A Data Prediction in Wireless Sensor Networks using Deep Learning-based RSA Algorithm
Anand Dohare1, Tulika2, Sweta Sachan3, B.Mallikarjuna4
1Anand Dohare, Research Scholar, SHAUTS, University, Allahabad, India,
2Dr. Mrs. Tulika, Assistant Professor, Department of Computer Science and Information Technology, Allahabad, India.
3Mrs. Sweta Sachan, Department of Computer Science and Information Technology, Allahabad, India.
4Dr. Basetty Mallikarjuna, Department of Computer Science and Information Technology, Allahabad, India.
Manuscript received on June 22, 2020. | Revised Manuscript received on June 30, 2020. | Manuscript published on July 10, 2020. | PP: 398-404 | Volume-9 Issue-9, July 2020 | Retrieval Number: 100.1/ijitee.I7135079920 | DOI: 10.35940/ijitee.I7135.079920
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In wireless sensor networks (WSN) data collection and gathering data from surroundings and removing the redundancy, process the appropriate data is a challenging task, in spite of low battery, less memory space, low computational speed, reduce the energy consumption are major research areas in WSN. Temperature sensor, humidity sensors are majorly used to climate monitoring, agricultural, humidity observation, due to energy consumption of sensors, low battery power, computational speed of sensors are used to maintain a long time is a crucial issue. To overcome such type of problems data prediction techniques are required, several data prediction, aggregation techniques are proposed in this issue and several research has been done, but not solved all challenging issues. In this paper proposed deep learning-based RSA algorithm to provide security and efficiently handle the data by using a feed-forward filter to remove the aggregated data, Least Mean Square (LMS) variable step-size method to remove error rate that will improve the energy consumption and size of the memory space for data collection, the experimental results proved that 98% predicted data and minimum error rate on cluster network as per considered (Intel Lab) data set.
Keywords: Wireless sensor networks, Prediction, Aggregation, Rdundant, LMS.
Scope of the Article: Deep Learning