The Reordering Technique

(Ema_1, Unmasking/Decryption) – Reordering is the opposite of shuffling. The only difference is that the Reordering must start from the end of the masked byte block and the last index of the key.

Below is the reordering technique (unmasking) applied to the previous shuffling technique (masking) example.

KEY & DATA INDEX	KEY VALUES	DATA BEFORE Reordering	Reordering (UNMASKING)																DATA RESULT
000	006	d													g			a	a
001	007	l		h													b		b
002	015	e												p		c			c
003	000	f						g							d				d
004	002	b								p				e					e
005	009	c	j										f						f
006	013	n										a						g	g
007	008	i									b						h		h
008	004	p								b	i								i
009	010	k							f				j						j
010	003	g						f	k										k
011	012	a			m		l												l
012	014	o				l	m												m
013	011	m			a							n							n
014	001	h		l		o													o
015	005	j	c													p			p
			015	014	013	012	011	010	009	008	007	006	005	004	003	002	001	000

Example of reordering in unmasking/decrypting stage

1. key(15)=5, j → data(5), c → data(15)
2. key(14)=1, h → data(1), l → data(14)
3. key(13)=11, m → data(11), a → data(13)
4. key(12)=14, o → data(14), l → data(12)
and so on.

As you can see from the example, at the end of the Reordering we can reconstruct the original order (the rightmost element for each row). Unlike the other techniques, in reordering we start from the end of the block and the end of the key.

From a computational point of view, all that is required is to overwrite the elements of an array, and the speed of execution is the same as for the shuffling technique.

Note
We have seen how, at a theoretical level, the masking and unmasking steps work to encrypt and decrypt a data source with Ema_0 and Ema_1, the techniques of replacement and substitution and shuffling and reordering are completely independent.
Ema_0 can be used for all types of data, especially for text files or data sources that are easily recognisable.
Ema_1 can be used for all types of data, but for plain text files, it should first be compressed or masked by Ema_0.
EMA uses both, Ema_0 and Ema_1 techniques at the same time.