Wireless sensor networks based multidimensional scaling position method

Mobile station (MS) localization which plays an important role in the Wireless Sensor Networks (WSNs) has received considerable attention. In this paper, a new framework based on subspace approach for positioning a MS at WSNs localization system with the use of time-of-arrival (TOA) measurements is introduced. Unlike ordinary multidimensional scaling (MDS) algorithm using eigendcomposition or inverse computation to estimate the MS position, a computationally simple weighting estimator is proposed by introducing lagrange multiplier and Mean square error (MSE) weighting matrix. Computer simulations are included to corroborate the theoretical development and to contrast the estimator performance with several conventional algorithms. It is shown that the new method with low computational complexity achieves a comparable localization performance for zero-mean white Gaussian range error at moderate noise level.