| Input: original cover-document , logo-watermark , fixed |
| set-size, , with = 2 (an inner-parameter). |
| Output: watermark key data-sequence, , based on |
| aggregation of (DAC) and second key generated by a |
| logical operation on and |
| (1) Convert to a character-stream, . |
| (2) Make a copy-document of the original cover-document, . |
| (3) Divide into -word sets according to the set-size . |
| (4) for = 1 to |
| get th word-set from . |
| add the Unicode values of all characters in each |
| word of set- to produce and |
| get next embedding-character e from |
| add the Unicode value of e to , and convert |
| and to binary values. |
| while bits left in to process do |
| perform XOR-classification using logical- |
| bitwise operations on the th bit-position of |
| and to generate an () result. |
| aggregate all () results produced by |
| passing through each of th bits (of first and |
| second words in current set) in turn. |
| end while |
| aggregate all bit-sequences produced by combining |
| result for each of the previous word-sets, into . |
| end for |
| (5) Obtain the first key-generated, , as the result of |
| aggregation of produced for all word-sets. |
| (6) Perform logical AND-operation between the and |
| to generate the second-key, . |
| (7) Output and for the whole document. |
