Artificial Bee Colony DLMS Beyond Mean Square Error Boundary in Ad-hoc WSN

A new Diffusion Artificial Bee Colony (DABC) heuristic algorithm is developed to estimate system parameters in Wireless Sensor Networks (WSNs). The main contribution is the incorporation of ABC algorithm in the traditional Diffusion Least Mean Square (DLMS) algorithm which leads to a better Mean Square Error (MSE) performance. The DABC algorithm shows excellent convergence beyond the noise variance boundary. In the diffusion stage, each node shares the local best cost function and corresponding local best particle position to immediate neighboring nodes. The extensive simulations show that the proposed DABC approach achieves excellent MSE improvement (MSD deterioration) in comparison to existing DLMS algorithms.