Dr Razi Arshad

Research Fellow in Security and Privacy



Research interests


Razi Arshad, Nassar Ikram (2013) Elliptic curve cryptography based mutual authentication scheme for session initiation protocol

The Session Initiation Protocol (SIP) is the most widely used signaling protocol for controlling communication on the internet, establishing, maintaining, and terminating the sessions. The services that are enabled by SIP are equally applicable in the world of multimedia communication. Recently, Tsai proposed an efficient nonce-based authentication scheme for SIP. In this paper, we do a cryptanalysis of Tsai’s scheme and show that Tsai’s scheme is vulnerable to the password guessing attack and stolen-verifier attack. Furthermore, Tsai’s scheme does not provide known-key secrecy and perfect forward secrecy. We also propose a novel and secure mutual authentication scheme based on elliptic curve discrete logarithm problem for SIP which is immune to the presented attacks.

Razi Arshad, Mudassar Jalil (2023) Comment on Nizam Chew, L.C.; Ismail, E.S. S-box Construction Based on Linear Fractional Transformation and Permutation Function. Symmetry 2020, 12, 826

The aim of this comment paper is to identify a technical error in the published article of Nizam Chew et al.

Hamza Ali Imran, Qaiser Riaz, Muhammad Zeeshan, Mehdi Hussain, Razi Arshad (2023) Machines Perceive Emotions: Identifying Affective States from Human Gait Using On-Body Smart Devices

Emotions are a crucial part of our daily lives, and they are defined as an organism’s complex reaction to significant objects or events, which include subjective and physiological components. Human emotion recognition has a variety of commercial applications, including intelligent automobile systems, affect-sensitive systems for customer service and contact centres, and the entertainment sector. In this work, we present a novel deep neural network of the Convolutional Neural Network - Bidirectional Gated Recurrent Unit (CNN-RNN) that can classify six basic emotions with an accuracy of above 95%. The deep model was trained on human gait data captured with body-mounted inertial sensors. We also proposed a reduction in the input space by utilizing 1D magnitudes of 3D accelerations and 3D angular velocities, which not only minimizes the computational complexity but also yields better classification accuracies. We compared the performance of the proposed model with existing methodologies and observed that the model outperforms the state-of-the-art.