Skip to main content
. 2022 Dec 2;22(23):9404. doi: 10.3390/s22239404
Algorithm 3 Data Screening Algorithm On the Node
  •  1:

    Given d¯si from the server, i = 1 …C, where C is the number of categories.

  • Input: 

    A new target data:

  •  2:

    Obtain ftj after feature extraction.

  •  3:

    Perform inference.

  •  4:

    if Its confidence level is greater than γnodes then

  •  5:

       for each ftj do

  •  5:

      Calculate the distance between each ftj and the d¯si of the corresponding pseudo-label category.

  •  5:

      Update the average distance dtemp once.

  •  6:

      if Equation (9) is satisfied then

  •  7:

       Output: upload ftj.

  •  8:

      end if

  •  9:

     end for

  •  10:

    end if