To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Geo-Location with Wireless Sensor Networks using Non-linear Optimization |
|
Author |
Abdellah Chehri, Paul Fortier, Pierre-Martin Tardif |
| Citation |
Vol. 8 No. 1 pp. 145-154 |
|
Abstract |
In this paper we propose a Reduced-Complexity Differential Evolution (DE) for localization of multi-hop sensor networks. In relatively smaller population size of DE algorithm (Np<50). We study scenarios where the nodes do not receive signal from anchors directly (i.e., multi-hop). The nodes with unknown positions use a new algorithm based on non-linear least squares (NLLS) method for deriving their estimated location. We consider the problem of minimizing a cost function based on distance measurements between these nodes and some definite anchors. We use Reduced Complexity Differential Evolution (RCDE) to find the global minimum of this cost function which corresponds to the nodes estimated coordinates. The main characteristic of the DE algorithm is that it can get close to the optimal solution with low complexity, if the steps of the algorithm are designed appropriately. In this paper we show that the DE optimization method can be used in ad hoc sensor networks to estimate the location of nodes accurately. The performance in terms of mean localization squared error is evaluated by simulations for different scenarios. The sensitivity of the estimated position to the uncertainty on distance measurements, radio transmission, and irregular area are also evaluated. |
|
Keywords |
Wireless Sensor Networks, Localization, Non-Linear Least Squares, Optimization, Differential Evolution (DE). |
|
URL |
