You might want to read this, to see how AES was evaluated:

http://mercury.webster.edu/aleshunas/COSC%205130/H-AES-Evaluation.pdf

Key size, vulnerability to known attacks, performance (computation time) with different types of data, flexibility of application to different platform types, randomness of cipher, cost, and so on. I think that in terms of strength, aside from key size, randomness is very important, even when encrypting repetitive plaintext.

This paper looks mostly at performance comparisons:

http://ibimapublishing.com/articles/CIBIMA/2009/600033/600033.pdf

The favorite these days still appears to be AES, and many of the others have been downgraded, such as DES. So we can expect AES to be scrutinized most, in the near future.

Not all cryptography algorithms rely on mathematical complexity. When quantum cryptography algorithms are analyzed, due to the different nature of the problem to be solved, different metrics are used (Eavesdropper equivocation, or success rate for example). The question posted is of very general type, so the spectrum of answers exists, without possibility for unique answer, given the state of the Cryptography science.

http://www.ijarse.com/images/fullpdf/1489208151_S2025ijarse.pdf

http://data.conferenceworld.in/ESM/P102-112.pdf

Read the following :-

CRYPTOGRAPHIC ALGORITHM METRICS

Norman D. Jorstad

Director, Technology Identification and Analyses Center

Landgrave T. Smith, Jr.

[email protected]

CRYPTOGRAPHIC ALGORITHM METRICS

ACKNOWLEDGMENTS

This paper is the work of a team composed of: Lt. Gen. Lincoln D. Faurer, USAF

(Retired), Mrs. Janice L. Freeman, Father M. Blake Greenlee, Mr. D. Richard Kuhn and Mr.

Charles W. Shadel. It is essentially a brief status report on a study in progress.

I.

PURPOSE

In the absence of generally accepted metrics in the public domain that could be used to

measure and specify cryptographic strength, a small working group agreed to explore the

possibility of developing an approach to cryptographic metrics

The purpose of this white paper is to report the results of the limited exploratory effort by

the above group that investigated the practicality of developing metrics for use in specifying the

strength of cryptographic algorithms. This paper only deals with a small sample of selected

symmetric cipher block encryption algorithms in the codebook mode and an asymmetric public

key algorithm. Other algorithms and cryptographic techniques for message integrity,

authentication, and digital signatures were not investigated. Only information available in the

public domain was used during this investigation

Some cryptosystems are not size-preserving, in particular the asymmetrical. Thence, the space overhead also ought to be incorporated as metrics if applicable.

Dear Ruaa;

The best and easiest way to evaluate the performance of any telecom. system is to correlate the data at both ends. check the Correlation chapter in any statistic book.

I think there are some measurement such as encryption time, correlation data

-The algorithm running time in terms of its upper, lower and average bound, the time for security element generation, time for encryption/decryption and signature ..etc.

-The space complexity, i.e the storage needs to store keys, and signatures, however regarding key size refer to NIST recommendations. generally, lightweight algorithms are preferable if the data not highly confidential.

You can evaluate an encryption algorithm by measuring its encryption and decryption time. one of the factors that may affect the above time is the key size.

Please check my papers:

1) A novel network security algorithm based on private key encryption

2)A hybrid network security algorithm based on Diffie Hellman and Text-to-Image Encryption algorithm

3)Secure network communication based on text-to-image encryption

4) Block cipher encryption for text-to-image algorithm

5) A novel network security algorithm based on encrypting text into a white-page image

6) Analyzing the efficiency of Text-to-Image encryption algorithm

7) An authenticated, secure, and mutable multiple‐session‐keys protocol based on elliptic curve cryptography and text‐to‐image encryption algorithm

and other papers on researchgate

Hope this being helpful