Path Loss Exponent Estimation and RSS Localization Using the Linearizing Variable Constraint

Received Signal Strength (RSS) localization has the attraction of requiring only simple receiver hardware. The basic idea is to find the position estimate that is most consistent with the received signal power measurements observed at multiple known locations in the area of interest. A fundamental assumption is that the signal power measurement is proportional to the product of the reference emitter power, eta(0), and a distance dependent term of the form d(-(alpha) over bar), where d is the emitter to receiver separation and (alpha) over bar is the path loss exponent. In practice, (alpha) over bar depends on the environment and can vary over a large range. There is often very limited a priori knowledge of (alpha) over bar and eta(0). available and the use of inaccurate values in the localization equations can adversely affect the accuracy of position estimates. This paper develops a new method, based on a linearizing variable constraint, for the estimation of (alpha) over bar. The estimation of the emitter position and of eta(0). then follows directly. This estimator yields a performance similar to a two-stage weighted least squares scheme, but requires much less computation.