scispace - formally typeset
Search or ask a question
Author

Mohsen Guizani

Bio: Mohsen Guizani is an academic researcher from Qatar University. The author has contributed to research in topics: Computer science & Cloud computing. The author has an hindex of 79, co-authored 1110 publications receiving 31282 citations. Previous affiliations of Mohsen Guizani include Jaypee Institute of Information Technology & University College for Women.


Papers
More filters
Proceedings ArticleDOI
01 Dec 2014
TL;DR: This paper relies on particle filtering to allocate the available bands among users in a distributed manner to propose an efficient spectrum and power allocation solution for a large scale dynamic spectrum access (DSA) systems.
Abstract: This paper proposes an efficient spectrum and power allocation solution for a large scale dynamic spectrum access (DSA) systems. Unlike conventional methods relying on optimization techniques which need huge computational capabilities and full information exchange, in this paper we rely on particle filtering to allocate the available bands among users in a distributed manner. Particle filter is based on the representation of the searched state, bands allocation per user in our case, by a set of particles. The Particle filter has the advantage, with comparison to Kalman-based filters, of its adaptivity to general scenarios (non-linear models, non-Gaussian noise, multi-modal distributions). Like Kalman-based filters, two model equations are needed for particle filter, (i) A state evolution equation to characterize the time evolution of the state. For our case, we derive a prediction equation of the channel allocation from the previous allocation from the channel fading temporal correlation, (ii) An observation equation which relates the observation, the Quality of Service in our case, to the channel allocation (state). This equation will be useful in the weighting and re-sampling phases of the filtering algorithm. The performances are analyzed in terms of the per user achieved throughput. In addition, comparison with performance when Q-learning is employed to show the efficiency of our approach.

11 citations

Journal ArticleDOI
TL;DR: A practical and privacy-preserving data aggregation scheme that can compute arbitrary aggregation functions without a TA is proposed that can ensure users’ anonymity and privacy protection, while on the other, the scheme is efficient in enabling participants to join or leave the system dynamically.

11 citations

Journal ArticleDOI
TL;DR: In this article , a framework for early diagnosis and fast classification of COVID-19 symptoms using multimodal deep learning techniques is proposed, which uses the selection algorithm of the pre-trained network to determine the best fusion model characterized by the pretrained chest X-ray and cough models.

11 citations

Journal ArticleDOI
TL;DR: The mathematical analysis and experimental results show that the GDNSA has good resistance to collusion attacks, and the CDNSA reduces the communication delay in spite of weakening a little anonymity.
Abstract: With the development of Internet applications, anonymous communication technology plays a very significant role in protecting personal privacy. As one of the most popular anonymous communication systems, I2P provides strong anonymity through its encryption and communication schemes. However, I2P does not consider the users’ preferences, which is difficult to meet the individual demands of specific users and then allows them to decide their anonymity. Thus, this paper proposes two novel user-oriented node selection algorithms that can effectively enhance the anonymity or reduce the communication delay over the I2P network. In order to choose proper nodes, we also investigate key factors to evaluate the nodes. Then, the basic node selection algorithm (BNSA) is proposed to group routing nodes and provide high-performance node candidates. Based on BNSA, the geographic-diversity-oriented node selection algorithm (GDNSA) and the communication-delay-oriented node selection algorithm (CDNSA) are proposed. These can improve the anonymity or communication performance of the I2P network. The GDNSA increases the attack difficulty by establishing tunnels that span multiple regions. In the meantime, the CDNSA reduces the communication delay of the tunnel by selecting the next hop node with the lowest communication delay. Finally, the mathematical analysis and experimental results show that the GDNSA has good resistance to collusion attacks, and the CDNSA reduces the communication delay in spite of weakening a little anonymity.

10 citations

Proceedings ArticleDOI
08 Jul 2014
TL;DR: This work designs a simple and efficient online scheme for scheduling cloud tasks requesting multiple resources, such as CPU and memory, and shows that the scheme outperforms the other schemes in terms of resource utilizations as well as average task queuing delays.
Abstract: We design a simple and efficient online scheme for scheduling cloud tasks requesting multiple resources, such as CPU and memory. The proposed scheme reduces the queuing delay of the cloud tasks by accounting for their execution time lengths. We also derive bounds on the average queuing delays, and evaluate the performance of our proposed scheme and compare it with those achievable under existing schemes by relying on real Google data traces. Using this data, we show that our scheme outperforms the other schemes in terms of resource utilizations as well as average task queuing delays.

10 citations


Cited by
More filters
Journal ArticleDOI
TL;DR: Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis.
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.).

13,246 citations

Christopher M. Bishop1
01 Jan 2006
TL;DR: Probability distributions of linear models for regression and classification are given in this article, along with a discussion of combining models and combining models in the context of machine learning and classification.
Abstract: Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

10,141 citations

01 Jan 2002

9,314 citations