Claims:The scope of the invention is defined by the following claims:

Claim:
1. A system/method for enhancing the security of sensors data by prediction mechanism, said system/method comprising the steps of:
a) The system should collect the data from the different sensors (1), thought the application (2).
b) The system stores all the encoded data in the database (3).
c) From the encoded data, we are going to analyze (4) the data whether it is modified or not in the data transmission.
d) Finally, we are going to perform the prediction mechanisms (5) from the analysis.
2. As per claim 1, the system collects many information from the sensors and perform the encryption mechanism and stored it on the database.
3. As mentioned in claim 1, the encoded data are analyzed and predict the data whether it is modified or not. , Description:Field of Invention
The present invention relates to, providing security to the sensors which are being used for IoT applications.
Background of the Invention
The objective of this invention is to improve the security for the sensors and make some prediction with the data received from the sensors. Whenever the data is being transmitted from a sensor to an application there will be some threat that takes place in that situation. In order to eradicate such type of things, our patent is being modelled. We will be defining some real time examples and relate them with the current invention. Let’s take some simple examples like automated dustbin. The dustbin will open automatically whenever a person goes closer to it and closes automatically when that person goes far from the bin. Each and every time the proximity sensors that are attached to the bin will send data to the application in the form of ‘0’ and ‘1’. With the help of that input, the bin will open and close [US2014/8682034B2].
If we have the data of ‘0’ and ‘1’, that means we can get to know that how many times the bin is opening and closing per day and at which timings it is opening. Generally, in a home bin are used to dispose waste materials. Humans on an average dispose food item at a percentage of 70% among the complete disposal garbage. With the information which we get from the sensors, there is a possibility of attacking a person with the help of some food items. No one thinks that, the attacks may be done in this way also. This is one of a life threat issue that need to be look into. There is no data analysis or prediction techniques that are being applied on this type of data.
There is already a work which is done on security to an IoT system in which security is provided to the application in order to not attacked by any intruders [US6930599B2]. And retrieving some video data from sensor as well is related in such a way of extracting the data only. With the help of providing security to that particular sensor we can secure the video data which is been transmitted from the sensor [US2013/0113932A1].
In (Chinnasamy et al [2020], Lecture Notes in Networks and Systems, 145, pp- 537-547), the authors are suggested to use Blowfish and ECC algorithm to enhance the security of the cloud computing. They have implemented this algorithm using JAVA based crypto libraries. The performance of the proposed method was analyzed in terms of encode and decoding time with (Chinnasamy et al [2018], 2018 Second International Conference on Inventive Communication and Computational Technologies (ICICCT), pp. 1717-1720). The (Chinnasamy et al [2018], Advances in Intelligent Systems and Computing, 145, pp-341-350), introduced the enhanced key generation method on RSA algorithms by using four prime numbers. They have used this algorithm to share the data in secure manner in health care applications.
Earlier there were many encryption algorithms that were used in order to control the theft of data. The data from the sensor will get transmitted to the application and there the data will get encrypted with some encryption algorithms, then used to get stored in the database. Here there are two challenges. The first one is attacking the database, stealing the encrypted data and decrypting the data manually with different methods. The second one is attacking the data while the data is getting transmitted by using some methods like man-in-the-middle attack. We are going to address the second attack which is being performed at the time of data transfer. In the first way some of the techniques like penetration testing, SQL Injection, etc may be used to penetrate into the database and copy the complete data that is present in the database. In the first type of attack the data is at least present in an encrypted form. That means in the first type, the risk percentage is upto 60%. But when it comes to second type of attack, the data is not in an encrypted form where the risk percentage will be increased.
Summary of the invention
Security will be provided to the data that is been extracted from the sensors. The security will be provided at two different stages, the first is the data that is getting generated at the sensors and the second is the analysis of the data that is getting generated. The data that is been generated at the sensors end will be redirected to analysis phase and with the result, we are going to perform prediction techniques. With the help of the result that is getting generated in the prediction phase will be used to train the future IoT devices.
Brief description of Drawing
In the figures which are illustrate about the invention.
Figure 1 Architecture diagram regarding the flow of work
Detailed description of the invention
In today’s day to day life, human life is becoming more and more machine dependent. Whenever there is a technology that is used, there the problems are being generated through the internet itself. Our complete proposal is about providing security to the machines that are linked with sensors and internet.

Security will be provided to the data that is been extracted from the sensors. The security will be provided at two different stages, the first is the data that is getting generated at the sensors and the second is the analysis of the data that is getting generated. The data that has been generated at the sensors end will be redirected to analysis phase and with the result, we are going to perform prediction techniques. With the help of the result that is getting generated in the prediction phase will be used to train the future IoT devices.

There will be a lot of problems as well as damages that can happen to the human life if we are not able to provide security to the machines that are connected to internet. Let’s see a scenario and go in depth explanation by relating that scenario to the patent. The production of self-drive cars is increasing day by day. The self-drive car runs on completely sensors and a specialized application. What are all the inputs that a sensor gets, it will transfer the same to the application. With that data only application is going to make a decision. If there is an obstacle, then the sensor passes the data to the application. Then application decides that whether to turn the car right side or left side.

We need to secure the data that is being generated by the sensor, data analysis and prediction need to performed on the data in order train future upcoming machines with this particular data. Providing security again has different scenarios. We are going to encrypt the data while the data is in transfer mode from sensor to the application. That data will get decrypted and go through data analysis and prediction phases. Here we are claiming the security of the data when it is under transmission and as well as prediction from the data that is being generated.

The output which is stored in the database is going to be an input in prediction phase. In the system memory phase, we are going to implement data analysis. After the analyzed data enters into the inference engine, then the prediction on top of that data is going to start. The main motive is to provide security to the data when the data is already under transmission phase. We don’t require any encryption techniques after the data reaches the database.

The encryption technique will be performed only when the data is in transfer mode itself, but not after the data gets transferred to the database. Attacking on the data can be performed while the data is getting transmitted by using some methods like man-in-the-middle attack. We are going to address the second attack which is being performed at the time of data transfer.
3 Claims & 1 Figure