Syed Muhammad Atif

Syed Muhammad Atif

Auto-generated profile, based on publicly available information

Syed Muhammad Atif

Graduate School of Science and Engineering, Karachi Institute of Economics and Technology, Korangi Creek, Pakistan

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 May 2019 in Pattern Recognition Letters
Verified
Journal Article
Closed access
SASyed Muhammad Atif
Syed Muhammad Atif
SQSameer Qazi
Sameer Qazi
NGNicolas Gillis
Nicolas Gillis

24 June 2022 in Neural Processing Letters
Verified
Journal Article
Open access
PDF file
SASyed Muhammad Atif
Syed Muhammad Atif
Shujaat KhanShujaat Khan
Shujaat Khan
INImran Naseem
Imran Naseem
RTRoberto Togneri
Roberto Togneri
Mohammed BennamounMohammed Bennamoun
Mohammed Bennamoun
A simple yet effective architectural design of radial basis function neural networks (RBFNN) makes them amongst the most popular conventional neural networks. The current generation of radial basis function neural network is equipped with multiple kernels which provide significant performance benefits compared to the previous generation using only a single kernel. In existing multi-kernel RBF algorithms, multi-kernel is formed by the convex combination of the base/primary kernels. In this paper, we propose a novel multi-kernel RBFNN in which every base kernel has its own (local) weight. This novel flexibility in the network provides better performance such as faster convergence rate, better local minima and resilience against stucking in poor local minima. These performance gains are achieved at a competitive computational complexity compared to the contemporary multi-kernel RBF algorithms. The proposed algorithm is thoroughly analysed for performance gain using mathematical and graphical illustrations and also evaluated on three different types of problems namely: (i) pattern classification, (ii) system identification and (iii) function approximation. Empirical results clearly show the superiority of the proposed algorithm compared to the existing state-of-the-art multi-kernel approaches.

14 July 2020 in IEEE Access
Verified
Journal Article
Open access
SQSameer Qazi
Sameer Qazi
FSFarah Sabir
Farah Sabir
BKBilal A Khawaja
Bilal A Khawaja
SASyed Muhammad Atif
Syed Muhammad Atif
MMMuhammad Mustaqim
Muhammad Mustaqim
The recent race for autonomous or ‘driverless’ vehicles, has spawned a lot of research in the area of Internet of Autonomous Vehicles (IAVs). With the advent of the latest technology fueled by Artificial Intelligence and Machine Learning, Autonomous Vehicles (AVs) can now determine the best possible route to a destination based on the current traffic situation and take dynamic driving decisions accordingly, while preventing accidents. Field trials for single autonomous vehicles have been largely successful. However, as more autonomous vehicles will be added to the intelligent transport networks, current research is now centered around their synergistic coexistence in the offering of network-centric and user-centric services. This development is governed by borrowing several concepts from the legacy Internet to address the problems of IAVs. In this paper, we present an extensive overview of the research challenges in the IAVs. Moreover, our contributions in this paper are that (i) We show how the network-oriented cooperative client-server model will give way to a more unorthodox and ‘selfish’ decentralized and peer-to-peer (P2P) model, for example in the offering of navigation services on the IAV. (ii) We discuss how centralized architecture will give way to more distributed architectures for real-time information propagation over the IAV. (iii) We discuss how network-centric policies will begin to shift to user-centric under more beneficial revenue models by offering network-assisted quality of service (QoS) provisioning. (iv) We discuss in detail how vehicle traffic grooming in the IAV would present as much of a challenge as in the legacy Internet. (v) We discuss the disruptive role of value-added services on the IAV, and (iv) Finally, we discuss the problem of cyber threats in the IAV just as in the legacy Internet.

1 December 2021 in Computer Networks
Verified
Journal Article
Closed access
SASyed Muhammad Atif
Syed Muhammad Atif
NGNicolas Gillis
Nicolas Gillis
SQSameer Qazi
Sameer Qazi
INImran Naseem
Imran Naseem

18 August 2019 in Electronics
Verified
Journal Article
Open access
PDF file
Muhammad NaveedMuhammad Naveed
Muhammad Naveed
SQSameer Qazi
Sameer Qazi
SASyed Muhammad Atif
Syed Muhammad Atif
BKBilal A Khawaja
Bilal A Khawaja
MMMuhammad Mustaqim
Muhammad Mustaqim
This research focuses on intelligent unmanned aerial vehicle (UAV)-based, real-time video surveillance to ensure better monitoring and security of remote locations over 4G-LTE cellular networks by maximizing end-user quality of experience (QoE). We propose a novel server-based crosslayer rate-adaptive scheme (SCRAS) for real-time video surveillance over 4G-LTE networks using UAVs. Our key contributions are: (1) In SCRAS, mobile UAVs having preprogrammed flight co-ordinates act as servers, streaming real-time video towards a remote client; (2) server-side video rate adaptation occurs in 4G-LTE based on the physical characteristics of the received signal conditions due to variations in the wireless channel and handovers; (3) SCRAS is fully automated and independent of client assistance for rate adaptation, as it is intended for real-time, mission-critical surveillance applications; (4) SCRAS ensures that during rate adaptation, the current video frame should not be damaged by completing the current group of packet (GoP) before adaptation. Our simulations in NS3 provide credible evidence that SCRAS outperforms recently proposed schemes in providing better QoE for real-time, rate-adaptive, video surveillance over 4G-LTE under varying channel quality and frequent handovers that occur during flight by UAVs.

1 November 2017
Verified
Conference Paper
Closed access
SASyed Muhammad Atif
Syed Muhammad Atif
SQSameer Qazi
Sameer Qazi
Today cloud computing dominates the range of services one can obtain from the Internet including Software as a service (SAAS), Platform as a service (PAAS) and Infrastructure as a Service (IAAS). These services increasingly provide ease to end user by providing a large number of applications. However, this versatility in services and applications make network planning, administration, management, security and assurances of service quality even more difficult. Hence, the significance of network traffic estimation from perspective of network planning and analysis has now become undeniable. The traffic estimation problem is often modeled as a system of linear equations Y=AX where the unobservable Origin Destination (OD) Traffic Matrix X has to be estimated from the observable Link Counts Matrix Y and Routing matrix A. However, this system of equations is under-constrained leading to difficulty in estimation of traffic matrix without knowledge of a prior estimate as a starting point. Recent research literature discusses several traffic matrix estimation techniques involving adoption of optimization approaches to satisfy hard space based constraints of the problem Y=AX which is computationally intensive, considerably expensive for real time traffic estimation. This paper proposes a novel light weight linear estimator for OD Traffic Matrix of large cloud computing platform suitable for real time prediction. Our contribution in this work is a novel algorithm which tries to incorporate these hard spaced based constraints into softer form which can be embedded into a fast, light weight linear estimator for dynamic estimation of OD Traffic Matrix.