Prediction based data reduction and controlled transmission in WSN for weather forecasting

Efficient utilization of energy resources is always the challenge in the field of Wireless Sensor Networks (WSNs). Besides energy utilization, the integration of internet of thing (IoT) with WSNs cause the issue of spectrum utilization in WSNs. There is an exponential growth in the data, which may b

Project Title

Prediction based data reduction and controlled transmission in WSN for weather forecasting

Project Area of Specialization

Internet of Things

Project Summary

Efficient utilization of energy resources is always the challenge in the field of Wireless Sensor Networks (WSNs). Besides energy utilization, the integration of internet of thing (IoT) with WSNs cause the issue of spectrum utilization in WSNs. There is an exponential growth in the data, which may be hindered if no control over the data generation is taken. One solution for both problems could be that the part of the sensed data can be predicted without triggering transmissions and congesting the wireless medium.

Our project proposes an energy efficient technique by using predictive models to predict future data which controls the data transmission. Instead of transmitting sensed data at every instant, we are implementing data prediction algorithm which continues to output data until or unless sensed data is changed which halts the prediction system, transmits the data and updates the sink node.

The system will contain 2 transmitting nodes and 1 receiving node. The prediction algorithm is applied on both sides, firstly on transmission side which predicts the data and transmit if and only if when the input data crosses the predefined error bound.

Project Objectives

The main objectives of this project are given below:

• Implementation of data prediction algorithm on the WSN nodes:
  As the WSN nodes constantly perform transmission of data, they require a large source of energy to make them work for longer period of time. The battery of a node can provide limited amount of energy to run them until they die out and needs to be replaced. One of the solution could be achieved by implementing the data prediction algorithms which can save the unnecessary transmission between WSN nodes so they can work for longer period of time.
   
• Controlled transmission based on error bound between sensed data and predicted data:
  As the sensor of the WSN nodes collect data, the algorithm calculates the error between the sensed data and the predicted data and control the transmission based on those calculations.
   
• Energy conservation and increasing the life span of a WSN node:
  In the literature on the WSN system (M.Arun raja, V. Malathi,“An LMS Based Data Reduction Technique for Energy Conservation in Wireless Sensor Network (WSN)”,Int.Journal Computer Technology & Applications, Vol. 3, pp. 1569-1576, July-August 2012), energy is most important factor of the transmission system, methods to conserve the energy are also needed. Energy conservation by the help of data prediction algorithms to increase the life of WSN node is one of the main objective of this project.

Project Implementation Method

• Data acquisition system collects the data from the sensor nodes and stores it in the memory unit of the node with help of Data logger which keeps it in a format for the algorithm.
   
• Data prediction algorithm utilizes the collected data to predict future values (Data) and compare the predicted value with the actual sensed value to make a decision whether to send the collected data or not.
   
• Sensor nodes will transmit only those values (Data) that differs from the actual sensed value. If the predicted value differs from the sensed data, the actual sensed value is transmitted so that the link node(receiver node) can adjust its values according to the new values the sensor node acquired.
   
• The Sensor nodes repeats the process until both values are made equal in which case the sensor nodes stops transmitting and predicts the values which should be equal to the value acquired by the sensor.
   
• Prediction algorithm will be implemented on both nodes i.e. sensor nodes and sink node so that it remains in-sync with each other and report the changes in values in real-time.

Benefits of the Project

The benefits provided by our project are as follows:

• Life span of WSN node will be increased before replacing it as most WSN nodes due to constant transmission die out rather quickly.
   
• Battery is effectively utilized by the proposed system thus making it cost-effective.
   
• System’s database makes a record of sensed parameters and these parameters can be utilized in other systems as well.
   
• The factories and industries that require constant surveillance could be achieved as nodes will work for longer period of time and as the values rise above the danger limit, they will alert the receiver station immediately.
   
• This project can provide a guideline to those projects that require a Time-series implementation of data transmission. 

Technical Details of Final Deliverable

The final deliverable will be a hardware based system of “Prediction based controlled transmission between WSN nodes” which would consists of 2 nodes with sensor to extract parameters and 1 node as receiver station.

The sensor nodes will send the sensed data to the receiver station in case of sudden change in values and the receiver station will update its parameters accordingly. After the change in parameters, the receiver station will continue to predict the value until it changes again.

Along with the hardware application of the project, we will also compare the parameters obtain from different prediction algorithms determine which produced the best results with help of software.

Final Deliverable of the Project

HW/SW integrated system

Type of Industry

Energy , Others , Telecommunication

Technologies

Internet of Things (IoT), Others, Clean Tech

Sustainable Development Goals

Industry, Innovation and Infrastructure, Responsible Consumption and Production

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com