Abstract
In this paper, a new localization approach is proposed based on the concept of Diffusion LMS (DLMS) to mitigate node and link failures in Incremental LMS (ILMS). It takes advantage of the known locations of powerful anchor nodes to estimate the location of tremendous unknown agents/nodes in Wireless Sensor Network (WSN). In addition, new techniques are developed according to the idea of Variable Step-Size (VSS) for ILMS and DLMS algorithms to extremely enhance the performance and convergence rate. Furthermore, the theoretical performance in terms of Mean Square Deviation (MSD) is derived and evaluated for DLMS and VSS-DLMS Algorithms. Simulation results show that the proposed VSS-ILMS and VSS-DLMS algorithms outperform the existing approach. From the Mean Square Error (MSE) point of view, they also have advantage of faster convergence in compare to the traditional ILMS and DLMS algorithms.
| Original language | English |
|---|---|
| Title of host publication | 2019 2nd IEEE Middle East and North Africa COMMunications Conference, MENACOMM 2019 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| ISBN (Electronic) | 9781728136875 |
| DOIs | |
| State | Published - Nov 2019 |
Publication series
| Name | 2019 2nd IEEE Middle East and North Africa COMMunications Conference, MENACOMM 2019 |
|---|
Bibliographical note
Publisher Copyright:© 2019 IEEE.
Keywords
- agents
- anchor nodes
- diffusion least mean square
- incremental least mean square
- localization
- variable step-size
- wireless sensor network
ASJC Scopus subject areas
- Information Systems and Management
- Safety, Risk, Reliability and Quality
- Signal Processing
- Computer Networks and Communications
- Hardware and Architecture