In recent past years, Deep Learning presented an excellent performance in different areas like image recognition, pattern matching, and even in cybersecurity. The Deep Learning has numerous advantages including fast solving complex problems, huge automation, maximum application of unstructured data, ability to give high quality of results, reduction of high costs, no need for data labeling, and identification of complex interactions, but it also has limitations like opaqueness, computationally intensive, need for abundant data, and more complex algorithms. In our daily life, we used many applications that use Deep Learning models to make decisions based on predictions, and if Deep Learning models became the cause of misprediction due to internal/external malicious effects, it may create difficulties in our real life. Furthermore, the Deep Learning training models often have sensitive information of the users and those models should not be vulnerable and expose security and privacy. The algorithms of Deep Learning and machine learning are still vulnerable to different types of security threats and risks. Therefore, it is necessary to call the attention of the industry in respect of security threats and related countermeasures techniques for Deep Learning, which motivated the authors to perform a comprehensive survey of Deep Learning security and privacy security challenges and countermeasures in this paper. We also discussed the open challenges and current issues.

No abstract

During the last two decades, free-space optical links got considerable importance due to their benefits of higher data rates, license free-spectrum, easy and rapid deployment and mobility. Free-space optical links use carrier frequency in the range of 20 THz to 375 THz (in near infrared (IR) region and visible band in wavelengths) to establish a communication link for terrestrial communication, inter-satellite links, deep space links, ground-tosatellite and satellite-to-ground links. Free-space optical links are also useful for different military applications, disaster recovery and last mile access. However, despite of having all these advantages the performance of freespace optical links depends upon the atmospheric conditions and parameters of system design. Geometrical losses of free-space optical links are directly related to parameters of system design or internal parameters. In this paper we analyzed different parameters of system design to minimize the geometrical losses. We presented the analysis of internal design parameters like divergence angle, diameter of receiver aperture, diameter of transmitter aperture, link distance and suggested the suitable parameters of system design.

Coronavirus disease , also known as Severe acute respiratory syndrome (SARS-COV2) and it has imposed deep concern on public health globally. Based on its fast-spreading breakout among the people exposed to the wet animal market in Wuhan city of China, the city was indicated as its origin. The symptoms, reactions, and the rate of recovery shown in the coronavirus cases worldwide have been varied . The number of patients is still rising exponentially, and some countries are now battling the third wave. Since the most effective treatment of this disease has not been discovered so far, early detection of potential COVID-19 patients can help isolate them socially to decrease the spread and flatten the curve. In this study, we explore state-of-the-art research on coronavirus disease to determine the impact of this illness among various age groups. Moreover, we analyze the performance of the Decision tree (DT), K-nearest neighbors (KNN), Naïve bayes (NB), Support vector machine (SVM), and Logistic regression (LR) to determine COVID-19 in the patients based on their symptoms. A dataset obtained from a public repository was collected and pre-processed, before applying the selected Machine learning (ML) algorithms on them. The results demonstrate that all the ML algorithms incorporated perform well in determining COVID-19 in potential patients. NB and DT classifiers show the best performance with an accuracy of 93.70%, whereas other algorithms, such as SVM, KNN, and LR, demonstrate an accuracy of 93.60%, 93.50%, and 92.80% respectively. Hence, we determine that ML models have a significant role in detecting COVID-19 in patients based on their symptoms.

Free-space optical links use modulated beam of light to transmit high amount of data from transmitter to receiver to get line-of-sight communication link. Free-space optical is cost effective solution to provide higher data rate to end-users. Free-space optical links are also considered as a better alternative to RF links due to their advantages of low power consumption and higher data rate of the range Gbps, and are highly secure to electromagnetic interference. Despite of having all these advantages free-space optical links are highly affected by the severe weather conditions like fog, rain, snow, smoke and dust or aerosol particles suspended in air. Fog is one of a major challenge for free-space optical to achieve carrier class availability and causes high amount of attenuation. In this paper we presented the performance analysis of free-space optical links by estimating signal-to-noise ratio, bit error rate, under fog conditions during winter season in Lahore, Pakistan.