoscope, RF Components: various antennas, gain/phase detectors, PLL, up/down convertors, RF switches, transmission lines, Directional Couplers, Impedance Transformers, RF power detectors, other components involved in Tx/Rx frontend.

 

This project: aims to develop a unifying model of IoT hardware and software in combination with the operational environment to provide a digitized replica of the whole system.

My role: has been to develop Non-Invasive methods to perform sensing using the RF signal. I have developed two new methods PhysioGram and CogniSenseTM from concept to TRL-5, utilizing an in-depth understanding of Communication, Control, and LTI System’s Theory.

v  PhysioGram Patent:

Babu D. Choudhary, Julie A. McCann, “A method and apparatus for Motional, Structural and Compositional characterization of a Physical System using Radio signal”. (In Process)

v  CogniSenseTM Patent:

Babu D. Choudhary, Julie A. McCann, “An apparatus and method for generating a motional signature indicative of motion of moving parts of a target machine”. (P109871PCT, Filing Date: 8 July 2016)

v  Awarded £15K for Proof-of-Concept followed by £150K for further development of CogniSenseTM by Imperial Innovations for commercial trials.

v  Impact: trials of CogniSenseTM are currently being setup with RS Components, Romax InSight, GSK, BOC etc.

 

Civil Infrastructure

Monitoring and Control

Imperial College London

Sponsor: EIT ICT Labs

Time: July 2015-Dec. 2015

Skills: Linux, RS232 Comport communication, Python script, TI Code Composer Studio (CCS), TI RF Studio.

PCB design, power regulation, Antenna placements, Near-field effect, Electromagnetic Interference (EMI) mitigation, Spectrum Analyser, Field Experiments.

 

This project: involved instrumentation of utility networks with IoT.

My role: I was the part of the connectivity team responsible for identifying the most appropriate LPWAN communication module for the urban environment. We have created a wireless testbed to evaluate various modules (LoRa, NWave, XBee, Sigfox and TI CC430), and deployed it around Imperial and Hyde Park to perform a comparative study.

Output: Publication

Sokratis Kartakis, Babu D. Choudhary, Alexander D. Gluhak, Lambros Lambrinos, and Julie A. McCann, “Demystifying low-power wide-area communications for city IoT applications”, In Proceedings of the Tenth ACM International Workshop on Wireless Network Testbeds, Experimental Evaluation, and Characterization (WiNTECH '16).

 

CubeSat: Adaptive communications and resource management for very large scale/very low-cost space systems

Imperial College London

Sponsor: UK Space Agency

Time: Feb. 2015-July 2015

Skills: TI CCS Studio, RTOS, TI RF Studio, Energia, Wireless MCU CC4305137 with multi-band radio, solar cells, Indoor Emulation of the test environment.

 

This: was an exploratory project to assess the feasibility of Thin-Film based IoT devices which are battery-free and solar powered. Such ultra-light, ultra-small, ultra-cheap and ultra-low-powered satellites could be a low-cost alternative for space exploration using a CubeSat class payload.

My Role: considering that any microchip circuit can be converted into a Thin-Film version, I have developed Adaptive and Predictive task scheduling algorithms with insight from off-line task profiling and solar-cell characteristics in order to optimize the performance of a swarm of micro-satellites in terms of its longevity and connectivity under the uncertainty of space weather.

 

FUSE: Floodplain Underground SEnsor Network

Imperial College London

Sponsor: NERC

Time: Aug. 2011-Jan. 2015

Lab Equipment: Oscilloscope, Logic Analyzer, Vector Analyzer

Processor: ARM Processor Board, Arch Linux, C/C++, multi-threading, interrupts for wake-up, communication interfaces -SPI, I2C, UART etc.

Peripherals: ADC/Relay Board, Real Time Clock, SD Card, Sensor excitation circuit, MOSFET switch, GPIO etc.

Wireless Modules: ISM band transceivers of different frequencies (Radiometrix: 173MHz, 433MHz, 458MHz and 869MHz) with different Tx power levels (0-500mW), licensed-band VHF module (Warwick X8200 5-Watt), SIMCOM GSM/GPRS modem, Adaptive MAC layer, Communication through cellular network using AT commands for SMS, Call, UDP, TCP, HTTP, FTP, Email etc..

Sensors:

Digital-Output- Decagon-5TE, Decagon MPS-6, SENSIRION - SHT75. Analog Output: Vegetronix VG400, IMSL Water Table Sensor

 

This project: was aimed to deploy an Underground Wireless Sensor Network in a floodplain along the Thames river for soil and vegetation monitoring in-order to quantify the hydro-ecological interactions.

My role: This was in collaboration with soil and environmental scientists, but I was solely responsible for the technology development from requirement-capturing to deployment with following activities involved.

v  Field experiments related to wireless communication through soil with various ISM band radio modules in different configurations under different soil conditions.

v  Evaluating various environmental sensors (Temperature, Conductivity, Moisture, Water-Potential etc.) to find best options.

v  Antenna Physics, Electromagnetic Modelling using MATLAB, understanding RF propagation through soil-water, Link-budget analysis, understanding energy flow from Battery >> Radio >> Transmission line >> Antenna >> Medium, the impact of the medium on the radio.

v  Node Design optimization considering the trade-offs around power consumption, longevity, size, cost, complexity etc., Building an IoT device from scratch by integrating appropriate components.

v  Extensive exposure of dealing with electronic circuits in the lab as well as in the field in extreme environments troubleshooting issues related to power, component failure, clock drift, EMI, interference, integrity issues, connectors, packaging etc.

v  Network topology planning, adaptive routing algorithms, Lifetime analysis, communication between cellular-gateway and cloud-server, over the air-updates.

Notable Outputs:

v  Sensing without Sensors: by monitoring the radio itself.

v  Saving idle state energy drainage (> 70%) by using a boot-up timer.

v  SleepSense: ultra-low-power ambient and accumulative sensing in a sleep state by replacing Crystal Oscillator with a customized sensor-coupled oscillator.

 

Remote Energy Metering System

ThinkClock Innovation Labs

Time: Sep. 2009-July 2011

Role: Technology Lead

Skills: MCU: Microchip PIC8F8722, GSM/GPRS: SIMCOM 300 module, ZigBee: Microchip MRF24J40MA, Metering IC MCP3909, Microchip IDE and C18 Compiler

 

Project: involves analog instrumentation of current & voltage sensing with temper protection mechanism and calculation of energy consumption in the load using MCU. Data from sensors of each phase are multiplexed into one SPI port of MCU. The controller MCU is interfaced with GSM/GPRS module through UART port and ZigBee & ethernet module through SPI ports. The energy consumption data is communicated to a central server (listening TCP, UDP or HTTP) periodically or on-demand through GSM/GPRS or pluggable ethernet in combination with ZigBee wireless mess network. A mechanism for two way communication with the meter is being developed using BOSH/SIP protocols.

 

Automotive Telematics System for Remote Tracking & Monitoring

ThinkClock Innovation Labs

Time: July 2009-July 2011

Role: Technology Lead

Skills: Java: J2SE, Servlet, JSP, J2ME, MySQL, SQL Toad, XML, Google Maps API: static/dynamic, Eclipse J2SE/J2EE, EclipseME, Mobile Tools for Java (MJT), JUnit, J2MEUnit, Sun Wireless Toolkit, Apache Tomcat, JDBC, Java SMS

 

My role: As a cofounder of the company, I was leading the product development team consisting of 7 people in close coordination with the marketing team. The project involved following modules.

(1). Automotive Telematics Unit: This H/W device was based on Siemens XT65 GPS/GPRS module with Java IMP profile.

(2). Telematics Engine: This is a S/W component running on the server responsible for receiving, parsing, interpreting the data using GIS coming from Automotive Telematics Unit via the cellular network. It supports TCP, UDP, HTTP, SMTP and SMS protocols for receiving data packets.

(3). User Presentation Module: This is a client-side S/W module using which user can connect to the Telematics Engine and locate & track vehicle on the GIS map with complete history. It is a Web-based Tracking client for PC users: a servlet fetching Google Maps using Javascript.

 

SIP Mediator Client for routing audio/video from IP Camera to Asterisk PBX Server

ThinkClock Innovation Labs

Time: June. 2008-July. 2009

Role: Software Developer

Skills: Linux, Asterisk PBX Server, SIP, codec conversion, Ekiga SIP Client, TCP/IP, UDP, HTTP

 

The project: involved routing of audio/video data from an IP Camera (acting as a TCP/UDP/HTTP server) to an  Asterisk PBX Server (Open source Telephony server compatible with SIP clients) by developing a Mediator SIP client on linux. The implementation was accomplished by me through inter-process communication (through shared memory access & semaphores) where one process retrieves audio/video data from IPCAM writing into a shared memory and other process fetching that data from shared memory to Ekiga SIP client (which routes to Asterisk). It also involved conversion of audio/video codecs because of compatibility reasons.

 

Software License Management System

SafeNet India (Now Gemalto)

Time: July 07-June 08

Role: Software Developer

Skills: C, C++, SQL, OpenSSL, Cryptographic Algo’s (AES, RSA, ECC, SHA-1, SHA-2, HMAC), Design Documentation (FDD)

 

The project: involved development of Software license management system targeted for software vendors to enable them to protect their applications using our client API and license management server.

I was responsible for design and development of Crypto-Layer to secure the communication between client and server in this project. During this project I explored and used various symmetric and asymmetric crypto-algorithms (AES, ECC, RSA, SHA, HMAC etc) and crypto tools like OpenSSL.

 

Hardware Implementation of ECC and RSA algorithms on FPGA

LESTER Labs, UBS France

Time: Jan 07-July 07

Role: Internship

Skills: C, C++, MS Visual Studio, Nios IDE, SOPC Builder, VHDL, OpenSSL, Cryptographic Algo’s (AES, RSA, ECC, SHA-1, SHA-2, HMAC)

 

This project: involved hardware implementation of ECC along with RSA to speed-up and enhance the security of public key crypto-systems using Altera FPGA and Nios II soft processor. Programming was done in C++ and then the code was built for target hardware (Altera FPGA) using Altera IDE. During this project, I have developed a new side-channel resistant ECC point multiplication algorithm. The computationally intensive part of the code (ECC point multiplication and addition) was implemented in hardware using VHDL.

 

Some Old Projects:

 

Collision Avoidance System for Smart Automobiles

[During 7th semester July'06-Dec'06, Won prize of 1 lakh INR  in Texas Instruments DSP design Contest 2006]

Guide: Prof. M. Radhakrishna, IIITA, Team: 2 members, Role: Project leader

 

Skills: MATLAB: simulink, real-time workshop, TI DSP C6713, Atmel MCU Atmega 169

 

System assists driver to have critical decisions through audio alerts during driving to avoid accidents. It provides information about distance and speed of vehicles in vicinity and also alerts when sees road hurdles hence making driving easy in traffic congestion and bad weather conditions. System can also function as black-box for accident analysis. System was implemented using TI DSP C6713, AVR Butterfly kit having ATmega 169 MCU and MATLAB DSP support.

 

Multiple Fundamental Frequency Estimation

[During 7th semester July'06-Dec'06, as an assigned work]

Guide: Prof. Michael Jordan, University of California at Berkeley, US, Team: alone

 

Skills: MATLAB: simulink, real-time workshop, audio signal processing

 

Project involves implementation of a state-of-art method for estimating the fundamental frequencies (F0) of several concurrent musical sounds using perceptually motivated method. The method consists of a computational model of the human auditory periphery, followed by a novel periodicity analysis mechanism. Estimation of multiple fundamental frequencies is achieved by canceling each detected sound from the mixture and by repeating the estimation for the residual.

 

ECC - Elliptic Curve Crypto-System

[During 7th semester July'06-Dec'06, as a course project]

Guide: Prof. R.C. Tripathi, IIITA, Team: 3 members, Role: Project leader

 

Skills: Java, Java Crypto-library, ECC, Sha-1, Sha-2

 

Elliptic Curve Cryptography is the next generation public-key cryptosystem for mobile/wireless environments. Compared to traditional cryptosystems like RSA, ECC offers equivalent security with smaller key sizes, which results in faster computations; lower power consumption, as well as memory and bandwidth savings. Project involves implementation of complete Crypto-System using three protocols ECDH, ECDSA and ECIES to speed up secure web transactions using JAVA library.

 

INFiANS - Intelligent Noise Filtering using Adaptive Noise Cancellation, Neural Networks and Blind Source Separation

[During 6th semester Jan'06-July'06, as a course Project, reached up to semifinal round of worldwide contest Windows Embedded Student Challenge 2006]

Guide: Prof. M. Radhakrishna, IIITA, Team: 4 members, Role: Project leader

 

Skills: MATLAB: simulink, real-time workshop, audio signal processing

 

This Project employs a mobile based solution for noise pollution control. It involves intelligent noise filtering techniques ANC, ANN and BSS to filter out unwanted sound signals and enables user…

 (a). To listen only sounds of interest ignoring crap sounds.

 (b). To be in complete peace in a noisy environment.

 (c). To have easy mobile conversation in crowded or cocktail party environment so that two side talkers will listen only each other’s voice irrespective of their environments.

 

A Kernel Level Overload Detection Mechanism for TCP Servers

[During May-June ’05 at CSE Department, IIT Bombay in KVPY summer program]

Guide: Prof. Varsha Apte, CSE, IITB, Team: alone

 

Skills: C, Linux (Debian/Suse), Linux Kernel, TCP/IP Layer, TCP/IP, UDP, HTTP

 

This project employs a kernel level overload detection mechanism for Web servers. Approach is based on SYN and Accept queues, which a TCP server maintains at TCP layer during three-way handshake. It also predicts the bottleneck resource (server machine or network). It was accomplished by modifying TCP/IP layer of the kernel code.

 

Sun Tracking System for maximizing the effectiveness of solar panel

[During 2nd semester Jan ’04 – June ’04, selected for IEEE INDICON-2004 Hardware Design Contest]

 

It is an electromechanical system which senses the direction of the Sun using 4-LDR sensors mounted on a cuboid and operates a DC motor to reorient the solar panel to keep it facing always towards Sun.

 

   Personal Information  

 

Full Name

 

Mr. Babu Devnarayan Choudhary

Father’s Name

Shri Kana Ram Saincha

Date of Birth

20th Aug, 1985

 

Current Address

 

Babu Devnarayan Choudhary

Department of Computing

Imperial College London,

South Kensington, London SW7 2AZ

 

Email: bdc@thinkclock.com 

b.choudhary@imperial.ac.uk

Phone: +44 (0)7517581192 ; +44 (0)7714051202

 

Permanent Address

 

Babu Devnarayan Choudhary  

S/o Shri. Kana Ram Saincha

Bera - Bharnda, Pipaliya Kalan

Teh. - Raipur [Marwar]

Dist. - Pali, Rajasthan, India

Pin - 306307

Phone - +91-2937-287191

 

Other Interests:

 

Volunteering for Social Care, Promoting Sustainable Agriculture in India, Yoga-Meditation

 

Co-Founder: Mera Gaon (https://www.facebook.com/meraga0n/)

 

This is a social engineering project which aims to raise the sensitivity and awareness of people towards our environment. We are developing aeroponic trees for roof-top gardens, with complete digital automation, which are cloud-connected and self-managed. This is an attempt to bring plants closer to people so that they can develop an intimate connection with them in their hearts and minds.

 

I hereby certify that all the information provided above is true to the best of my knowledge.

 

-Babu Devnarayan Choudhary