This paper deals with a network configuration considering priori estimation error variance and communication energy in sensor networked feedback system. Each sensor transmits information to a fusion center via multi-hop communication network. We propose a novel network configuration algorithm that each sensor node transmits information with a same dimension. The transmitted information is merged in each sensor node and consists of measurement data with time delay. Then the priori estimate can be minimum variance estimation. The proposed algorithm achieves sub-optimal network topology with minimum energy and a desired variance. Experimental results show an effectiveness of the proposed method.