In the smart grid, data communication between smart meters and utility servers should be authentic, private, have integrity while being accessible. To mitigate the risks of potential attacks, securing these two-way communications is crucial. Equally important is maintaining near real-time communication and avoiding significant delays when extra security levels are involved. Existing research on smart grids has not simultaneously tackled the issues of security, communication speed, and network scalability. In this work, we propose a novel delay-optimized blockchain solution for securing cryptographic communication between consumers and the utility in a smart grid. Our solution, based on EOS smart contracts, Edge computing, asymmetric Cryptographic functions, and Group signatures ($E E C G$), treats data communication as transactions that are asymmetrically encrypted and signed in groups before being stored on the EOS blockchain, ensuring confidentiality, privacy, availability, and low cost. The use of edge computing reduces the computational burden of smart meters, increases transaction speed, enhances data privacy, and improves scalability. Furthermore, an optimization problem for associating smart meters with edge nodes is formulated to minimize data exchange and processing delays over the blockchain, facilitating near real-time secure data access.

Smart grids are suscceptible to security vulnerabilities of cyber-physical systems due to the heterogeneity of their interconnected components. There are high risks associated with potential attacks targeting the two-way communication between the smart meters and the utility servers. It is vital to ensure that data communicated between consumers and the utility is not tampered with and is authentic, private, and available. Conventional security measures in traditional communication and network systems fail to secure the data communication aspect in the complex network that composes the advanced metering infrastructure (AMI). In this work, we propose ECC: a novel prevention approach based on Ethereum smart contracts, asymmetric cryptographic functions, and cloud services for securing the two-way communication between smart meters and utility servers. Ethereum blockchain is utilized as a building block where communicated data is treated as transactions encrypted and stored in a distributed fashion to ensure data availability, confidentiality, and privacy. We also augment the Ethereum architecture with cloud services to extend the number of allowable transactions, ensure the availability of the electricity data, and reduce the cost associated with Ethereum transactions. The conducted experiments illustrate the efficacy of ECC in terms of the achieved security properties. This paper shows that the Ethereum Blockchain coupled with Cloud services can improve the efficiency of a system solely based on the Ethereum Blockchain.

Learning response generation models constitute the main component of building open-domain dialogue systems. However, training open-domain response generation models requires large amounts of labeled data and pre-trained language generation models that are often nonexistent for low-resource languages. In this article, we propose a framework for training open-domain response generation models in low-resource settings. We consider Dialectal Arabic (DA) as a working example. The framework starts by warm-starting a transformer-based encoder-decoder with pre-trained language model parameters. Next, the resultant encoder-decoder model is adapted to DA by employing self-supervised pre-training on large-scale unlabeled data in the desired dialect. Finally, the model is fine-tuned on a very small labeled dataset for open-domain response generation. The results show significant performance improvements on three spoken Arabic dialects after adopting the framework’s three stages, highlighted by higher BLEU and lower Perplexity scores compared with multiple baseline models. Specifically, our models are capable of generating fluent responses in multiple dialects with an average human-evaluated fluency score above 4. Our data is made publicly available.

Dialogue generation is the automatic generation of a text response, given a user’s input. Dialogue generation for low-resource languages has been a challenging tasks for researchers. However, the advancements in deep learning models have made developing conversational agents that perform the tasks of dialogue generation not only possible, but also effective and helpful in many applications spanning a variety of domains. Nevertheless, work on conversational bots for low-resource languages such as the Arabic language is still limited due to various challenges, including the language structure, vocabulary, and the scarcity of its data resources. Meta-learning has been introduced before in the natural language processing (NLP) realm and showed significant improvements in many tasks; however, it has rarely been used in natural language generation (NLG) tasks and never in Arabic NLG. In this work, we propose a meta-learning approach for Arabic dialogue generation for fast adaptation on low-resource domains, namely, Arabic. We start by using existing pre-trained models; we then meta-learn the initial parameters on high-resource dataset before finetuning the parameters on the target tasks. We prove that the proposed model that employs meta-learning techniques improves generalization and enables fast adaptation of the transformer model on low-resource NLG tasks. We report gains in the BLEU-4 and improvements in Semantic textual Similarity (STS) metrics when compared to the existing state-of-the-art approach. We also do a further study on the effectiveness of the meta-learning algorithms on the response generation of the models.

With the advances in Natural Language Processing (NLP), the industry has been moving towards human-directed artificial intelligence (AI) solutions. Recently, chatbots and automated news generation have captured a lot of attention. The goal is to automatically generate readable text from tabular data or web data commonly represented in Resource Description Framework (RDF) format. The problem can then be formulated as Data-to-text (D2T) generation from structured non-linguistic data into human-readable natural language. Despite the significant work done for the English language, no efforts are being directed towards low-resource languages like the Arabic language. This work promotes the development of the first RDF data-to-text (D2T) generation system for the Arabic language while trying to address the low-resource limitation. We develop several models for the Arabic D2T task using transfer learning from large language models (LLM) such as AraBERT, AraGPT2, and mT5. These models include a baseline Bi-LSTM Sequence-to-Sequence (Seq2Seq) model, as well as encoder-decoder transformers like BERT2BERT, BERT2GPT, and T5. We then provide a detailed comparative study highlighting the strengths and limitations of these methods setting the stage for further advancement in the field. We also introduce a new Arabic dataset (AraWebNLG) that can be used for new model development in the field. To ensure a comprehensive evaluation, general-purpose automated metrics (BLEU and Perplexity scores) are used as well as task-specific human evaluation metrics related to the accuracy of the content selection and fluency of the generated text. The results highlight the importance of pre-training on a large corpus of Arabic data and show that transfer learning from AraBERT gives the best performance. Text-to-text pre-training using mT5 achieves second best performance results even with multilingual weights.

In many real-world scenarios, machine learning models fall short in prediction performance due to data characteristics changing from training on one source domain to testing on a target domain. There has been extensive research to address this problem with Domain Adaptation (DA) for learning domain invariant features. However, when considering advances for time series, those methods remain limited to the use of hard parameter sharing (HPS) between source and target models, and the use of domain adaptation objective function. To address these challenges, we propose a soft parameter sharing (SPS) DA architecture with representation learning while modeling the relation as non-linear between parameters of source and target models and modeling the adaptation loss function as the squared Maximum Mean Discrepancy (MMD) . The proposed architecture advances the state-of-the-art for time series in the context of activity recognition and in fields with other modalities, where SPS has been limited to a linear relation. An additional contribution of our work is to provide a study that demonstrates the strengths and limitations of HPS versus SPS. Experiment results showed the success of the method in three domain adaptation cases of multivariate time series activity recognition with different users and sensors.

One key objective of artificial intelligence involves the continuous adaptation of machine learning models to new tasks. This branch of continual learning is also referred to as lifelong learning (LL), where a major challenge is to minimize catastrophic forgetting, or forgetting previously learned tasks. While previous work on catastrophic forgetting has been focused on vision problems; this work targets time-series data. In addition to choosing an architecture appropriate for time-series sequences, our work addresses limitations in previous work, including the handling of distribution shifts in class labels. We present multi-objective learning with three loss functions to minimize catastrophic forgetting, prediction error, and errors in generalizing across label shifts, simultaneously. We build a multi-task autoencoder network with a hierarchical convolutional recurrent architecture. The proposed method is capable of learning multiple time-series tasks simultaneously. For cases where the model needs to learn multiple new tasks, we propose sequential learning, starting with tasks that have the best individual performances. This solution was evaluated on four benchmark human activity recognition datasets collected from mobile sensing devices. A wide set of baseline comparisons is performed, and an ablation analysis is run to evaluate the impact of the different losses in the proposed multi-objective method. The results demonstrate an up to 4% performance improvement in catastrophic forgetting compared to the use of loss functions in state-of-the-art solutions while demonstrating minimal losses compared to upper bound methods of traditional fine-tuning (FT) and multi-task learning (MTL).

Convolutional Neural Networks (CNN) have achieved a resounding success especially in computer vision and collaborative filtering. The general trend in CNN architectures has been to build deeper networks with a substantial number of convolution filters and several large feature maps. As a result, most of the current CNN inference routines are highly compute-intensive and have significant storage requirements. Field Programmable Gate Arrays (FPGAs) are among the most popular choices for accelerating CNN inference workloads as they can perform complex and massively parallel jobs. Recently, notable efforts have been made to distribute CNN inference workloads across multiple FPGAs [1]. These strategies, however, do not take into account variations in computational complexity across different layers of a CNN resulting in suboptimal performance gains. This work proposes an optimal distribution of CNN layers across different FPGA nodes while accounting for each layer’s performance to achieve maximum overall throughput.

Smart grids are being continually adopted as a replacement of the traditional power grid systems to ensure safe, efficient, and cost-effective power distribution. The smart grid is a heterogeneous communication network made up of various devices such as smart meters, automation, and emerging technologies interacting with each other. As a result, the smart grid inherits most of the security vulnerabilities of cyber systems, putting the smart grid at risk of cyber-attacks. To secure the communication between smart grid entities, namely the smart meters and the utility, we propose in this paper a communication infrastructure built on top of a blockchain network, specifically Ethereum. All two-way communication between the smart meters and the utility is assumed to be transactions governed by smart contracts. Smart contracts are designed in such a way to ensure that each smart meter is authentic and each smart meter reading is reported securely and privately. We present a simulation of a sample smart grid and report all the costs incurred from building such a grid. The simulations illustrate the feasibility and security of the proposed architecture. They also point to weaknesses that must be addressed, such as scalability and cost.

Epilepsy is a chronic medical condition that involves abnormal brain activity causing patients to lose control of awareness or motor activity. As a result, detection of pre-ictal states, before the onset of a seizure, can be lifesaving. The problem is challenging because it is difficult to discern between electroencephalogram signals in pre-ictal states versus signals in normal inter-ictal states. There are three key challenges that have not been addressed previously: (1) the inconsistent performance of prediction models across patients, (2) the lack of perfect prediction to protect patients from any episode, and (3) the limited amount of pre-ictal labeled data for advancing machine learning methods. This article addresses these limitations through a novel approach that uses adversarial examples with optimized tuning of a combined convolutional neural network and gated recurrent unit. Compared to the state of the art, the results showed an improvement of 3x in model robustness as measured in reduced variations and superior accuracy of the area under the curve, with an average increase of 6.7%. The proposed method also exhibited superior performance with other advances in the field of machine learning and customized for epilepsy prediction including data augmentation with Gaussian noise and multitask learning.

In this paper, we address the problem of recognizing the semantic human activities through the analysis of large dataset collected from users’ sensor-based smartphones. Our approach is unique in terms of covering a large number of activities that users could possibly engage in, and considering the multi-level-based classification model. Our model has three properties that never seemed to be addressed by existing approaches dealing with the same problem. These are: (1) comprehensiveness — in terms of the activity set, (2) accuracy — in terms of the activity classification, and (3) applicability — in terms of flexibility in being applied in real-life settings. Current approaches do not tackle all these properties. When tested on realistic dataset, our multi-level-based model achieved promising results despite the large number of activities being considered. When compared to similar approaches, our approach achieved comparable results in terms of accuracy and outperformed them in terms of the activity types, environment and settings covered, comprehensiveness, and applicability.

Success of Natural Language Processing (NLP) models, just like all advanced machine learning models, rely heavily on large -scale lexical resources. For English, English WordNet (EWN) is a leading example of a large-scale resource that has enabled advances in Natural Language Understanding (NLU) tasks such as word sense disambiguation, question answering, sentiment analysis, and emotion recognition. EWN includes sets of cognitive synonyms called synsets, which are interlinked by means of conceptual-semantic and lexical relations and where each synset expresses a distinct concept. However, other languages are still lagging behind in having large-scale and rich lexical resources similar to EWN. In this article, we focus on enabling the development of such resources for Arabic. While there have been efforts in developing an Arabic WordNet (AWN), the current version of AWN has its limitations in size and in lacking transliteration standards, which are important for compatibility with Arabic NLP tools. Previous efforts for extending AWN resulted in a lexicon, called ArSenL, that overcame the size and the transliteration standard limitation but was limited in accuracy due to the heuristic approach that only considered surface matching between the English definitions from the Standard Arabic Morphological Analyzer (SAMA) and EWN synset terms, and that resulted in inaccurate mapping of Arabic lemmas to EWN’s synsets. Furthermore, there has been limited exploration of other expansion methods due to expensive manual validation needed. To address these limitations of simultaneously having large-scale size with high accuracy and standard representations, the mapping problem is formulated as a link prediction problem between a large-scale Arabic lexicon and EWN, where a word in one lexicon is linked to a word in another lexicon if the two words are semantically related. We use a semi-supervised approach to create a training dataset by finding common terms in the large-scale Arabic resource and AWN. This set of data becomes implicitly linked to EWN and can be used for training and evaluating prediction models. We propose the use of a two-step Boosting method, where the first step aims at linking English translations of SAMA’s terms to EWN’s synsets. The second step uses surface similarity between SAMA’s glosses and EWN’s synsets. The method results in a new large-scale Arabic lexicon that we call ArSenL 2.0 as a sequel to the previously developed sentiment lexicon ArSenL. A comprehensive study covering both intrinsic and extrinsic evaluations shows the superiority of the method compared to several baseline and state-of-the-art link prediction methods. Compared to previously developed ArSenL, ArSenL 2.0 included a larger set of sentimentally charged adjectives and verbs. It also showed higher linking accuracy on the ground truth data compared to previous ArSenL. For extrinsic evaluation, ArSenL 2.0 was used for sentiment analysis and showed, here, too, higher accuracy compared to previous ArSenL.

The vision towards 5G and beyond is to provide remarkable performance enhancements that enable the launch of new services and markets in different industry verticals. Example scenarios of those services include indoor hotspot, broadband access in a crowd, and dense urban; many of which require very high data rates. Traffic offloading and device-to-device cooperation have been leveraged to expand the system capacity and coverage through using heterogeneous network technologies. In this work, we shed the light on the significant gains incurred when predicted short term network performance is factored into network splitting decisions over multiple wireless interfaces, while guaranteeing a desired quality of experience to end users. Accordingly, we allow dynamic use of multiple network interfaces taking into consideration their energy requirement and price models to deliver premium services and minimize the overall energy consumption and total cost. We develop a novel and efficient real-time traffic splitting approach that makes use of predicted bit rate of each network interface in addition to device-to-device cooperation to decide on the amount of video traffic to be delivered on each interface at every time slot. The proposed approach is validated and evaluated under realistic network conditions. Simulation results demonstrate substantial gains in terms of energy consumption, data cost and quality of user experience as compared to multiple alternative solutions.

Opinion-mining or sentiment analysis continues to gain interest in industry and academics. While there has been significant progress in developing models for sentiment analysis, the field remains an active area of research for many languages across the world, and in particular for the Arabic language, which is the fifth most-spoken language and has become the fourth most-used language on the Internet. With the flurry of research activity in Arabic opinion mining, several researchers have provided surveys to capture advances in the field. While these surveys capture a wealth of important progress in the field, the fast pace of advances in machine learning and natural language processing (NLP) necessitates a continuous need for a more up-to-date literature survey. The aim of this article is to provide a comprehensive literature survey for state-of-the-art advances in Arabic opinion mining. The survey goes beyond surveying previous works that were primarily focused on classification models. Instead, this article provides a comprehensive system perspective by covering advances in different aspects of an opinion-mining system, including advances in NLP software tools, lexical sentiment and corpora resources, classification models, and applications of opinion mining. It also presents future directions for opinion mining in Arabic. The survey also covers latest advances in the field, including deep learning advances in Arabic Opinion Mining. The article provides state-of-the-art information to help new or established researchers in the field as well as industry developers who aim to deploy an operational complete opinion-mining system. Key insights are captured at the end of each section for particular aspects of the opinion-mining system giving the reader a choice of focusing on particular aspects of interest.

While cloud computing has provided major benefits by maximizing the use of resources within a cloud, the current solutions still face many challenges. In this paper, we propose performance enhancements for cloud computations, provided by integrating hardware acceleration into the computation services. We extend the Hadoop framework by adding provisions for hardware acceleration using Field Programmable Gate Arrays (FPGAs). Hardware acceleration using energy efficient FPGAs is used as a service within the clouds or to offload computations when needed. It can provide additional sources of revenues, reduced operating costs, and increased resource utilization. We developed a custom Hadoop system and tested four applications commonly used for machine learning and deep learning. The results show the benefits of hardware acceleration and the high performance gains in smaller execution times and energy consumption.

In recent years, there has been a significant growth of context-aware applications, which extract the user's context from multiple embedded sensors in smartphones and wearable sensors. However, running multiple context-aware applications simultaneously causes extensive battery drainage for mobile devices. To alleviate the energy limitation in multi-context setting, we propose EGO: an ontology-based framework for group sensor selection to achieve synergy across applications while trading off energy consumption, accuracy, and delay in context recognition. A new context recognition ontology is designed to capture context recognition models. It captures the alternative groups of sensors for each context and the parameters for context recognition models. EGO includes an adaptive sensor selection mechanism that selects the appropriate sensors based on the current user state, the available resources, and the requested accuracies by running applications. The framework provides an open architecture that allows integration with other sensor optimization modules for sensor scheduling. EGO is validated using a real test-bed implemented on an Android platform that provides accessibility to on-board mobile sensors and external sensors. Results show that EGO provides better trade-off than previous state-of-the-art methods. Furthermore, EGO provides 63 percent energy saving with comparable accuracy when compared to the most accurate group of sensors.

Convolutional Neural Network (ConvNet or CNN) algorithms are characterized by a large number of model parameters and high computational complexity. These two requirements have made it challenging for implementations on resource-limited FPGAs. The challenges are magnified when considering designs for low-end FPGAs. While previous work has demonstrated successful ConvNet implementations with high-end FPGAs, this article presents a ConvNet accelerator design that enables the implementation of complex deep ConvNet architectures on resource-constrained FPGA platforms aimed at the IoT market. We call the design “FeatherNet” for its light resource utilization. The implementations are VHDL-based providing flexibility in design optimizations. As part of the design process, new methods are introduced to address several design challenges. The first method is a novel stride-aware graph-based method targeted at ConvNets that aims at achieving efficient signal processing with reduced resource utilization. The second method addresses the challenge of determining the minimal precision arithmetic needed while preserving high accuracy. For this challenge, we propose variable-width dynamic fixed-point representations combined with a layer-by-layer design-space pruning heuristic across the different layers of the deep ConvNet model. The third method aims at achieving a modular design that can support different types of ConvNet layers while ensuring low resource utilization. For this challenge, we propose the modules to be relatively small and composed of computational filters that can be interconnected to build an entire accelerator design. These model elements can be easily configured through HDL parameters (e.g., layer type, mask size, stride, etc.) to meet the needs of specific ConvNet implementations and thus they can be reused to implement a wide variety of ConvNet architectures. The fourth method addresses the challenge of design portability between two different FPGA vendor platforms, namely, Intel/Altera and Xilinx. For this challenge, we propose to instantiate the device-specific hardware blocks needed in each computational filter, rather than relying on the synthesis tools to infer these blocks, while keeping track of the similarities and differences between the two platforms. We believe that the solutions to these design challenges further advance knowledge as they can benefit designers and other researchers using similar devices or facing similar challenges. Our results demonstrated the success of addressing the design challenges and achieving low (30%) resource utilization for the low-end FPGA platforms: Zedboard and Cyclone V. The design overcame the limitation of designs targeted for high-end platforms and that cannot fit on low-end IoT platforms. Furthermore, our design showed superior performance results (measured in terms of [Frame/s/W] per Dollar) compared to high-end optimized designs.

Monitoring context depends on continuous collection of raw data from sensors which are either embedded in smart mobile devices or worn by the user. However, continuous sensing constitutes a major source of energy consumption; on the other hand, lowering the sensing rate may lead to missing the detection of critical contextual events. In this paper, we propose VCAMS: a Viterbi-based Context Aware Mobile Sensing mechanism that adaptively finds an optimized sensing schedule to decide when to trigger the sensors for data collection while trading off the sensing energy and the delay to detect a state change. The sensing schedule is adaptive from two aspects: 1) the decision rules are learned from the user's past behavior, and 2) these rules are updated over real time whenever there is a significant change in the user's behavior. VCAMS is validated using multiple experiments, which include evaluation of model success when considering binary and multi-user states. We also implemented VCAMS on an Android-based device to estimate its computational costs under realistic operational conditions. Test results show that our proposed strategy provides better trade-off than previous state-of-the-art methods under comparable conditions. Furthermore, the method provides 78 percent energy saving when compared to continuous sensing.

Deep neural networks have gained tremendous attention in both the academic and industrial communities due to their performance in many artificial intelligence applications, particularly in computer vision. However, these algorithms are known to be computationally very demanding for both scoring and model learning applications. State-of-the-art recognition models use tens of millions of parameters and have significant memory and computational requirements. These requirements have restricted the users of deep neural network applications to high-end, expensive, and power hungry IoT platforms to penetrate the deep learning markets. This paper presents work at the leading edge intersection of several evolving technologies, including emerging IoT platforms, Deep Learning, and Field-programmable Gate Array (FPGA) computing. We demonstrate a new minimalist design methodology that minimizes the utilization of FPGA resources and can run deep learning algorithms with over 60 million parameters. This makes particularly suitable for resource-constrained, low-end FPGA platforms.

Ultra dense networks and device-to-device communications are expected to play a major role in 5G networks to meet tremendous traffic requirements. In our work, we address traffic offloading in dense device-to-device cooperative heterogeneous networks with focus on use cases where a very large number of users request simultaneously common streaming content from a remote server with quality of service guarantees. We formulate an optimization problem to maximize the number of users served and reduce the number of access points deployed while satisfying a set of system constraints. The solution determines the best strategy for downloading the content either over long range connectivity from the access points or short range connectivity from peer mobile devices. Results are presented for various scenarios in a stadium setting to demonstrate the significant gains of optimized traffic offloading in ultra dense wireless networks.

Accurate sentiment analysis models encode the sentiment of words and their combinations to predict the overall sentiment of a sentence. This task becomes challenging when applied to morphologically rich languages (MRL). In this article, we evaluate the use of deep learning advances, namely the Recursive Neural Tensor Networks (RNTN), for sentiment analysis in Arabic as a case study of MRLs. While Arabic may not be considered the only representative of all MRLs, the challenges faced and proposed solutions in Arabic are common to many other MRLs. We identify, illustrate, and address MRL-related challenges and show how RNTN is affected by the morphological richness and orthographic ambiguity of the Arabic language. To address the challenges with sentiment extraction from text in MRL, we propose to explore different orthographic features as well as different morphological features at multiple levels of abstraction ranging from raw words to roots. A key requirement for RNTN is the availability of a sentiment treebank; a collection of syntactic parse trees annotated for sentiment at all levels of constituency and that currently only exists in English. Therefore, our contribution also includes the creation of the first Arabic Sentiment Treebank (A r S en TB) that is morphologically and orthographically enriched. Experimental results show that, compared to the basic RNTN proposed for English, our solution achieves significant improvements up to 8% absolute at the phrase level and 10.8% absolute at the sentence level, measured by average F1 score. It also outperforms well-known classifiers including Support Vector Machines, Recursive Auto Encoders, and Long Short-Term Memory by 7.6%, 3.2%, and 1.6% absolute respectively, all models being trained with similar morphological considerations.

While research on English opinion mining has already achieved significant progress and success, work on Arabic opinion mining is still lagging. This is mainly due to the relative recency of research efforts in developing natural language processing (NLP) methods for Arabic, handling its morphological complexity, and the lack of large-scale opinion resources for Arabic. To close this gap, we examine the class of models used for English and that do not require extensive use of NLP or opinion resources. In particular, we consider the Recursive Auto Encoder (RAE). However, RAE models are not as successful in Arabic as they are in English, due to their limitations in handling the morphological complexity of Arabic, providing a more complete and comprehensive input features for the auto encoder, and performing semantic composition following the natural way constituents are combined to express the overall meaning. In this article, we propose A R ecursive Deep Learning Model for O pinion M ining in A rabic (AROMA) that addresses these limitations. AROMA was evaluated on three Arabic corpora representing different genres and writing styles. Results show that AROMA achieved significant performance improvements compared to the baseline RAE. It also outperformed several well-known approaches in the literature.

In a previous work, we presented a system which combines active solid state drives and reconfigurable FPGAs (which we called reconfigurable active SSD nodes, or simply RASSD nodes) into a storage-compute node that can be used by a cloud data-center to achieve accelerated computations while running data intensive applications. To hide the complexity of accessing RASSD nodes from applications, we proposed in another work a middleware framework which handles all low-level interactions with the hardware. The Middleware Server (MWS), which manages a group of RASSD nodes, has the role of bridging the connection between a client and the nodes. In this paper, we present extensions to the MWS to enable it to operate within a collaborative cloud environment, and we develop a model to evaluate the performance of the collaborative MWS. This model represents a study of the utilization of three hardware resources of the MWS: CPU, memory, and network interface. For each, we derive the parameters that affect its operations, and propose formulas for its utilization. The results describe the capacity of a MWS, and hence can be used to decide on the number of MWSs in a collaborative cloud datacenter.

Modern mobile pervasive applications focus on context awareness that monitors a diverse range of personal domains. In order to infer contextual information, most of these applications require the collection of raw data from sensors which are either embedded in personal smartphones or worn by the user. Critical context-aware applications rely on continuous accurate monitoring of the user's current context. Continuous sensing mechanisms in sensors cost high energy consumption to support accurate contextual detection. Hence, there is a trade-off between the classification accuracy and the energy consumption. In this paper, we exploit the advantages of Deep Neural Network (DNN) with ensemble classification of other complementary machine learning approaches to determine the best sensor sampling frequency for the recognition of a given context. DNN relies on raw data for classification while the other complementary methods (such as Decision Tree and Naïve Bayes) use feature recognition to classify data. Therefore, our approach provides a range of granularity from raw data. We prove the robustness of our approach in experiments which show high accuracy in context recognition. In addition, real experiments demonstrate the energy gains of the proposed algorithm which reach 87% reduction in energy consumption when compared to continuous sensing.

This article introduces a sentiment analysis approach that adopts the way humans read, interpret, and extract sentiment from text. Our motivation builds on the assumption that human interpretation should lead to the most accurate assessment of sentiment in text. We call this automated process Human Reading for Sentiment (HRS). Previous research in sentiment analysis has produced many frameworks that can fit one or more of the HRS aspects; however, none of these methods has addressed them all in one approach. HRS provides a meta-framework for developing new sentiment analysis methods or improving existing ones. The proposed framework provides a theoretical lens for zooming in and evaluating aspects of any sentiment analysis method to identify gaps for improvements towards matching the human reading process. Key steps in HRS include the automation of humans low-level and high-level cognitive text processing. This methodology paves the way towards the integration of psychology with computational linguistics and machine learning to employ models of pragmatics and discourse analysis for sentiment analysis. HRS is tested with two state-of-the-art methods; one is based on feature engineering, and the other is based on deep learning. HRS highlighted the gaps in both methods and showed improvements for both.

Huge amounts of data and personal information are being sent to and retrieved from web applications on daily basis. Every application has its own confidentiality and integrity policies. Violating these policies can have broad negative impact on the involved company's financial status, while enforcing them is very hard even for the developers with good security background. In this paper, we propose a framework that enforces security-by-construction in web applications. Minimal developer effort is required, in a sense that the developer only needs to annotate database attributes by a security class. The web application code is then converted into an intermediary representation, called Extended Program Dependence Graph (EPDG). Using the EPDG, the provided annotations are propagated to the application code and run against generic security enforcement rules that were carefully designed to detect insecure information flows as early as they occur. As a result, any violation in the data's confidentiality or integrity policies is reported. As a proof of concept, two PHP web applications, Hotel Reservation and Auction, were used for testing and validation. The proposed system was able to catch all the existing insecure information flows at their source. Moreover and to highlight the simplicity of the suggested approaches vs. Existing approaches, two professional web developers assessed the annotation tasks needed in the presented case studies and provided a very positive feedback on the simplicity of the annotation task.

Heterogeneous networks are expected to play a major role towards meeting the exploding traffic demand over cellular systems. Particularly, existing WiFi hotspots will be dynamically utilized to offload the traffic of cellular mobile subscribers. This will be further facilitated by forthcoming advances in mobile device capabilities that will include the ability to operate multiple wireless interfaces simultaneously. To this end, we focus in this work on cellular/WiFi heterogeneous networks with traffic splitting where a mobile device can utilize existing cellular and WiFi links simultaneously to achieve various performance gains. We propose a multi-objective approach for traffic splitting that captures the tradeoffs between throughput maximization on one hand and battery energy minimization on the other hand. We evaluate the proposed approach using parameters determined via experimental measurements using Samsung Galaxy SIII mobile devices. Results are presented for various scenarios in order to quantify and analyze the throughput-energy tradeoffs of traffic splitting in cellular/WiFi heterogeneous networks.

In this article, we propose new extensions to Hadoop to enable clusters of reconfigurable active solid-state drives (RASSDs) to process streaming data from SSDs using FPGAs. We also develop an analytical model to estimate the performance of RASSD clusters running under Hadoop. Using the Hadoop RASSD platform and network simulators, we validate our design and demonstrate its impact on performance for different workloads taken from Stanford's Phoenix MapReduce project. Our results show that for a hardware acceleration factor of 20×, compute-intensive workloads processing 153MB of data can run up to 11× faster than a standard Hadoop cluster.

The goal of this brief is to present a unique top-down design methodology for developing energy-aware algorithms based on energy profiling. The key idea revolves around identifying and measuring components of code with high energy consumption. There are two major contributions of this brief: 1) a method for identifying components with high energy consumption in compute-intensive applications. To this end, we target operations called kernels, which are frequently used operations in the algorithm; 2) a method for estimating software energy for the identified software components, in particular for kernels and load/store operations. The energy evaluation method involves isolated code with assembly injection. Furthermore, to ensure reliable results, we use physical energy measurements conducted on specially instrumented circuit boards to provide actual and not just simulated measurements. To evaluate the proposed methods, we conducted two case studies using data mining algorithms: K-nearest neighbors and linear regression. The results highlight the contributions of kernels and memory energy to total energy.

A novel and rapidly growing area of research concerns data-intensive applications and the technical challenges that accompany it. One of those challenges is developing approaches and mechanisms that render high performance in processing and storing data. We joined this research effort by proposing a reconfigurable active solid state drives (RASSD) system that deals with such applications, through employing basic hardware, namely FPGA's connected to SSD's, to service the above applications as processing nodes, and take advantage of the close proximity between storage and processing. In this paper, we propose an intelligent middleware system for interfacing workstation-based and mobile applications to the distributed RASSD system. In order to provide high performance in terms of time and functionality, the middleware manages the data processing on the RASSD nodes through special pieces of code that we call drivelets, along with FPGA configuration files (bitstreams). Another important responsibility of the proposed middleware architecture lies in the unguided management of applications' flows, where it uses an intelligent script-parsing mechanism to turn one general request from the client into a sequence of operations needed to generate the required results. The middleware design allows for the integration of mobile applications into the overall architecture of the RASSD system, and allowing them to run data intensive applications that otherwise it is unfeasible for them to execute. We validate our design by comparing it to an existing middleware architecture, and present two use-cases with their results and discussion.

The proliferation of powerful smart devices is revolutionizing mobile computing systems. A particular set of applications that is gaining wide interest is recommender systems. Recommender systems provide their users with recommendations on variety of personal and relevant items or activities. They can play a significant role in today's life whether in E-commerce or for daily decisions that we need to make. We introduce a hybrid approach for solving the problem of finding the ratings of unrated items in a user-item ranking matrix through a weighted combination of user-based and item-based collaborative filtering. The proposed technique provides improvements in addressing two major challenges of recommender systems: accuracy of recommender systems and sparsity of data by simultaneously incorporating users' correlations and items ones. The evaluation of the system shows superiority of the solution compared to stand-alone user-based collaborative filtering or item-based collaborative filtering.

Cloud computing is increasingly becoming a desirable and foundational element in international enterprise computing. There are many companies which design, develop, and offer cloud technologies. However, cloud providers are still like lone islands. While current cloud computing models have provided significant benefits of maximizing the use of resources within a cloud, the current solutions still face many challenges including the lack of cross-leverage of available resources across clouds, the need to move data between clouds in some cases, and the lack of a global efficient cooperation between clouds. In [1], we addressed some of these challenges by providing an approach that enables various cloud providers to cooperate in order to execute, together, common requests. In this paper, we illustrate several enhancements to our work in [1] which focus on integrating hardware acceleration with the cloud services. We extend the Hadoop framework by adding provisions for hardware acceleration with Field Programmable Gate Arrays (FPGAs) within the cloud, for multi-cloud interaction, and for global cloud management. Hardware acceleration is used to offload computations when needed or as a service within the clouds. It can provide additional sources of revenues, reduced operating costs, and increased resource utilization. We derive a mathematical model for evaluating the performance of the most important entity in our system under various conditions.

Port scanning is the most popular reconnaissance technique attackers use to discover services they can break into. Port scanning detection has received a lot of attention by researchers. However a slow port scan attack can deceive most of the existing Intrusion Detection Systems (IDS). In this paper, we present a new, simple, and efficient method for detecting slow port scans. Our proposed method is mainly composed of two phases: (1) a feature collection phase that analyzes network traffic and extracts the features needed to classify a certain IP as malicious or not. (2) A classification phase that divides the IPs, based on the collected features, into three groups: normal IPs, suspicious IPs and scanner IPs. The IPs our approach classify as suspicious are kept for the next (K) time windows for further examination to decide whether they represent scanners or legitimate users. Hence, this approach is different than the traditional approach used by IDSs that classifies IPs as either legitimate or scanners, and thus producing a high number of false positives and false negatives. A small Local Area Network was put together to test our proposed method. The experiments show the effectiveness of our proposed method in correctly identifying malicious scanners when both normal and slow port scan were performed using the three most common TCP port scanning techniques. Moreover, our method detects malicious scanners that are otherwise not detected using well known IDSs such as Snort.

The most challenging issue in WiMAX network planning is to measure and enhance the Quality of Service (QoS) of WiMAX networks. In this paper, a comprehensive WiMAX simulator is proposed to evaluate the performance of the system. The key parts of the simulator are described including end-to-end communication path, traffic generation, Medium Access Control (MAC) and Physical (PHY) layers, resource allocation, frame construction and configuration options such as Adaptive Modulation and Coding (AMC). Several experiments are conducted to assess different scenarios while varying one or more of the following: input traffic size, traffic load, presence of fragmentation and AMC. The results show the scalability of the system as it can support a large number of users while showing the real-life representation of the traffic models. The simulations also show the flexibility of implementing AMC schemes according to desired distributions. The high accuracy of the simulator is shown by comparing the simulator results to theoretical expected values.

In this paper, we propose to combine active solid state drives and reconfigurable FPGAs into a storage-compute node to use as a building block in a distributed, high performance computation platform for data intensive applications. We propose a complete framework for middleware functionality through an API abstraction layer that hides the complexity of accessing and processing data stored on these distributed nodes, thus allowing programmers to focus on the application, and not the underlying specialized architecture. The application in turn is re-architected to maximize its performance by delegating selected computations down to the storage-compute node. We present preliminary results measured on a real hardware prototype of a single-node. These results show that our proposed architecture provides more than a 2× improvement in performance over non-reconfigurable active-disk architectures that use electromechanical disks for storage and a 6× improvement in performance over a platform that performs the computation on the middle server.

Minimizing energy consumption is an urgent and challenging problem. As in any communication system, high energy efficiency in WiMAX systems should be maintained by increasing resource efficiency. Thus, WiMAX resources should be properly utilized by optimizing the construction of downlink (DL) bursts. This paper proposes an energy-efficient scheme that maximizes the use of resources at the base station (BS) by reducing the energy wasted caused by sending padding bits instead of useful data. The problem was formulated as nonlinear integer programming model. Due to the complexity of the problem, this paper presents first the formulation of the base model for optimal DL bursts construction problem assuming the packet is represented by one burst. Then, the formulation is expanded to allow the representation of packets by several bursts. The results show an improvement in data packing that maximizes the utilization of frames, and minimizes energy wastage.

The increasing demand for mobile devices and high performance computing has made energy consumption a main issue in computer technology. Mobile devices require extended battery life, but the available technology still puts limits on the need for recharging the devices. High performance computing has a high price tag on energy for compute-intensive applications such as data mining. As a result, optimizations at various layers of the computer platform are becoming necessary to minimize energy usage or extend the time before a battery needs to be recharged. This paper focuses on back-propagation neural network algorithm, one of the popular compute-intensive data mining algorithms. The goal is to present a design methodology for developing an energy aware algorithm. The key idea revolves around identifying operations called kernels, which are frequently used in the algorithm, and that can be implemented in hardware. Optimizing these kernels for performance or energy would then lead to a major impact in these areas. These kernels are analyzed for their impact on the overall application energy using energy-based asymptotic analysis. The methodology then considers additional optimizations not related to kernels, but are specific to the back-propagation algorithm. Suggestions are provided to improve the performance and reduce energy consumption. Experiments show that there are significant potentials in energy reduction through the use of alternative lower energy kernels or through custom optimizations with tradeoffs in the accuracy of the results.

Abstract Wired and wireless networks are being attacked and hacked on continuous basis. One of the critical pieces of information the attacker needs to know is the open ports on the victim's machine, thus the attacker does what is called port scanning. Port scanning is considered one of the dangerous attacks that intrusion detection tries to detect. Snort, a famous network intrusion detection system (NIDS), detects a port scanning attack by combining and analyzing various traffic parameters. Because these parameters cannot be easily combined using a mathematical formula, fuzzy logic can be used to combine them; fuzzy logic can also reduce the number of false alarms. This paper presents a novel approach, based on fuzzy logic, to detect port scanning attacks. A fuzzy logic controller is designed and integrated with Snort in order to enhance the functionality of port scanning detection. Experiments are carried out in both wired and wireless networks. The results show that applying fuzzy logic adds to the accuracy of determining bad traffic. Moreover, it gives a level of degree for each type of port scanning attack. Copyright © 2010 John Wiley & Sons, Ltd.

Artificial intelligence and wireless technologies have progressed rapidly in the last decade driven by technology evolution and new emerging applications. The automotive industry has the opportunity to benefit from progress in these fields. In this paper, we propose an extensible software framework that can be implemented on all vehicles. It comprehends the capabilities supporting wireless communication and analytics for collaborative intelligence among vehicles and data mining solutions. The architecture has four key modules for: interface to on-board controls, external wireless communication, artificial intelligence, and an internal virtual bus. The design is implemented as a framework so that new applications can be developed with minimal additions to achieve high end solutions. We demonstrate the effectiveness of the architecture with vehicle safety applications, and illustrate the use of collaborative intelligence. A potential implementation is proposed with multi-threads lending itself to parallel programming and high performance computing for real-time response.

We illustrate how VSpyware — Vehicular Spyware — may jeopardize the integrity of vehicular systems. We propose a complete framework to protect vehicles against this threat based on a generic five-level protection scheme and customize it for the standardized and open specifications of AUTOSAR. We then inspect the vulnerabilities of the embedded operating systems, specifically OSEK OS, which is adopted by AUTOSAR, and propose methods to implement protection at this level. Finally, we show how our design thwarts VSpyware and VMalware attacks and protects the privacy and security of drivers and passengers.

Minimal configuration and quick deployment of ad hoc networks make it suitable for numerous applications such as emergency situations, border monitoring, and military missions, etc. For such ad hoc networks to fulfill their mission in a timely manner, they should be able to establish a connection between nodes and to maintain this connection until the communication halts. Establishing a connection is achieved by using a routing protocol, and maintaining it is achieved by having a resilient fault tolerant network. In this paper, we propose a network design scheme that incorporates these features. We first propose a special network topology that is unique in terms of how nodes are interconnected. After constructing the initial topology, we propose a distributed routing protocol that allows any two sites to communicate by traversing at most 2 nodes regardless of the network size. We conducted both simulation study and theoretical analysis; the results show that the proposed scheme is resilient to network dynamics and has high quality as well as efficient routing.

Abstract In the communication sector, the optimal objective is to equate quality and cost. The technologies that best serve these objectives are Wireless Access Technologies since they are easily deployed and capable of reaching and serving customers everywhere in a cost effective way. In this paper, we examine the communication options, and account for a country's geography to propose optimal WiMax planning keeping in mind the security concerns that are inherent in wireless communication. To perform WiMax radio network planning, we use a network simulation tool from ATDI called ICS Telecom. Our approach offers all users a minimum bandwidth of 1.4 Mbps as well as a coverage that exceeds 90% for all indoor users in the area under study. We optimize the network by iteratively minimizing the number of base stations required, and equivalently minimizing the cost, while maximizing the coverage for the subscribers. We also analyze the impact of security on the performance of WiMax. More specifically, we use well‐known simulation software called Qualnet to simulate a WiMax environment under different security protocols and encryption scenarios. We then analyze the results to determine the impact of the added security features on the data rates between the base station(s) subscribers. The results of our proposed WiMax planning approach and the conducted security experiments showed that efficient deployment and coverage plans could be achieved for big cities as well as for rural areas. Copyright © 2009 John Wiley & Sons, Ltd.