Work place: Imam Hossein University/ Department of Computer Engineering,Tehran, Iran



Research Interests: Computational Science and Engineering, Computer Architecture and Organization, Network Architecture, Network Security, Distributed Computing, Data Structures and Algorithms


Arash Ghafouri is a researcher of computer engineering in the school of computer engineering of Imam Hossein University. He was born in Tehran, Iran in 1990. He received his Master of Science in computer engineering with area of study of distributed systems from Iran University of Science and Technology in 2014. His current research interests include distributed and high performance computing, virtualization technology, network security, and data science.

Author Articles
IAMIN: An Infrastructure-aware Middleware for the Reconfiguration of Unstructured Peer-to-Peer Overlay Networks

By Morteza Alizadeh Arash.Ghafouri Mohsen Sharifi

DOI:, Pub. Date: 8 May 2019

Peer to peer networks have become one of the most popular networking methods because of their flexibility and many use cases such as file sharing and distributed computations. Unstructured overlay peer to peer networks are one of key components of peer to peer systems that are considerable because of their low cost in network construction and maintenance. One of the main challenges in unstructured peer to peer overlay networks is the topology mismatch between overlay network and the underlying physical infrastructure. The root of this challenge is lack of awareness about peers in the network infrastructure during connection to and disconnection from overlay network, in addition to the neighbor selection mechanism in the overlay network. Different types of awareness of network infrastructure includes awareness of the location of internet service providers. Also awareness of proximity, geographical location and resources of peers. In this article we present a middleware which configures overlay network by using public measurements and the estimated delay among peers in order to have the most conformity with the topology of physical infrastructure. To evaluate the performance, our middleware is implemented on the top of Gnutella which is an unstructured overlay peer-to-peer network. Our simulations show that our middleware enhances the conformity of overlay network to the topology of physical network infrastructure. In addition, it improved the average throughput and the average delay.

[...] Read more.
A Power-based Method for Improving the ODMRP Protocol Performance in Mobile Ad-hoc Networks

By Arash.Ghafouri Ahmad Ghasemi Mohammad Reza. Hasani Ahangar

DOI:, Pub. Date: 8 Mar. 2018

Mobile AD hoc Networks as a special type of wireless networks have received special attention due to having special features such as no need for central management, no need for infrastructure and high mobility capability and can be used in cases where creating an effective communication infrastructure is not cost-effective or is practically impossible, such as conferences, such as battles and communications after natural disasters. Several routing protocols are proposed for these networks. ODMRP protocol is one of the most famous and used protocols in Mobile AD hoc networks. This study was carried out aimed to discuss this routing protocol and then provide a new routing method for this protocol for increasing its efficiency. In the ODMRP protocol, the optimal route is selected based on the shortest route. In wireless communications and getting the nodes away from each other, the received signal levels are weakened and may result in loss of data, and in practice, the shortest path that works based on the number of hops loses its effectiveness. In the proposed protocol, the route is selected based on the received signal strength level. According to the simulation results, the performance of the proposed protocol increases by decreasing control overhead and increasing the packet delivery rate compared to the original protocol.

[...] Read more.
On a Novel Grid Computing-Based Distributed Brute-force Attack Scheme (GCDBF) By Exploiting Botnets

By Mohammad Reza. Hasani Ahangar Mohammad Reza. Esmaeili Taba Arash.Ghafouri

DOI:, Pub. Date: 8 Jun. 2017

Brute-force attacks are known to be the promising way to break into even most complicated systems by trying every possible permutation of the keys. But since cryptosystems began to use longer and more complex keys, brute-force attacks has lost their usability, because of relatively high complexity of trying every possible permutation with respect to computational power and computation time that was available to crypto breakers. Although computational power is increasing continuously, its increasing rate is less than that of key length and complexity. Having these assumptions in mind, it is infeasible for centralized traditional computing architectures with limited computation power to break into modern cryptosystem by compromising the key with implementing schemes like conventional brute-force. In this paper authors aim for devising a novel brute-force scheme which integrates a modern computing architecture (grid computing) with botnets in order to perform brute-force attacks with lower computation time and lower equipment cost for individual cryptobreakers who have no access to supercomputers. In summary, GCDBF uses a portion of computation power of each of the infected nodes belonging to a botnet in a grid-based environment in order to process a portion of total workload of a brute-force attack which is needed for breaking a specific key. This approach neutralizes the need of acquiring supercomputers for individual hackers while reducing the required time for breaking the key because of using grid computing architecture. For the purpose of evaluation, GCDBF is implemented in different scenarios to prove its performance in comparison to centralized brute-force scheme.

[...] Read more.
Other Articles