Speed Optimised AES-GCM N Rajitha #1 , R Sridevi *2 # Research Scholar, Department of Computer Science & Engineering JNTUHCEH, Hyderabad, Telangana, India 1 rajitha2k2@yahoo.co.in * Professor, Department of Computer Science & Engineering JNTUHCEH, Hyderabad, Telangana, India 2 sridevirangu@jntuh.ac.in Abstract—The Advanced Encryption Standard (AES) in Galois Counter Mode (GCM) has gained significant popularity as it is can be implemented in a parallelised and pipelined way particularly in the hardware. The multiplication operations in AES can be optimised using PCLMULQDQ as in the case of Intel Processors and also using Karatsuba Algorithm. This paper illustrates the existing works on optimisations of AES-GCM using Karatsuba Offman Algorithm. I. INTRODUCTION Reconfigurable computing involves the use of reconfigurable devices, such as the field programmable gate arrays (FPGAs), for computing purposes [1]. Processing power is an important requirement not only for scientific computing applications but for interactive applications that include image processing, voice recognition, video streaming etc. Hence there is always demand on the computation units to implement these applications. One of the ways to support such huge processing requirement is to use reconfigurable computing systems. A Reconfigurable computing system usually consists of one or more processors and a reconfigurable fabric [2]. The processor executes the sequential or noncritical code, while code that can be executed by processing units is mapped to the reconfigurable fabric. Furthermore, the parts of the computation that contain high degree of parallelism can be mapped to the reconfigurable unit. AES-GCM is one of the algorithms that support green computing. GHASH and AES operations can be run in parallel and thus, it has been the choice for implementation. Authenticated Encryption is a common requirement in many security applications as it guarantees both privacy and integrity. Several dedicated modes for authenticated encryption have been devised so far. The most popular such mode is the AES in Galois Counter Mode which is optimized for high performance[6]. Galois Counter Mode provides confidentiality, integrity and authenticity assurances on the data. Two functions in GCM are authenticated encryption (AE), authenticated decryption (AD) which includes GHASH. It can be efficiently pipelined and parallelized in hardware for implementation of these functions. Fig. 1. AES-GCM Authenticated Encryption operation [7] II. STEPS IN AES-GCM ALGORITHM 1. Let H=E(K ,0 128 ). 2. Define a block J 0 , as follows: if len(IV) = 96, then let J 0 = IV || 0 31 ||1. if len(IV) ≠ 96, then let s = 128 ڿlen(IV)/128ۀ−len(IV), and let J 0 =GHASH H (IV||0s+64|| ڿlen(IV)ۀ64 ). 3. C=GCTR k (inc32(J 0 ), P) 4. Let u=128 ڿlen(C)/128 ۀ−len(C) and let v=128 ڿlen(A)/128 ۀ−len(A). 5. Define a block, S, as follows: S=GHASH H (A || 0 u || C || 0 v || [len(A)] 64 || [len(C)] 64 ). 6. Let T = MSB t (GCTR k (J0, S)). Vol. 14 ICETCSE 2016 Special Issue International Journal of Computer Science and Information Security (IJCSIS) ISSN 1947-5500 [https://sites.google.com/site/ijcsis/] 34 Proceedings of 3rd International Conference on Emerging Technologies in Computer Science & Engineering (ICETCSE 2016) V. R. Siddhartha Engineering College, Vijayawada, India, October 17-18, 2016