Distributed Multi-task APA over Adaptive Networks Based on Partial Diffusion

Distributed multi-task adaptive strategies are useful to estimate multiple parameter vectors simultaneously in a collaborative manner. The existed distributed multi-task strategies use diffusion mode of cooperation in which during adaptation step each node gets the cooperation from it neighborhood nodes but not in the same cluster and during combining step each node combines the intermediate estimates of it neighboring nodes that belong to the same cluster. For this the nodes need to transmit the intermediate estimates to its neighborhood. In this paper we propose an extension to the multi-task diffusion affine projection algorithm by allowing partial sharing of the entries of the intermediate estimates among the neighbors. The proposed algorithm is termed as multi-task Partial diffusion Affine projection Algorithm (multi-task Pd-APA) which can provide the trade-off between the communication performance and the estimation performance. The performance analysis of the proposed multi-task partial diffusion APA algorithm is studied in mean and mean square sense. Simulations were conducted to verify the analytical results.