AU2020102637A4 - A technique for traffic prediction and congestion control in iot networks using machine learning - Google Patents

A technique for traffic prediction and congestion control in iot networks using machine learning Download PDF

Info

Publication number
AU2020102637A4
AU2020102637A4 AU2020102637A AU2020102637A AU2020102637A4 AU 2020102637 A4 AU2020102637 A4 AU 2020102637A4 AU 2020102637 A AU2020102637 A AU 2020102637A AU 2020102637 A AU2020102637 A AU 2020102637A AU 2020102637 A4 AU2020102637 A4 AU 2020102637A4
Authority
AU
Australia
Prior art keywords
congestion
traffic
congestion control
machine learning
technique
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
AU2020102637A
Inventor
Vijay Anant Athavale
Pradeep Bedi
Paladuga Satish Rama Chowdary
Kanwalvir Singh Dhindsa
Sukhpreet Kaur
Sushil Kumar
Yogesh Kumar
Makhan Kumbhkar
Sandeep Sharma
Rohit TRIPATHI
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chowdary Paladuga Satish Rama Dr
Dhindsa Kanwalvir Singh Dr
Kaur Sukhpreet Dr
Sharma Sandeep Dr
Tripathi Rohit Dr
Original Assignee
Chowdary Paladuga Satish Rama Dr
Dhindsa Kanwalvir Singh Dr
Kaur Sukhpreet Dr
Sharma Sandeep Dr
Tripathi Rohit Dr
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chowdary Paladuga Satish Rama Dr, Dhindsa Kanwalvir Singh Dr, Kaur Sukhpreet Dr, Sharma Sandeep Dr, Tripathi Rohit Dr filed Critical Chowdary Paladuga Satish Rama Dr
Priority to AU2020102637A priority Critical patent/AU2020102637A4/en
Application granted granted Critical
Publication of AU2020102637A4 publication Critical patent/AU2020102637A4/en
Ceased legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/012—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from other sources than vehicle or roadside beacons, e.g. mobile networks
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0133—Traffic data processing for classifying traffic situation
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G08G1/0141—Measuring and analyzing of parameters relative to traffic conditions for specific applications for traffic information dissemination
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/12—Messaging; Mailboxes; Announcements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/80—Services using short range communication, e.g. near-field communication [NFC], radio-frequency identification [RFID] or low energy communication
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W84/00—Network topologies
    • H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks

Landscapes

  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Traffic Control Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A TECHNIQUE FOR TRAFFIC PREDICTION AND CONGESTION CONTROL IN IOT NETWORKS USING MACHINE LEARNING ABSTRACT At present, all the real time applications in the world are deployed with smart technology using internet of things. Particularly in transportation system, any congestion in the traffic affects the transportation of goods, medically equipped vehicles, tourist logistics management, etc. For unmanned vehicle, 5G network with wireless sensor deployed for communication of data, especially in internet of things enabled smart city. The data input are observed by sensors of highly configured cameras, mobile devices and radio frequency identification tags. The information are grouped as datasets, for example, vehicle number, speed of all the vehicles, timings, paths and current congestion status. After grouping of datasets, it is deployed with machine learning algorithm. A supervised learning involving statistics, logistic regression model which is a classification algorithm is applied to predict the probability of target parameter. A threshold set can be used to observe the probability and classified into classes with logistic regression model. This prediction model for traffic analysis provides good accuracy and efficiency involving smart technology in the city managing the transportation system. It predicts the traffic congestion in advance and congestion control is performed from the notification received in smart devices at end user. 1| P a g e A TECHNIQUE FOR TRAFFIC PREDICTION AND CONGESTION CONTROL IN IOT NETWORKS USING MACHINE LEARNING Drawings NO OF VEHICLES ROAD CAMERAVEHICLE NO DATA COLLECTION SPEED WIREESS - SENSORS 5G NETWORK) TIME PATHS CURRENT STA~t" OF1 CONGESTION Fig. I DATA COLLECTION

Description

A TECHNIQUE FOR TRAFFIC PREDICTION AND CONGESTION CONTROL IN IOT NETWORKS USING MACHINE LEARNING
Drawings
NO OF VEHICLES ROAD CAMERAVEHICLE NO DATA COLLECTION SPEED
WIREESS - SENSORS 5G NETWORK) TIME
PATHS
CURRENT STA~t" OF1 CONGESTION
Fig. I DATA COLLECTION
A TECHNIQUE FOR TRAFFIC PREDICTION AND CONGESTION CONTROL IN IOT NETWORKS USING MACHINE LEARNING
Description
Field of the Invention.
The Field of invention is related to machine learning approach to train data sets collected from various sensors so that traffic congestion can be predicted in advance and congestion control deployed in IOT networks.
Background of the invention.
At present, all the city locations are deployed with smart technology enabled with internet of things. This has paved way for advancement of techniques to find solution of commonly faced problems every day around us.
Traffic is the most important one that has to be analyzed and controlled in the real time. If the prediction and control is inefficient, it may degrade the transport management system of the city. Even today and earlier, manual traffic control that has being in use have many drawbacks affecting the common people to reach the destination in right time. In addition, even the travellers who may be on a business trip or on tourism, the logistics cannot manage and provide visits to destined place at appropriate time.
At present, all the real time applications are deployed with smart technology using internet of things. By prior prediction of the traffic prediction, the congestion control can be performed on iot networks. The information has to be collected fast, processed quicker, analyzed and made predictions soon. 5G networks are more suitable to enable faster action for any real time response when it has to be handled critically. Especially for unmanned aerial vehicle that involves recent mobile technology need 5G wireless communication for internet of things for managing traffic congestion.
As it involved 5G networks for wireless sensor networks, if congestion control is not efficiently handled, it may cause delay and packet loss. So efficient method of controlling congestion has to be made to have low delay and high speed of data transfer.
The data sources at all angles in all the roadmaps have to be gathered. There is a necessity to deploy high end cameras with good configuration, mobile devices and radio frequency identification tags for data collection. It has many datas or the information and has to be grouped
11 P a g e as datasets, for example, vehicle number, speed of all the vehicles, timings, paths and current congestion status.
As for the traffic prediction in advance, though datasets are grouped, there is a need to deploy an efficient algorithm. Initially, neural networks was deployed, but it consumes more time for prediction. It was overcome by using machine learning algorithm as a technique for traffic prediction and congestion control in internet of things networks.
For training and modeling of the datasets, different techniques in machine learning though can be deployed, logistic regression found to be having a highest recall and accuracy for the real time traffic prediction and congestion control.
Using SCTP approach with decision tree in machine learning algorithm also provides less accuracy than the logistic regression technique for traffic analysis. Compared to random forest, support vector machine and multilayer perceptron, logistic regression shows better results for traffic prediction and congestion control.
A logistic regression is a classification algorithm which is applied to predict the probability of target parameter. It is a supervised learning involving statistics. A threshold set can be used to observe the probability and classified into classes with logistic regression model. It is the most suitable and best prediction model for traffic analysis that can bring about good accuracy and efficiency involving smart technology in the city managing the transportation system.
Objects of the Invention
The first objective is to predict the traffic congestion status in advance in smart technology city with internet of things network using machine learning. The second objective is to deploy congestion control with machine learning. The third objective is to deploy logistic regression model to achieve efficient accurate analysis quicker.
Machine learning algorithm is deployed to analyse different datasets captured in different location and in different time. It requires a classification of the datasets to make prediction and control congestion. It is a need to determine and analyse the optimal parameter that were observed with mobile sensors. The statistical analysis is required to have efficient and accurate results. So when compared to random forest, support vector machine and multilayer perceptron, logistic regression shows better results for traffic prediction and congestion control. A threshold is defined prior as test set can be used to compare with trained set to observe the probability of the target parameter and classified into classes with logistic regression model. It is the most suitable and best prediction
21Page model for traffic analysis that can bring about good accuracy and efficiency involving smart technology in the city managing the transportation system.
Summary of the Invention
Transportation management has become importance as it is one of the key monitoring in the smart technology city. Any congestion in the traffic affects the transportation of good, medically equipped vehicles, tourist logistics management, etc. For unmanned vehicle, 5G network with wireless sensor deployed for communication of data, especially in internet of things enabled smart city. The data input are observed by sensors of highly configured cameras, mobile devices and radio frequency identification tags. It has many information and has to be grouped as datasets, for example, vehicle number, speed of all the vehicles, timings, paths and current congestion status. After grouping of datasets, it is deployed with machine learning algorithm. A logistic regression model is a classification algorithm which is applied to predict the probability of target parameter. It is a supervised learning involving statistics. A threshold set can be used to observe the probability and classified into classes with logistic regression model. It is the most suitable and best prediction model for traffic analysis that can bring about good accuracy and efficiency involving smart technology in the city managing the transportation system. It predicts the traffic congestion in advance and congestion control is performed from the notification received in smart devices at end user.
Detailed Description of the Invention
Fig. 1 illustrates the data collection of traffic management Fig. 2 illustrates the process flow block diagram of machine learning deployed in traffic management IOT networks.
Detailed Description of the Invention
The fig.1 illustrates the data collection in deploying the technique to predict traffic in advance and congestion control in internet enabled smart technology city. Traffic which is one of the most important management system that has to be analyzed and controlled in the real time. If the prediction and control is inefficient, it may not only degrade the transport management system of the city but also lead to many losses. Even today and earlier, manual traffic control that has being in use have many drawbacks affecting the common people to reach the destination in right time. Particularly in transportation system, any congestion in the traffic affects the transportation of goods, medically equipped vehicles, tourist logistics management, etc. In addition, even the travellers who may be on a business trip or on tourism, the logistics cannot manage and provide visits to destined place at appropriate time. For unmanned vehicle, 5G network with wireless sensor deployed for communication of data, especially in internet of things enabled smart city. The
31Page data input are observed by sensors of highly configured cameras, mobile devices and radio frequency identification tags. Later, the information has to be grouped as datasets, for example, number of vehicles on road, vehicle number, speed of all the vehicles, timings, paths and current congestion status.
Fig.2 illustrates the process flow block diagram of the traffic prediction and congestion control in internet of things enabled city with machine learning algorithm. Transportation management has become importance as it is one of the key monitoring in the smart technology city. Any congestion in the traffic affects the movement of goods, medically equipped vehicles, tourist logistics, etc. At present, all the real time applications are deployed with smart technology using internet of things. By prior prediction of the traffic prediction, the congestion control can be performed on iot networks. The information has to be collected fast, processed quicker, analyzed and made predictions soon. 5G networks are more suitable to enable faster action for any real time response when it has to be handled critically. 5G network with wireless sensor deployed for communication of data in internet of things enabled smart city. The data sources at all angles in all the roadmaps have to be gathered. The data input are collected by sensors of highly configured cameras, mobile devices and radio frequency identification tags. It has many information and has to be grouped as datasets, for example, vehicle number, speed of all the vehicles, timings, paths and current congestion status. After grouping of datasets, it is deployed with machine learning algorithm. A logistic regression model is a classification algorithm which is applied to predict the probability of target parameter. It is a supervised learning involving statistics. A threshold set can be used to observe the probability and classified into classes with logistic regression model. It is the most suitable and best prediction model for traffic analysis that can bring about good accuracy and efficiency involving smart technology in the city managing the transportation system. The machine learning deployed is trained and modeled with datasets, so that it can make predictions that have a highest recall and accuracy for the real time traffic prediction and congestion control. It predicts the traffic congestion in advance and congestion control is performed from the notification received in smart devices at end user.
41Page

Claims (2)

A TECHNIQUE FOR TRAFFIC PREDICTION AND CONGESTION CONTROL IN IOT NETWORKS USING MACHINE LEARNING CLAIMS I/We Claim:
1. Traffic prediction and congestion control performed in internet of things enabled smart technology city. 2. 5G networks for the communication of data, processing and performing machine learning algorithm. 3. Data collected or captured with unmanned aerial wireless sensors with highly configured cameras, mobile devices and radio frequency identification tags to capture the information namely number of vehicles on road, vehicle number, speed of all the vehicles, timings, paths and current congestion status. 4. A high configured computer with high end processing systems equipped to perform the process of traffic congestion control. 5. Grouping of datasets performed to train the machine to model and make predictions. 6. Logistic regression a supervised learning approach involving statistics, which is a classification algorithm is applied to predict the probability of target parameter by classifying the classes. 7. Smart devices to receive the notification of the status of the congestion in advance and perform the congestion control.
1 Pag e
A TECHNIQUE FOR TRAFFIC PREDICTION AND CONGESTION CONTROL IN 08 Oct 2020
IOT NETWORKS USING MACHINE LEARNING
Drawings 2020102637
Fig. 1 DATA COLLECTION
Fig.
2 PROCESS FLOW BLOCK DIAGRAM
AU2020102637A 2020-10-08 2020-10-08 A technique for traffic prediction and congestion control in iot networks using machine learning Ceased AU2020102637A4 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU2020102637A AU2020102637A4 (en) 2020-10-08 2020-10-08 A technique for traffic prediction and congestion control in iot networks using machine learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
AU2020102637A AU2020102637A4 (en) 2020-10-08 2020-10-08 A technique for traffic prediction and congestion control in iot networks using machine learning

Publications (1)

Publication Number Publication Date
AU2020102637A4 true AU2020102637A4 (en) 2020-11-26

Family

ID=73458055

Family Applications (1)

Application Number Title Priority Date Filing Date
AU2020102637A Ceased AU2020102637A4 (en) 2020-10-08 2020-10-08 A technique for traffic prediction and congestion control in iot networks using machine learning

Country Status (1)

Country Link
AU (1) AU2020102637A4 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113271569A (en) * 2021-06-09 2021-08-17 南京万般上品信息技术有限公司 Intelligent vehicle-mounted communication method based on 5G mmWaves
CN113726557A (en) * 2021-08-09 2021-11-30 国网福建省电力有限公司 Network transmission control optimization method based on flow demand
CN117692917A (en) * 2024-02-01 2024-03-12 苏州抖文信息科技有限公司 Relay control system for wireless communication network

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113271569A (en) * 2021-06-09 2021-08-17 南京万般上品信息技术有限公司 Intelligent vehicle-mounted communication method based on 5G mmWaves
CN113726557A (en) * 2021-08-09 2021-11-30 国网福建省电力有限公司 Network transmission control optimization method based on flow demand
CN117692917A (en) * 2024-02-01 2024-03-12 苏州抖文信息科技有限公司 Relay control system for wireless communication network

Similar Documents

Publication Publication Date Title
AU2020102637A4 (en) A technique for traffic prediction and congestion control in iot networks using machine learning
US10451712B1 (en) Radar data collection and labeling for machine learning
JP6709283B2 (en) Detection and analysis of moving vehicles using low resolution remote sensing images
Peixoto et al. A traffic data clustering framework based on fog computing for VANETs
Banerjee et al. A survey on IoT based traffic control and prediction mechanism
Alipour-Fanid et al. Machine learning-based delay-aware UAV detection over encrypted Wi-Fi traffic
Ikiriwatte et al. Traffic density estimation and traffic control using convolutional neural network
WO2020185209A1 (en) Radar data collection and labeling for machine-learning
US20210357767A1 (en) Automated knowledge infusion for robust and transferable machine learning
Li et al. A transfer double deep Q network based DDoS detection method for internet of vehicles
Alonso et al. Predicting flight departure delay at Porto Airport: A preliminary study
US11165648B1 (en) Facilitating network configuration testing
Rashid et al. SEIS: A spatiotemporal-aware event investigation framework for social airborne sensing in disaster recovery applications
Ferdous et al. Intelligent traffic monitoring system using VANET infrastructure and ant colony optimization
CN114663739A (en) Environment-specific model delivery
Salahdine et al. Short-term traffic congestion prediction with deep learning for lora networks
Borah Detecting Background Dynamic Scenes using Naive Bayes Classifier Analysis Compared to CNN Analysis
Zhang et al. Edge-assisted learning for real-time UAV imagery via predictive offloading
Pasero et al. Artificial neural networks to forecast air pollution
Chow et al. Flare: detection and mitigation of concept drift for federated learning based IoT deployments
Sepulcre et al. Exploiting context information for estimating the performance of vehicular communications
Baruah et al. Online learning and prediction of data streams using dynamically evolving fuzzy approach
Sehrawat et al. Performance Evaluation of Machine Learning Algorithms applied in SD-VANET for Efficient Transmission of Multimedia Information
US20240037419A1 (en) Accuracy of multivariate approach for time-series based forecasting
US20230188936A1 (en) Locationing System that Utilizes Read Data of a Reader of a Device

Legal Events

Date Code Title Description
FGI Letters patent sealed or granted (innovation patent)
MK22 Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry