Scribd
Upload
93 views10 pages

Machine Learning Prediction Approach To Enhance Congestion Control in 5G IoT Environment

Uploaded by

Rashika Khanna
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
93 views10 pages

Machine Learning Prediction Approach To Enhance Congestion Control in 5G IoT Environment

Uploaded by

Rashika Khanna
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 10

ML PREDICTION APPROACH

TO ENHANCE CONGESTION
CONTROL IN A 5G IOT
ENVIRONMENT
IP PROJECT; CSE B ROLL NUMBERS 48, 104, 116
BASE PAPER FUNDAMENTALS
• Decrease in the lifetime and energy of a WSN (wireless sensor network) causes packet loss, bandwidth
degradation and delays.
• There have been many methods proposed for congestion control like the scalable distributed cluster
based congestion control scheme.
• Congestion control influences diverse applications, such as event-based, continuous sensing, query-
driven, and hybrid applications.
• Studies developed a novel method (expert framework) to estimate the end-to-end round-trip time using a
machine learning approach.
• This reduces number of packets transmitted and increases throughput.
BASE PAPER FUNDAMENTALS
• Conventional congestion control mechanism are used to satisfy the demands of 5G Internet of Things
(IoT) nodes/sensors networks.
• The aim of applying the machine learning method to congestion control is to predict the optimal
parametric setting and path to send/receive data.
• In this work, a machine learning approach is utilized for prediction based on support vector regression
using two parameters, namely, file transfer history and the measurement of simple path properties.
• This study presents a new methodology for improving the congestion control mechanism based on the
machine learning decision tree (DT) approach.
• The Stream Control Transmission Protocol (SCTP) is used in this environment.
BASE PAPER FUNDAMENETALS
• The DT approach classifies data into smaller subsets and uses it to predict the nature.
BASE PAPER FUNDAMENTALS
• The congestion control algorithm works on the transport layer.
• Transmitting and drawing of messages, error detection, and the retrieval of misplaced messages, can be
achieved provided that the SCTP engages throughout this layer.
• A transmitter connected to the recipient receives messages about acknowledgements to predict and
observe cwnd and ssthread.
BASE PAPER FUNDAMENTALS-
ML APPLIED HERE
About this Project:
• This is an ongoing project where multiple algorithms are being researched on to provide the most
suitable application.
• One starting point for the given project is the congestion control in TCP networks.
• We can simulate what happens in a smaller scale using that approach.
• The machine learning codes for this particular part, in congestion control is any type of network is also
an ongoing practice with some codes having 92% probability in control like the DT C4.5.
TCP Congestion Control:
Considerations:
• To obtain the optimal congestion window management, there are multiple factors to be considered.
• These include high throughput, high queue size, high congestion window and low packet loss.
• Node readings are inserted into a table to predict using models.
• The decision graph which is a result of the given DT approach shows the best path for congestion
control.
• This method is highly scalable and it can provide a clear representation of all cases.
Requirements:
• Simulation Environment: Ns2
• Software Requirements: Ns2, xgraph, nsnam
• Hardware Requirements: Os: Ubuntu 16.04

You might also like