Zafar Mehmood

Mr. Zafar Mehmood


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
11 January 2023 in Sustainability
Verified
Journal Article
Open access
PDF file
ZMZafar Mehmood
Zafar Mehmood
BCBenmao Cheng
Benmao Cheng
NBNaveed Anwer Butt
Naveed Anwer Butt
GRGhani rehman
Ghani rehman
MZMuhammad Zubair
Muhammad Zubair
MAMUAHMMAD ASLAM
MUAHMMAD ASLAM
The main problem for both the utility companies and the end-used is to efficiently schedule the home appliances using energy management to optimize energy consumption. The microgrid, macro grid, and Smart Grid (SG) are state-of-the-art technology that is user and environment-friendly, reliable, flexible, and controllable. Both utility companies and end-users are interested in effectively utilizing different heuristic optimization techniques to address demand-supply management efficiently based on consumption patterns. Similarly, the end-user has a greater concern with the electricity bills, how to minimize electricity bills, and how to reduce the Peak to Average Ratio (PAR). The Home Energy Management Controller (HEMC) is integrated into the smart grid, by providing many benefits to the end-user as well to the utility. In this research paper, we design an efficient HEMC system by using different heuristic optimization techniques such as Genetic Algorithm (GA), Binary Particle Swarm Optimization (BPSO), and Wind Driven Optimization (WDO), to address the problem stated above. We consider a typical home, to have a large number of appliances and an on-site renewable energy generation and storage system. As a key contribution, here we focus on incentive-based programs such as Demand Response (DR) and Time of Use (ToU) pricing schemes which restrict the end-user energy consumption during peak demands. From the results figures, it is clear that our HEMC not only schedules all the appliances but also generates optimal patterns for energy consumption based on the ToU pricing scheme. As a secondary contribution, deploying an efficient ToU scheme benefits the end-user by paying minimum electricity bills, while considering user comfort, at the same time benefiting utilities by reducing the peak demand. From the graphs, it is clear that HEMC using GA shows better results than WDO and BPSO, in energy consumption and electricity cost, while BPSO is more prominent than WDO and GA by calculating PAR.

22 December 2022 in Sensors
Verified
Journal Article
Open access
PDF file
GRGhani rehman
Ghani rehman
MZMuhammad Zubair
Muhammad Zubair
IQIqbal Qasim
Iqbal Qasim
ABAfzal Badshah
Afzal Badshah
ZMZafar Mehmood
Zafar Mehmood
Syeda Fizzah JilaniSyeda Fizzah Jilani
Syeda Fizzah Jilani
Since several Internet of Things (IoT) applications have been widely deployed on unstable wireless networks, such as the Delay Tolerant Network (DTN), data communication efficiency in DTN remains a challenge for IoT applications. Vehicular Delay Tolerant Network (VDTN) has become one of DTN’s potential applications, in which the network experiences connectivity interruption due to the lack of an end-to-end relay route. VDTNs focus on node cooperation to achieve this goal. As a result, it is essential to ensure that almost all network nodes adopt the protocol to preserve network performance. This is a challenging task since nodes may diverge from the basic protocol to optimize their effectiveness. This article presents an Efficient Monitoring System (EMS) to detect and respond to just selfish nodes to minimize their entire network and data communication efficacy. The scheme is based on a network-wide cooperative sharing of node reputation. It is also necessary to increase overall network efficiency by tracking selfish nodes. The NS-2 simulator is used to run this experimental setup. Simulation results indicate that the proposed scheme performs better in terms of probability of package delivery, package delivery delay, energy consumption, and amount of packet drops. For 80% selfish nodes in the network, the packet delivery of EMS is 37% and 31% better than SOS and IPS. Similarly, the average delivery delay of EMS is 22% and 18% lower than SOS and IPS when 80% selfish nodes are incorporated in the network.

24 August 2022 in Multimedia Tools and Applications
Verified
Journal Article
Closed access
PDF file
GRGhani rehman
Ghani rehman
MHMuhammad INAM UL HAQ
Muhammad INAM UL HAQ
MZMuhammad Zubair
Muhammad Zubair
ZMZafar Mehmood
Zafar Mehmood
Madhusudan SinghMadhusudan Singh
Madhusudan Singh
Dhananjay SinghDhananjay Singh
Dhananjay Singh

25 August 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
ZMZafar Mehmood
Zafar Mehmood
LJLeila Jamel
Leila Jamel
DSDina Ahmed Salem
Dina Ahmed Salem
Imran AshrafImran Ashraf
Imran Ashraf
Several issues are there to prevent the traditional classifiers from getting an acceptable performance level while learning from multi-class problems. One of the main problems is the unequal distribution of samples, which significantly reduces the efficiency of the underlying classifier when combined with incompatible optimization benchmarks and data overlapping phenomena. The classifier performance is compromised beyond the expected level by the combined effects of imbalanced distribution and sample overlapping around the class boundaries. This problem worsens with the increase in the number of classes in the multi-class scenario. Despite having a more significant combined effect on classifier performance, the combined effects of imbalanced data and overlapping questions have been given the least attention in the research. To improve models’ learning from imbalanced multi-class and overlapping of shared attributes issues, this work introduces SVM++, a modified version of support vector machines (SVM). Comprising of three steps, Algorithm-1 finds and splits the training set into overlapping and non-overlapping samples. Algorithm-2 then separates the overlapped data into the Critical-1 and Critical-2 regions. The Critical-1 region consists of overlapped samples, sharing similar characteristics, which is the main cause of degraded classification performance. In the third step, an algorithm based on the mean of the maximum and minimum distance of the Critical-1 region samples is proposed by improving the traditional SVM kernel mapping function to a higher dimension. Thirty real datasets with various imbalances and degrees of overlap are utilized to compare our suggested algorithms’ supremacy with the state-of-the-art classifiers.

22 August 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
ZMZafar Mehmood
Zafar Mehmood
NBNaveed Anwer Butt
Naveed Anwer Butt
GRGhani rehman
Ghani rehman
MZMuhammad Zubair
Muhammad Zubair
MAMUAHMMAD ASLAM
MUAHMMAD ASLAM
Syeda Fizzah JilaniSyeda Fizzah Jilani
Syeda Fizzah Jilani
The classification of imbalanced and overlapping data has provided customary insight over the last decade, as most real-world applications comprise multiple classes with an imbalanced distribution of samples. Samples from different classes overlap near class boundaries, creating a complex structure for the underlying classifier. Due to the imbalanced distribution of samples, the underlying classifier favors samples from the majority class and ignores samples representing the least minority class. The imbalanced nature of the data—resulting in overlapping regions—greatly affects the learning of various machine learning classifiers, as most machine learning classifiers are designed to handle balanced datasets and perform poorly when applied to imbalanced data. To improve learning on multi-class problems, more expertise is required in both traditional classifiers and problem domain datasets. Some experimentation and knowledge of hyper-tuning the parameters and parameters of the classifier under consideration are required. Several techniques for learning from multi-class problems have been reported in the literature, such as sampling techniques, algorithm adaptation methods, transformation methods, hybrid methods, and ensemble techniques. In the current research work, we first analyzed the learning behavior of state-of-the-art ensemble and non-ensemble classifiers on imbalanced and overlapping multi-class data. After analysis, we used grid search techniques to optimize key parameters (by hyper-tuning) of ensemble and non-ensemble classifiers to determine the optimal set of parameters to enhance the learning from a multi-class imbalanced classification problem, performed on 15 public datasets. After hyper-tuning, 20% of the dataset samples are synthetically generated to add to the majority class of each respective dataset to make it more overlapped (complex structure). After the synthetic sample’s addition, the hyper-tuned ensemble and non-ensemble classifiers are tested over that complex structure. This paper also includes a brief description of tuned parameters and their effects on imbalanced data, followed by a detailed comparison of ensemble and non-ensemble classifiers with the default and tuned parameters for both original and synthetically overlapped datasets. We believe that the underlying paper is the first kind of effort in this domain, which will furnish various research aspects to with a greater focus on the parameters of the classifier in the field of learning from imbalanced data problems using machine-learning algorithms.

20 March 2023 in Sustainability
Verified
Journal Article
Open access
PDF file
GRGhani rehman
Ghani rehman
MZMuhammad Zubair
Muhammad Zubair
Wael Hosny Fouad AlyWael Hosny Fouad Aly
Wael Hosny Fouad Aly
HFHaleem Farman
Haleem Farman
ZMZafar Mehmood
Zafar Mehmood
NBNaveed Anwer Butt
Naveed Anwer Butt
The term “Internet of Things” (IoT) refers to an architecture in which digital objects have identification, sensing, connectivity, and processing capabilities that allow them to connect with other devices as well as perform tasks on the internet. There are many applications of IoT, among which Vehicle Delay-Tolerant Networks (VDTNs) are one of the best known. This new generation of vehicular networks can be applied in a variety of circumstances. For example, it can be employed to make data connections possible in densely crowded cities and as well as in remote and sparsely populated places with weak connectivity. These environments are characterized by frequent network partitioning, inconsistent connectivity, considerable propagation delays, high error rates, and short contact duration. Most of these behaviours are due to node selfishness. This task is crucial because selfish behaviour by nodes may make other nodes hesitant to cooperate. Selfish nodes have significant negative impacts on the effectiveness and efficiency of the network as a whole. To solve these issues, cooperative strategies that motivate nodes to share their resources must be considered. Important contributions to cooperation for vehicular networks are presented in this article, which investigates the effects of six different cooperative techniques on network performance and makes corresponding suggestions for their use in IoT-based VDTNs. Across all simulations, our results show that the studied strategies are all able to increase overall network performance by improving throughput and packet delivery probability, which in turn reduces average packet delivery time, energy consumption, overhead ratio, and the number of packets dropped.