The optimal filtering problem for multi-sensor networked systems with unknown channel interferences and packet losses is studied. There are possible packet dropouts and unknown channel interferences during the data transmissions from the sensors to the filter. The phenomena of packet losses are described by a set of random variables satisfying Bernoulli distributions. The prediction values of the lost measurements are used as the compensations when packet losses occur. In the absence of any information about the unknown channel interferences, an optimal filter independent of unknown channel interferences is designed based on the linear unbiased minimum variance estimation criterion. The stability of the proposed filter is analyzed. A simulation example verifies the effectiveness of the proposed algorithm.