Accuracy Improvement of SSSD Circular Positioning in Cellular Networks

Location determination of a mobile subscriber (MS) to support location-based services has become an area of highly active research over the past few years. The stationary signal-strength-difference (SSSD)-based positioning technique has been proposed as a cost-effective low-complexity solution. However, its accuracy is limited due to the error in reference distance estimation. This paper contributes a simple and efficient method to minimize the error in the reference distance and consequently reduce the error in MS location determination. The proposed method is based on a standard parameter separation least-square algorithm by following the convergences of the gradient-descent algorithm to determine the MS location. Accuracy evaluation is implemented at 55 different test locations in an urban environment under a commercial cellular network. Experimental results clearly demonstrated that the proposed method performs superior positioning accuracy as compared with our previous works and the trilateration positioning method.