Muhammad Nasir Khan

Bio: Muhammad Nasir Khan is an academic researcher from University of Lahore. The author has contributed to research in topics: Communications system & Bandwidth (signal processing). The author has an hindex of 10, co-authored 50 publications receiving 326 citations. Previous affiliations of Muhammad Nasir Khan include National University of Science and Technology & University of South Australia.

Skin Lesion Segmentation from Dermoscopic Images Using Convolutional Neural Network.

Abstract: Clinical treatment of skin lesion is primarily dependent on timely detection and delimitation of lesion boundaries for accurate cancerous region localization. Prevalence of skin cancer is on the higher side, especially that of melanoma, which is aggressive in nature due to its high metastasis rate. Therefore, timely diagnosis is critical for its treatment before the onset of malignancy. To address this problem, medical imaging is used for the analysis and segmentation of lesion boundaries from dermoscopic images. Various methods have been used, ranging from visual inspection to the textural analysis of the images. However, accuracy of these methods is low for proper clinical treatment because of the sensitivity involved in surgical procedures or drug application. This presents an opportunity to develop an automated model with good accuracy so that it may be used in a clinical setting. This paper proposes an automated method for segmenting lesion boundaries that combines two architectures, the U-Net and the ResNet, collectively called Res-Unet. Moreover, we also used image inpainting for hair removal, which improved the segmentation results significantly. We trained our model on the ISIC 2017 dataset and validated it on the ISIC 2017 test set as well as the PH2 dataset. Our proposed model attained a Jaccard Index of 0.772 on the ISIC 2017 test set and 0.854 on the PH2 dataset, which are comparable results to the current available state-of-the-art techniques.

A Protection Scheme for Multi-Terminal VSC-HVDC Transmission Systems

Abstract: High-voltage direct-current (HVDC) power transmission is becoming increasingly important due to steadily rising need for bulk power delivery and interconnected power transmission and distribution systems. DC grids are vulnerable to dc faults, which lead to a rapid rise in dc fault currents. The dc faults must be cleared within the timeframe of milliseconds to avoid the collapse of the HVDC system. In the event of primary protection (PP) failure, back-up protection (BP) must be applied to clear the fault. In this paper, a novel algorithm based on a Naive Bayes classifier is proposed to determine threshold levels and operational time frames for primary and back-up protection in multi-terminal voltage source converter-based HVDC. Local voltage and currents are measured to detect and identify the kind of fault. A four-terminal HVDC transmission system is developed in PSCAD/EMTDC and is subjected to line–line faults at different locations and time, to assess the designed protection schemes. Results show that a relaying algorithm effectively detects the fault and expedite the primary protection operation. On malfunctioning of PP, BP is accelerated in a short delay of 0.2 ms. Furthermore, the relaying algorithm provides faster protection compared with techniques available in the literature. The resulting reduced fault clearance time truncates the maximum fault current and, inevitably, leads to reduced power ratings required for dc grid equipment.

Adaptive hybrid free space optical/radio frequency communication system

Abstract: Wireless communication has achieved lot of attention and the demand is continually increasing day by day. Radio frequency (RF) is highly attracted by various wireless communication applications. The RF spectrum is already very crowded and the rapid increase in the use of wireless services has led the problems of RF spectrum exhaustion and eventually RF spectrum deficit. Free space optical (FSO) communication is a viable technology with a plenty of bandwidth, license-free spectrum and interference free link. On the other hand, FSO channel is severely corrupted by atmospheric turbulence and non-predictive weather scenarios. We suggest a hybrid FSO/RF communication system in our previous research, which can mitigate the issues of the individual links. In this research, we investigate the performance of the proposed adaptive system for reliable data transmission. We develop modulation and power adaptive schemes for maximizing the mutual information. The proposed adaptive system is compared with non-adaptive system, which gives 2.75 dB gain for the joint power and 0.75 dB gain for the separate power constraint.

Maximizing Throughput of Hybrid FSO-RF Communication System: An Algorithm

Abstract: Electromagnetic spectrum is too cluttered to add additional broadband channels of high data rate and bandwidth. Free space optical (FSO) communication is one of the most dominant optical wireless communication systems, which provides huge and licensed free spectrum, non-interfering link, and high data rate. Despite of having above-mentioned advantages, the FSO links are very sensitive to bad weather conditions (i.e., fog, snow, dust, and their combination). Therefore, there is a dire requirement to develop a system that can overcome issues of individual communication system and adjust the current demand of new broadband channels. Adaptive FSO-radio frequency (RF) communication system is proposed to tackle the individual channel issues. A novel throughput maximization algorithm (TMA) is developed for the adaptive hybrid FSO-RF channel. Performance of the proposed algorithm for adaptive hybrid communication system is analyzed considering the regular and right-regular low density parity check (LDPC) code under various weather conditions. Simulation results show that the TMA performs well under all weather conditions and gives a performance gain of up to 2.25 dB considering the right-regular LDPC code. From the presented results, it is also seen that the suggested communication system optimizes mapping schemes, bit proportions for each channel, and puncturing ratios adaptively for different degree variable nodes in worst weather conditions.

Signal dependent Gaussian noise model for FSO communications

Abstract: Free space optical (FSO) communication is an emerging technology which can fulfil the current demands of additional broadband channels. It offers enormous bandwidth, less expensive setup and highly secure links. But it has some potential drawbacks as well i.e. its links deteriorate significantly due to the atmospheric turbulence and weather conditions (fog or cloud). The detection of the incoming light signals in FSO communications is done by photodiodes. The most preferable photodiodes are the photodetectors (avalanche photodiodes (APD)/positive-intrinsic-negative (PIN)). APDs are normally used in high performance FSO links where the noise shows different distribution for bits Os and Is rather than the Gaussian distribution. A signal dependent Gaussian noise (SDGN) model is proposed that incorporate the implementation of APDs for high performance communication links. We investigate the use of ON-OFF keying (OOK) which shows a good performance in terms of achieving optical power gains. Expressions are derived for the log-likelihood ratio (LLR) mapping of a received bit, the uncoded bit error rate (BER) and the ergodic capacity. We also provide the analysis of the FSO optimum/sub-optimum detectors.

Machine learning

Abstract: Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Dynamics and Control

Abstract: 1. Control Methodology 2. Dynamical Systems 3. Applications to Social and Environmental Problems

HVDC Transmission: Technology Review, Market Trends and Future Outlook

Abstract: HVDC systems are playing an increasingly significant role in energy transmission due to their technical and economic superiority over HVAC systems for long distance transmission. HVDC is preferable beyond 300–800 km for overhead point-to-point transmission projects and for the cable based interconnection or the grid integration of remote offshore wind farms beyond 50–100 km. Several HVDC review papers exist in literature but often focus on specific geographic locations or system components. In contrast, this paper presents a detailed, up-to-date, analysis and assessment of HVDC transmission systems on a global scale, targeting expert and general audience alike. The paper covers the following aspects: technical and economic comparison of HVAC and HVDC systems; investigation of international HVDC market size, conditions, geographic sparsity of the technology adoption, as well as the main suppliers landscape; and high-level comparisons and analysis of HVDC system components such as Voltage Source Converters (VSCs) and Line Commutated Converters (LCCs), etc. The presented analysis are supported by practical case studies from existing projects in an effort to reveal the complex technical and economic considerations, factors and rationale involved in the evaluation and selection of transmission system technology for a given project. The contemporary operational challenges such as the ownership of Multi-Terminal DC (MTDC) networks are also discussed. Subsequently, the required development factors, both technically and regulatory, for proper MTDC networks operation are highlighted, including a future outlook of different HVDC system components. Collectively, the role of HVDC transmission in achieving national renewable energy targets in light of the Paris agreement commitments is highlighted with relevant examples of potential HVDC corridors.