ABSTRACT. Breast cancer is the most diagnosed type of cancer as per the data that is collected by World Health Organization (WHO) within the past few years. Over 600.000 deaths were recorded in 2021 due to breast cancer. Breast cancer screening is done using 2D and 3D mammography, but MRIs and Ultrasounds are also used in certain conditions. The diagnosis from the screenings are not always accurate as a practitioner has to physically look at the digital images to find any signs of cancer. Essentially, each diagnosis has a variable chance of a false-positive or a false-negative. Many CAD (computer aided detection) systems have been developed for the assistance of a practitioner with the diagnosis. However, in the past years, Deep Neural Networks (DNN) have seen a spike and the models are being used to aid the breast cancer screening. Data shows a possibility of reaching AUC values as high as 0.99 under ideal conditions when the training dataset is cleaned of noise and properly pre-processed and in some studies, the accuracy and sensitivity is even compared to that of a practitioner’s, with the DNN model outperforming in numbers across the board. After performing a literature review on similar work, we have trained a model of our own on a publicly available dataset (MIAS) reaching promising results of an AUC 0.87 and Accuracy of 0.88 with the initial model built on a DenseNet121 architecture.

ABSTRACT. For overhead power distribution, Aerial Bundle Cables often referred to as Aerial Bundled Conductors or simply ABCs are used. With the insulation of the cables, the ABCs offer better protection from electrical pilferage; subsequently, these cables are of interest to metropolitan areas with electrical pilferage problems. ABCs utilized in Karachi metropolitan by M/s K-Electric, instead of the standard copper cables to prevent electricity from theft. Nonetheless, because of the relevantly new and infrequent use of the cables in Karachi, the life and performance of these Aerial Bundle Cables aren’t known for Karachi; which has a harsh climate due to being on the coastal line. The identification of the degradation rate of the ABCs is of great interest to various factors, such as the state of the environment, thermal, moisture, mechanical loading conditions, etc. The implementation of suitable non-destructive testing (NDT) techniques is beneficial for the RUL prediction to assess the cable condition and in turn degradation rates of the cables. The NDT and environmental data will enable degradation trend prediction by maintaining a historical database. The degradation trend information can lead to predicting the remaining useful life (RUL) of the ABCs. This research work focuses on the degradation trend prediction of ABCs using sophisticated Artificial Intelligence (AI) model.

ABSTRACT. Power-limited devices (or sensors) constrain the deployment of modern IoT networks, such as Next-Generation Industrial IoT (NG-IIoT). These networks are envisioned as the key enablers to facilitate connectivity to billions of devices for applications like smart industries, smart healthcare, etc. This paper presents an optimal uplink communication protocol for a power-limited unlicensed sensor operating among numerous licensed sensors communicating in a time division multiple access (TDMA)-based scheme. The transmission of the power-limited sensor is ensured by employing the non-orthogonal multiple access (NOMA) technique during the timeslot of a licensed sensor. To ensure energy-efficient communication, we maximize the throughput of the power-limited sensor using a deep reinforcement learning (DRL) framework recognized as a combined experience replay deep deterministic policy gradient (CER-DDPG) algorithm. Our simulation results demonstrated that the CER-DDPG-based communication protocol outperforms the benchmark schemes, such as DDPG and stochastic algorithms, in terms of throughput.

ABSTRACT. As Internet of Things (IoT) is evolving very rapidly which is producing massive amount of data and normally all this data is shifted towards the cloud for further execution which puts burden on the cloud and doesn’t fulfill QoS requirements. In order to overcome this problem computing facility were extended towards the edge of the network and the concept is known as Fog computing. Due to heterogeneity of fog nodes and dynamic user requirements, task scheduling in fog computing has become the big challenge. This paper covers the literature review which classify the task scheduling algorithm in terms of approaches, it highlights the parameters mostly used, it also covers the analysis among the available simulation tools and highlight the open issues and challenges which will help the researchers for their problem formulation.

ABSTRACT. Networking industry has changed a lot in the past few years and as SDN being the new emerging technology of the future, security concerns have risen enormously. Control plane is separated from data plane in SDN providing the network administrators with more control by pushing application codes into the network also making it more vulnerable to attacks. DoS/DDoS are the most dominant attacks that a network of an organization faces these days, because of the unsupervised network subscriptions and easily available attack tools/applications. This research aims to integrate the traditional open source IDS with Software Defined Networks and test their capabilities against malicious traffic injected by an attacker. In this paper, we compared two open source IDS Snort and Suricata with SDN and then deliberately injected malicious traffic. We time stamped the DoS/DDoS attack in both scenarios to compare Snort and Suricata true positive and false negative rates. From the results obtained we concluded that used IDS were successfully integrated with a prototype emulated network on Mininet and OpenDaylight being the remote controller and after injecting attacks into the network.

ABSTRACT. A way behind the past where teaching and learning were only by video, images, and 2D animations. Now we have live sessions where a classroom is bounded on just a laptop screen. What happened if that live session and 3D imaging are presented in the form of real sculpture by using Augmented Reality? In this paper, we are introducing a low-cost 3D display that presents both recorded and live sessions. Also, this display will be controlled with our fingertips. This application covers sectors of education, medical imaging, and advertisements with good effects and cheap cost.

ABSTRACT. Water shortage is the major issue in the modern world. Conservation is urgently needed as a result of the population's growing need for water on a minute-by-minute basis. Every aspect of life has been automated thanks to information and technology. Automation is now simpler because to the expansion of the Internet of Things (IoT). Through IoT, all the parts may now be connected. It has applications in the monitoring and management of equipment. IoT has given humans the ability to remotely operate a variety of gadgets. We provide a model to manage and monitor the water flow in order to conserve water using the wonders of IoT. The proposed methodology uses water tanks with ultrasonic sensors to measure the liquid level and contrast it with the depth of the container. The storage system's water level is detected by the system. In this case, sensors are installed in the tank at various levels. A microcontroller is attached to the sensors (ESP32). The sensors gauge the water level and transmit that information to the microprocessor, which updates the mobile application and LCD with the tank condition. Using an ultrasonic sensor, the user can manually override the mobile application's operation of the pump. This technology will give users a mobile application and buzzer to let them know how their tanks are doing, making it simple for them to operate. The system can be enhanced to ensure that dams operate effectively.

ABSTRACT. Due to increase in the data breaches over the last few years and even the initial layers of authentication failed to stop the data breaches its been crucial to take a step forward and bust the cyber frauds and maintain the confidentiality and integrity of the user.To Achieve the cyber goals in this paper we presented an idea of eliminating a central authority by keeping a decentralized system so that the attacks on server can be totally eliminated and therefore To tackle the data breaches by implementing the blockchain architecture ,each node will be user and will be storing the information of the previous node in the network.to add one more layer of authentication that is Biometric .Data of the user will remain secured and if the biometric identity is compromised losses will be very less compared to centralized system also the information of biometric data is way more secured.

ABSTRACT. Disasters are unpredictable phenomena that occur in different periods, cause damage and sometimes loss of life. There are various types of disasters that include flooding, drought, wildfires, earthquakes, and building collapse. Some of them happen naturally while others are caused due to human ignorance and affect the infrastructure. This research focuses on identification of human beings trapped in rubble of a collapsed building to increase the chance of survival using the microwave technique. The proposed methodology is used to estimate the digging in the correct direction that is difficult to discover human lives. Radio frequency waves are transmitted that pass through the rubble and detect humans by the reflected signal. The change due to transmitting and receiving signals is observed via an antenna that triggers an alarm to indicate the human presence. This technique is implemented using a robot that can travel difficult to access places. The system has been tested at small scale with satisfactory results.

ABSTRACT. Congestion is an unavoidable issue in modern, scientific, and commercial Wireless Sensor Network (WSN) applications, where irregular and bursty traffic increased sending rate. The congestion is aggravated due to the coherent nature of traffic, which affects the continuous flow of data. The underlying node becomes a bottleneck, which results in degradation of network performance (increased packet loss, decreased packet delivery ratio (PDR), and increased delay). This paper presents a novel N3PID (Non-dominated Sorting Genetic Algorithm III (NSGA-III) Proportional-Integral-Derivative (PID)) controller to compute the optimal sending rate for a source node that can assist to control congestion. PID is employed to attain an optimal sending rate for the source node by taking the error (difference of instantaneous and expected queue) as input. Moreover, a reference-based NSGA-III is exercised to provide a more accurate response by efficiently tuning the controller parameters of traditional PID. The proposed controller and a few counterparts are implemented using computer simulation in NS3 (Network Simulator version 3) software. The network performance is analyzed to prove the enhancement of the N3PID controller through a simulation study, which shows increased PDR and substantially decreased delay and packet loss as compared with N3PID.

ABSTRACT. In this project, voice recognition controls the wheelchair. This invention helps disabled, ill, or elderly wheelchair users since the design allows sick or disabled people to navigate by voice. Disabled people need wheelchairs because of weak upper limbs. Since a 24-hour caretaker is unrealistic, a voice-controlled motorized wheelchair is beneficial. As many sick or disabled persons as possible must have inexpensive access. Simple voice-activated wheelchair prototype can be designed using a voice-controlled home automation system so elderly and disabled persons can manage AC power and appliances.

ABSTRACT. Some of the most fundamental aspects of cyber security include authentication, confidentiality, integrity, and security. In this survey paper, we provide a survey on different existing Internet of things (IoT) devices like smart home environments, smart buildings, smart factories, etc. based on three factors of authentication. The solution of IoT applications anticipated by wireless networks to legitimate clients and users. For making secure IoT solutions for users, numerous mechanisms and protocols and cryptographic techniques have been studied which are proposed by some researchers. Things need a proper authentication system. We have studied different author’s research that is based on Physically Secure lightweight protocol, multi-Gateway secure IoT Environments protocol, Tele biometric objects (TAO) with biometric sensors and telecommunication networks, and Multi-Factor Authentication that prevent for the users to avoid threats and man-in-the-middle attacks, etc.