A novel approach (non-cryptography) for SECURE storage

                               Of data on External media and Lossless Retrieval

Executive Summary:

 

We use the technique of encoding, obfuscation and a novel approach to solve the problem using non-cryptography technique so that management of key is not required, we will come up with simple and yet robust logic so that the data is secured and easily convertible.

Any approach to secure the data needs the data to be unrecognizable by jumbling or tampering them with the logic that’s quite tough to break but its always risk to use a single logic/algorithm.

So we come up with the solution through which we would divide the data into ‘N’ chunks that would be decided upon the size of data and then we would select randomly ‘M’ number of algorithm from the ‘Z’ number of available one so that it literally impossible to break the whole data. Due to the fact that the data is collection of different chunks and each chunk will be using different encoding way, hence the data will be totally jumbled and hence secured.

 

Introduction:

The challenge of securing the data is critical one and if we don’t need to use the key or any cryptography technique then we need to use some way to manipulate the data in our application in such a way that the data is encoded differently each time with different approach. we need to select the solution which handles speed, length and security perfectly and hence considering the criticality we suggest to use chunk encoding technique where the logic will be not disclosed.

 

Methodology:

           We would create the Application which will have ALGORITHM POOL (may be 15-20), now this could be a single Algorithm with different parameters or all together different algorithms. Any storage to External device will be done via that application and to get the data back we need that application.

Steps involved

Encoding (when data is stored on drive) by our application

1. Data is divided into different chunks depending on the size of Data and is stored onfirst chunk.
2. Now each chunk is encoded with the random algorithm from the pool which will be defined as number or an id to further decode it
3. Data is totally jumbled in such a way that the same word will be different on different chunk of data as different algorithm is been used.

 Decoding (when data is extracted from the drive) by our application

1. When data is passed on algorithm then the number of chunk is identified.
2. Each chunk has the number or id for the application through which the data is encoded

We will do some research on whether to go for available algorithm or create a new one,

Whether we should use single algorithm with different parameters or different algorithm

This would be the taska and depending upon the complexity of the problem we can select the algorithm.

Just for simplicity an example:

   Approach 1

We use an algorithm where the vowels are converted to different alphabets

Parameter 1

A -> w

E -> ^

I ->  *

O -> h

U -> 2

 

Parameter 2

 

A -> r

E -> &

I ->  4

O -> 7

U -> _

     

      .

      .

      .

      .

      .

Parameter 15

 

 

 

So This will be our ALGORITHM POOL.

 

We have Data, which would be divided into chunk and with the above logic say for example

Parameter 1 is used , then in this case

 

 

 Every vowels will be converted to different syllables as defined above

So for a paragraph having a word ‘Rakesh’ will be changed to ‘Rwk^sh’ and the same word on other chunk would be, if applied parameter 2 logic then

‘Rakesh’ -> ‘Rrk&sh’ and on other paragraph its value will changed hence it would be totally jumbled.

 

We will try to select the correct Algorithm so that speed length and security is at optimal.