Dallas, Texas, United States Contact Info
6K followers 500+ connections

Join to view profile

About

Full Time Data Scientist with a demonstrated history of working in the higher education…

Activity

Join now to see all activity

Experience & Education

  • Walmart

View Mohit’s full experience

See their title, tenure and more.

or

By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.

Licenses & Certifications

Publications

  • Detection and Classification of Traffic Signs using Inception in ResNet Model

    IJMCS- ISST Journal of Mathematics & Computing System Vol.9 No.1 (Jan.–Jun. 2018)

    Other authors

  • Enhanced ADHOC Routing Algorithm for MANET based on intelligent selection

    IEEE

    Have got the results soon going to be published.

    So far, there have been several research papers and advance algorithms which targets the power consumption, energy factor, link failures, black & grey hole attacks on the network but very few have focused on the flooding and the congestion that is the main problem in AODV for not having the promising results as it degrades all the performance parameters in small as well large topologies and it further degrades it if the network is loaded…

    Have got the results soon going to be published.

    So far, there have been several research papers and advance algorithms which targets the power consumption, energy factor, link failures, black & grey hole attacks on the network but very few have focused on the flooding and the congestion that is the main problem in AODV for not having the promising results as it degrades all the performance parameters in small as well large topologies and it further degrades it if the network is loaded with the data packets and RREQ packets. Therefore, a need was felt to start a research on the reasons of the flooding in AODV and the ways it could be controlled. The second thing was the fairness that AODV does not exploit to avoid the link failures.
    So, by using the pre-existing knowledge from some advance algorithms and research papers we decided to target the flooding in AODV and came up with a very promising algorithm, where we used a cue of pre-existing algorithm to write our algorithm.
    So, we have created an algorithm, which will efficiently reduce the flooding and thereby, improves the throughput, packet delivery ratio, latency in network, increase the fairness in network, i.e. nodes will can make path from source to destination will also get chance to participate in the communication.
    Advantages

    1). Short overhead
    2). It can be used to unicast or multicast packets
    3). Easily handles the dynamic behavior

Courses

  • Automation

    -

  • Embedded Systems

    -

  • Nanodegree Program in Self Driving Car Udacity

    -

Projects

  • Human level Traffic Sign prediction and detection using SVM and CNN (Inception + Resnet model)

  • Blink n Match game on Altera FPGA

    SD card implementation on Altera development board and audio implementation for background effects in game.

  • Intelligent Chatbot

    -

    Sequence to sequence modelling using LSTM on TensorFlow framework to support intelligent communication (chatbot), based on social networks data. Achieved 85% accuracy of generating relevant reply.

  • Behaviour Cloning

    -

    This project was implemented in python, data was collected using unity car simulator and thereafter used the unity simulator and connected its to web sockets so that picture collected by camera on the roof of the car will be pass into the trained model in order to get decisive wheel angle in order to control the car.

  • Finding lanes on road for self driving car

    -

    Used python and open cv for finding lanes on road for self driving car. Work is done on test video of moving car.

  • Image Similarity Detection Tool - Python, Scipy, Matplotlib, Numpy, SSIM, RMS, PSNR

    -

    Design a tool by calculating average of Structural Similarity Index, Root Mean Square and Peak signal to noise ratio between 2 images and achieve the average match accuracy of ~94%.

  • Enhanced AODV Algorithm in NS-2 Simulator

    -

  • X-MAC RDC Algorithm in Cooja Simulator

    -

Languages

  • English

    -

  • Hindi

    -

More activity by Mohit

View Mohit’s full profile

  • See who you know in common
  • Get introduced
  • Contact Mohit directly
Join to view full profile

Other similar profiles

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Others named Mohit Punia

57 others named Mohit Punia are on LinkedIn

See others named Mohit Punia

Add new skills with these courses

See all courses