Background: Large Language Models (LLMs) are predominantly trained and aligned in ways that reinforce Westerncentric epistemologies and socio-cultural norms, leading to cultural homogenization and limiting their ability to reflect global civilizational plurality. Existing benchmarking frameworks fail to adequately capture this bias, as they rely on rigid, closed-form assessments that overlook the complexity of cultural inclusivity. Objectives: To address this cultural bias problem, we introduce WorldView-Bench, a benchmark designed to evaluate Global Cultural Inclusivity (GCI) in LLMs by analyzing their ability to accommodate diverse worldviews. Methods: Our approach is grounded in the Multiplex Worldview proposed by Senturk et al., which distinguishes between Uniplex models, reinforcing cultural homogenization, and Multiplex models, which integrate diverse perspectives. WorldViewBench measures Cultural Polarization, the exclusion of alternative perspectives, through free-form generative evaluation rather than conventional categorical benchmarks. We implement applied multiplexity through two intervention strategies: (1) Contextually-Implemented Multiplex LLMs, where system prompts embed multiplexity principles, and (2) Multi-Agent System (MAS)-Implemented Multiplex LLMs, where multiple LLM agents representing distinct cultural perspectives collaboratively generate responses. Results: Our results demonstrate a significant increase in Perspectives Distribution Score (PDS) entropy from 13% at baseline to 94% with MAS-Implemented Multiplex LLMs, alongside a shift toward positive sentiment (67.7%) and enhanced cultural balance. Conclusions: The success of multiplex-aware evaluation in WorldView-Bench demonstrates that cultural bias in LLMs can be meaningfully measured and mitigated through structured worldview diversity. We expect this to pave the way for more inclusive, globally representative, and ethically aligned AI systems.

The advent of Generative Artificial Intelligence (GenAI) has sparked significant interest in education, offering ways to support learning, personalize student experiences, and boost engagement. Generative AI holds the promise of transforming education with personalized learning, instant feedback, and assistance with complex problem-solving. However, its integration into classrooms requires careful management due to ethical concerns, misinformation risks, and potential misuse. While many articles explore the potential of generative AI in education, empirical studies on its real-world classroom use are limited. This paper presents an experience report on deploying a customized GPT-powered chatbot at Qatar University to support learning in two undergraduate courses: Data and Computer Communications Networks (technical) and Computer Ethics. Working across these diverse courses allows a thorough analysis of generative AI's strengths and weaknesses in different academic contexts, offering a comprehensive evaluation of its applicability and effectiveness. We used a mixed-methods approach with over 100 students, collecting quantitative and qualitative data via questionnaires to assess the chatbot's impact. Findings are analyzed with established theoretical frameworks to contextualize the pedagogical impact of generative AI, aligned with UNESCO guidelines for ethical integration. This paper details the chatbot's technical customization to meet course-specific needs, provides evidence-based insights into practical challenges and opportunities, and offers strategic recommendations for effective AI-assisted pedagogy. Directions for further research are also outlined to explore and refine the role of generative AI in classroom settings. By examining two distinct courses, this study demonstrates how generative AI can be adapted across academic disciplines for more nuanced applications in education.

As large language models (LLMs) like GPT-4 and Llama 3 become integral to educational contexts, concerns are mounting over the cultural biases, power imbalances, and ethical limitations embedded within these technologies. Though generative AI tools aim to enhance learning experiences, they often reflect values rooted in Western, Educated, Industrialized, Rich, and Democratic (WEIRD) cultural paradigms, potentially sidelining diverse global perspectives. This paper proposes a framework to assess and mitigate cultural bias within LLMs through applied multiplexity. Multiplexity, inspired by Senturk et al. and rooted in Islamic and other wisdom traditions, emphasizes the coexistence of diverse cultural viewpoints, supporting a multilayered epistemology that integrates empirical sciences and normative values. Our analysis reveals that LLMs frequently exhibit cultural polarization in both overt responses and subtle contextual cues. To address these biases, we propose two strategies: Contextually-Implemented Multiplex LLMs, which embed multiplex principles directly into the system prompt, and Multi-Agent System (MAS)-Implemented Multiplex LLMs, where multiple LLM agents representing distinct cultural viewpoints collaboratively generate balanced responses. Our findings demonstrate that as mitigation strategies evolve from contextual prompting to MAS-implementation, cultural inclusivity markedly improves, evidenced by a significant rise in the Perspectives Distribution Score (PDS) and a PDS Entropy increase from 3.25% at baseline to 98% with the MAS-Implemented Multiplex LLMs. Sentiment analysis shows a shift towards positive sentiment across cultures, with the MAS-Implemented Multiplex LLMs achieving 0% negative sentiment. This study establishes a baseline for assessing and fostering cultural inclusivity in educational AI, laying the groundwork for a globally pluralistic approach that respects diverse cultural perspectives.

Multi-Agent Large Language Models (LLMs) are gaining attention for their ability to harness collective intelligence in complex problem-solving, decision-making, and planning tasks. This aligns with the wisdom of crowds concept, where diverse agents collectively generate effective solutions, making them well-suited for educational settings. Senior design projects, pivotal in engineering education, integrate theoretical knowledge with practical application, fostering critical thinking, teamwork, and real-world problem-solving skills. These projects often involve multidisciplinary considerations and conflicting objectives, such as optimizing technical performance while addressing ethical, social, and environmental concerns. In this paper, we explore a framework where distinct LLM agents embody expert perspectives, including problem formulation, system complexity, societal and ethical considerations, and project management. These agents engage in rich, collaborative dialogues, leveraging multi-agent system principles like coordination, cooperation, and negotiation. Prompt engineering is employed to create diverse personas, simulating human engineering teams and incorporating swarm AI principles to balance contributions efficiently. To evaluate the framework, we analyzed six senior capstone project proposals from engineering and computer science, comparing Multi-Agent and single-agent LLMs using metrics developed with engineering faculty and widely used NLP-based measures. These metrics assess technical quality, ethical considerations, social impact, and feasibility, aligning with the educational objectives of engineering design. Our findings suggest that Multi-Agent LLMs can provide a richer, more inclusive problem-solving environment compared to single-agent systems with 89% alignment with engineering-faculty scores, offering a promising tool for enhancing the educational experience of engineering and computer science students by simulating the complexity and collaboration of real-world engineering and computer science practice. By supporting senior design projects, this tool not only aids in achieving academic excellence but also prepares students for the multifaceted challenges they will face in their professional engineering careers. We have open-sourced our framework for further development and adaptation on GitHub11Copilot is available at GitHub Repository: https://github.com/AbdullahMushtaq78/Multi-Agent-SDP-Copliot.

Abstract Semantic understanding of roadways is a key enabling factor for safe autonomous driving. However, existing autonomous driving datasets provide well-structured urban roads while ignoring unstructured roadways containing distress, potholes, water puddles, and various kinds of road patches i.e., earthen, gravel etc. To this end, we introduce Road Region Segmentation dataset (R2S100K)—a large-scale dataset and benchmark for training and evaluation of road segmentation in aforementioned challenging unstructured roadways. R2S100K comprises 100K images extracted from a large and diverse set of video sequences covering more than 1000 km of roadways. Out of these 100K privacy respecting images, 14,000 images have fine pixel-labeling of road regions, with 86,000 unlabeled images that can be leveraged through semi-supervised learning methods. Alongside, we present an Efficient Data Sampling based self-training framework to improve learning by leveraging unlabeled data. Our experimental results demonstrate that the proposed method significantly improves learning methods in generalizability and reduces the labeling cost for semantic segmentation tasks. Our benchmark will be publicly available to facilitate future research at https://r2s100k.github.io/ .

The concept of the Metaverse has sparked great interest as a futuristic virtual space that provides immersive experiences and social interactions through digital avatars. However, using avatars in the Metaverse raises privacy concerns that require innovative solutions to ensure the safety of users. In this article, we conducted a systematic review using the PRISMA method to identify work discussing privacy issues related to avatars in the Metaverse and efforts to provide safer virtual environments for users. Our review revealed two main avatar-related privacy issues: threats related to the user’s identity, such as the disclosure of personal details, and social threats, such as harassment. We also reviewed different proposed solutions to these problems, categorized into three main groups: altering user representation, providing safety options to users, and leveraging AI techniques to detect and mitigate issues. While these solutions promise a safer Metaverse for users, there are inherent limitations that require further advancements. To fill the existing research gap and create a safer Metaverse, we suggest improving safety features for users, finding a balance between user experience and privacy, increasing the use of AI to create safer environments while considering user concerns, and enhancing reporting methods available in the Metaverse. Our review emphasizes the need for more advanced research and development to tackle avatar privacy challenges in the Metaverse.

To combat the rapid spread of misinformation on social networks, automated misinformation detection systems based on deep neural networks (DNNs) have been developed. However, these tools are often proprietary and lack transparency, which limits their usefulness. Furthermore, privacy concerns limit data sharing by data owners as well as by data-driven misinformation-detection services. Although data encryption techniques can help address privacy concerns in DNN inference, there is a challenge to the seamless integration of these techniques due to the encryption errors induced by cascaded encrypted operations, as well as a mismatch between the tools used for DNNs and cryptography. In this paper, we make two-fold contributions. First, we study the noise bounds of homomorphic encryption (HE) operations as error propagation in DNN layers and derive two properties that, if satisfied by the layer, will considerably reduce the output error. We identify that$L_{2}$regularization and sigmoid activation satisfy these properties and validate our hypothesis, for instance, replacing ReLU with sigmoid reduced the output error by$10^{6}\times$(best case) to$10\times$(worst case). Second, we extend the Python encryption library TenSeal by enabling the automatic conversion of a TensorFlow DNN into an encryption-compatible DNN with a few lines of code. These contributions are significant as encryption-friendly DL architectures are sorely needed to close the gap between DL-in-research and DL-in-practice.

This article outlines the architecture of autonomous driving and related complementary frameworks from the perspective of human comfort. The technical elements for measuring autonomous vehicle (AV) user comfort and psychoanalysis are listed here. At the same time, this article introduces the technology related to the structure of automatic driving and the reaction time of automatic driving. We also discuss the technical details related to the automatic driving comfort system, the response time of the AV driver, the comfort level of the AV, motion sickness, and related optimization technologies. The function of the sensor is affected by various factors. Since the sensor of automatic driving mainly senses the environment around a vehicle, including “the weather” which introduces the challenges and limitations of second-hand sensors in AVs under different weather conditions. The comfort and safety of autonomous driving are also factors that affect the development of autonomous driving technologies. This article further analyzes the impact of autonomous driving on the user’s physical and psychological states and how the comfort factors of AVs affect the automotive market. Also, part of our focus is on the benefits and shortcomings of autonomous driving. The goal is to present an exhaustive overview of the most relevant technical matters to help researchers and application developers comprehend the different comfort factors and systems of autonomous driving. Finally, we provide detailed automated driving comfort use cases to illustrate the comfort-related issues of autonomous driving. Then, we provide implications and insights for the future of autonomous driving.

The metaverse is a nascent concept that envisions a virtual universe, a collaborative space where individuals can interact, create, and participate in a wide range of activities. Privacy in the metaverse is a critical concern as the concept evolves and immersive virtual experiences become more prevalent. The metaverse privacy problem refers to the challenges and concerns surrounding the privacy of personal information and data within Virtual Reality (VR) environments as the concept of a shared VR space becomes more accessible. Metaverse will harness advancements from various technologies such as Artificial Intelligence (AI), Extended Reality (XR) and Mixed Reality (MR) to provide personalized and immersive services to its users. Moreover, to enable more personalized experiences, the metaverse relies on the collection of fine-grained user data that leads to various privacy issues. Therefore, before the potential of the metaverse can be fully realized, privacy concerns related to personal information and data within VR environments must be addressed. This includes safeguarding users’ control over their data, ensuring the security of their personal information, and protecting in-world actions and interactions from unauthorized sharing. In this paper, we explore various privacy challenges that future metaverses are expected to face, given their reliance on AI for tracking users, creating XR and MR experiences, and facilitating interactions. Moreover, we thoroughly analyze technical solutions such as differential privacy, Homomorphic Encryption, and Federated Learning and discuss related sociotechnical issues regarding privacy.

Recent works have shown that deep learning (DL) models can effectively learn city-wide crowd-flow patterns, which can be used for more effective urban planning and smart city management. However, DL models have been known to perform poorly on inconspicuous adversarial perturbations. Although many works have studied these adversarial perturbations in general, the adversarial vulnerabilities of deep CFP models in particular have remained largely unexplored. In this paper, we perform a rigorous analysis of the adversarial vulnerabilities of DL-based CFP models under multiple threat settings, making three-fold contributions; 1) we propose CaV-detect by formally identifying two novel properties—ConsistencyandValidity—of the CFP inputs that enable thedetection of standard adversarial inputs with 0% false acceptance rate (FAR); 2) we leverage universal adversarial perturbations and an adaptive adversarial loss to present adaptive adversarial attacks to evade CaV-detect defense; 3) we propose CVP, aConsistent,Valid andPhysically-realizable adversarial attack, that explicitly inducts the consistency and validity priors in the perturbation generation mechanism. We find out that although the crowd-flow models are vulnerable to adversarial perturbations, it is extremely challenging to simulate these perturbations in physical settings, notably when CaV-detect is in place. We also show that CVP attack considerably outperforms the adaptively modified standard attacks in FAR and adversarial loss metrics. We conclude with useful insights emerging from our work and highlight promising future research directions.

This paper presents a primer for learners interested in self-study of robotics, a multidisciplinary field that involves the design, construction, and operation of robots having wide-ranging applications in industries such as manufacturing, automation, business, and education. Acquiring knowledge and skills in robotics can lead to expertise in the highly in-demand Internet of Things (IoT) and cyber-physical systems (CPS) industries. The paper presents a curated curriculum based on existing resources, covering fundamental concepts and technologies such as kinematics, control systems, sensor design and integration, and robotics programming. It also discusses various resources including online courses, textbooks, software tools, hardware components, and real or virtual robots to support self-study, as well as offering tips for success such as setting clear goals, applying knowledge through exercises and projects, seeking support from a community of learners or experts, staying up-to-date with new developments, and seeking additional resources or help when needed. The field of robotics is complex and wide-ranging, with a high bar to entry, especially for those without access to engineering or computing courses. Our recommendations aim to help those interested in self-study in robotics maximize their learning and progress towards their goals.

Engineering education is constantly evolving to keep up with the latest technological developments and meet the changing needs of the engineering industry. One promising development in this field is the use of generative artificial intelligence technology, such as the ChatGPT conversational agent. ChatGPT has the potential to offer personalized and effective learning experiences by providing students with customized feedback and explanations, as well as creating realistic virtual simulations for hands-on learning. However, it is important to also consider the limitations of this technology. ChatGPT and other generative AI systems are only as good as their training data and may perpetuate biases or even generate and spread misinformation. Additionally, the use of generative AI in education raises ethical concerns such as the potential for unethical or dishonest use by students and the potential unemployment of humans who are made redundant by technology. While the current state of generative AI technology represented by ChatGPT is impressive but flawed, it is only a preview of what is to come. It is important for engineering educators to understand the implications of this technology and study how to adapt the engineering education ecosystem to ensure that the next generation of engineers can take advantage of the benefits offered by generative AI while minimizing any negative consequences.

This study explores the topics and trends of teaching AI ethics in higher education, using Latent Dirichlet Allocation as the analysis tool. The analyses included 166 courses from 105 universities around the world. Building on the uncovered patterns, we distil a model of current pedagogical practice, the BAG model (Build, Assess, and Govern), that combines cognitive levels, course content, and disciplines. The study critically assesses the implications of this teaching paradigm and challenges practitioners to reflect on their practices and move beyond stereotypes and biases.

The Metaverse has introduced a vast virtual environment where Artificial Intelligence Generated Content (AIGC) plays a crucial role in content creation. However, the increasing popularity of AIGC has raised concerns regarding its potential misuse and the need for effective detection methods to ensure content integrity. Security and privacy threats in the Metaverse have garnered significant attention, but the specific risks of AIGC in this context remain understudied. This paper aims to analyze the potential risks of AIGC in the Metaverse and explore mitigation techniques. By addressing these challenges, the paper contributes to a comprehensive understanding of AIGC's implications in the Metaverse.

Self-driving cars have gained widespread attention in recent years due to their potential to revolutionize the transportation industry. However, their success critically depends on the ability of reinforcement learning (RL) algorithms to navigate complex environments safely. In this paper, we investigate the potential security risks associated with end-to-end deep RL (DRL) systems in autonomous driving environments that rely on visual input for vehicle control, using the open-source Duckietown platform for robotics and self-driving vehicles. We demonstrate that current DRL algorithms are inherently susceptible to attacks by designing a general state adversarial perturbation and a reward tampering approach. Our strategy involves evaluating how attacks can manipulate the agent's decision-making process and using this understanding to create a corrupted environment that can lead the agent towards low-performing policies. We introduce our state perturbation method, accompanied by empirical analysis and extensive evaluation, and then demonstrate a targeted attack using reward tampering that leads the agent to catastrophic situations. Our experiments show that our attacks are effective in poisoning the learning of the agent when using the gradient-based Proximal Policy Optimization algorithm within the Duckietown environment. The results of this study are of interest to researchers and practitioners working in the field of autonomous driving, DRL, and computer security, and they can help inform the development of safer and more reliable autonomous driving systems.

People with visual impairments use different kinds of assistive technologies in their daily lives for various activities such as navigation, reading texts, etc. Technological advancements in recent years have enabled developers to actively deploy and expeditiously operate assistive applications in embedded devices. In this paper, we have proposed an assistive wearable system called VisualAid+ for people with visual impairments. Leveraging the power of Tiny ML and Edge AI, a portable wearable assistive system is developed for people with visual impairments for object detection and visual scene narration. A hierarchical approach has been followed to take advantage of the complex prediction model (TensorFlow model), and Lite models (TensorFlow Lite and TensorFlow Lite Micro). This makes the system capable of on-device, as well as server-side, inference. The proposed VisualAid+ system consists of a Raspberry Pi device, a computer as a server, a power source, and two cameras mounted on a wearable glass where one camera is embedded with an ESP32 microcontroller. The camera will capture the scenes in front of the user and transfer the images to ESP32 as well as Raspberry Pi. A TensorFlow Lite Micro person detection model is employed in the microcontroller and a Lite object detection model is employed in the Raspberry Pi. These two models are efficient in terms of both memory and processing time. The person detection model will look for whether any person is present in front of the user or not. If any person is detected, it will notify the user via audio feedback. The object detection model can recognize 80 different types of objects from the images and speak out the names of detected objects. Furthermore, the system will provide the audio narration of visual scenes (image captioning to speech) to the user. The image narration model is employed in the server and requires internet connectivity.

AI-XR metaverses leverage artificial intelligence (AI) and extended reality (XR) technology to create immersive environments that replicate and enhance the real world. The transformative impact of the metaverse as an advanced AI-XR platform is widely recognized in the realms of learning, economics, and social activities. However, integrating the meta-verse into our daily lives raises significant ethical and legal challenges. This paper provides a comprehensive examination of associated risks, including privacy and copyright violation, behavior modification, and surveillance. It explores existing methodologies and promising solutions to address these issues while identifying avenues for further research. In conclusion, the paper summarizes key findings and observations, underscoring the importance of a synergistic approach that combines technical and regulatory measures to safeguard the ethical and legal rights of users and co-creators, promoting the development of prosocial AI-XR metaverses.

Protecting the privacy of personal information, including emotions, is essential, and organizations must comply with relevant regulations to ensure privacy. Unfortunately, some organizations do not respect these regulations, or they lack transparency, leaving human privacy at risk. These privacy violations often occur when unauthorized organizations misuse machine learning (ML) technology, such as facial expression recognition (FER) systems. Therefore, researchers and practitioners must take action and use ML technology for social good to protect human privacy. One emerging research area that can help address privacy violations is the use of adversarial ML for social good. Evasion attacks, which are used to fool ML systems, can be repurposed to prevent misused ML technology, such as ML-based FER, from recognizing true emotions. By leveraging adversarial ML for social good, we can prevent organizations from violating human privacy by misusing ML technology, particularly FER systems, and protect individuals' personal and emotional privacy. In this work, we propose an approach called Chaining of Adversarial ML Attacks (CAA) to create a robust attack that fools misused technology and prevents it from detecting true emotions. To validate our proposed approach, we conduct extensive experiments using various evaluation metrics and baselines. Our results show that CAA significantly contributes to emotional privacy preservation, with the fool rate of emotions increasing proportionally to the chaining length. In our experiments, the fool rate increases by 48% in each subsequent chaining stage of the chaining targeted attacks (CTA) while keeping the perturbations imperceptible ($\epsilon = 0.0001$).

With the advent of modern smart grid networks, advanced metering infrastructure provides real-time information from smart meters (SM) and sensors to energy companies and consumers. The smart grid is indeed a paradigm that is enabled by the Internet of Things (IoT) and in which the SM acts as an IoT device that collects and transmits data over the Internet to enable intelligent applications. However, IoT data communicated over the smart grid could however be maliciously altered, resulting in energy theft due to unbilled energy consumption. Machine learning (ML) techniques for energy theft detection (ETD) based on IoT data are promising but are nonetheless constrained by the poor quality of data and particularly its imbalanced nature (which emerges from the dominant representation of honest users and poor representation of the rare theft cases). Leading ML-based ETD methods employ synthetic data generation to balance the training the dataset. However, these are trained to maximise average correct detection instead of ETD. In this work, we formulate an energy-aware evaluation framework that guides the model training to maximise ETD and minimise the revenue loss due to mis-classification. We propose a convolution neural network with positive bias (CNN-B) and another with focal loss CNN (CNN-FL) to mitigate the data imbalance impact. These outperform the state of the art and the CNN-B achieves the highest ETD and the minimum revenue loss with a loss reduction of 30.4% compared to the highest loss incurred by these methods.

Deep Learning (DL) algorithms have shown wonders in many Natural Language Processing (NLP) tasks such as language-to-language translation, spam filtering, fake-news detection, and comprehension understanding. However, research has shown that the adversarial vulnerabilities of deep learning networks manifest themselves when DL is used for NLP tasks. Most mitigation techniques proposed to date are supervised—relying on adversarial retraining to improve the robustness—which is impractical. This work introduces a novel, unsupervised detection methodology for detecting adversarial inputs to NLP classifiers. In summary, we note that minimally perturbing an input to change a model’s output—a major strength of adversarial attacks—is a weakness that leaves unique statistical marks reflected in the cumulative contribution scores of the input. Particularly, we show that the cumulative contribution score, called CF-score, of adversarial inputs is generally greater than that of the clean inputs. We thus propose Con-Detect—a Contribution based Detection method—for detecting adversarial attacks against NLP classifiers. Con-Detect can be deployed with any classifier without having to retrain it. We experiment with multiple attackers—Text-bugger, Text-fooler, PWWS—on several architectures—MLP, CNN, LSTM, Hybrid CNN-RNN, BERT—trained for different classification tasks—IMDB sentiment classification, fake-news classification, AG news topic classification—under different threat models—Con-Detect-blind attacks, Con-Detect-aware attacks, and Con-Detect-adaptive attacks—and show that Con-Detect can reduce the attack success rate (ASR) of different attacks from 100% to as low as 0% for the best cases and ≈70% for the worst case. Even in the worst case, we note a 100% increase in the required number of queries and a 50% increase in the number of words perturbed, suggesting that Con-Detect is hard to evade.

The use of artificial intelligence (AI) at the edge is transforming every aspect of the lives of human beings from scheduling daily activities to personalized shopping recommendations. Since the success of AI is to be measured ultimately in terms of how it benefits human beings, and that the data driving the deep learning-based edge AI algorithms are intricately and intimately tied to humans, it is important to look at these AI technologies through a human-centric lens. However, despite the significant impact of AI design on human interests, the security and trustworthiness of edge AI applications are not foolproof and ethicalneither foolproof nor ethical; Moreover, social norms are often ignored duringin the design, implementation, and deployment of edge AI systems. In this paper, we make the following two contributions: Firstly, we analyze the application of edge AI through a human-centric perspective. More specifically, we present a pipeline to develop human-centric embedded machine learning (HC-EML) applications leveraging a generic human-centric AI (HCAI) framework. Alongside, we also analyzediscuss the privacy, trustworthiness, robustness, and security aspects of HC-EML applications with an insider look at their challenges and possible solutions along the way. Secondly, to illustrate the gravity of these issues, we present a case study on the task of human facial emotion recognition (FER) based on AffectNet dataset, where we analyze the effects of widely used input quantization on the security, robustness, fairness, and trustworthiness of an EML model. We find that input quantization partially degrades the efficacy of adversarial and backdoor attacks at the cost of a slight decrease in accuracy over clean inputs. By analyzing the explanations generated by SHAP, we identify that the decision of a FER model is largely influenced by features such as eyes, alar crease, lips, and jaws. Additionally, we note that input quantization is notably biased against the dark skin faces, and hypothesize that low-contrast features of dark skin faces may be responsible for the observed trends. We conclude with precautionary remarks and guidelines for future researchers.

In recent years, advancements in machine learning (ML) techniques, in particular, deep learning (DL) methods have gained a lot of momentum in solving inverse imaging problems, often surpassing the performance provided by hand-crafted approaches. Traditionally, analytical methods have been used to solve inverse imaging problems such as image restoration, inpainting, and superresolution. Unlike analytical methods for which the problem is explicitly defined and the domain knowledge is carefully engineered into the solution, DL models do not benefit from such prior knowledge and instead make use of large datasets to predict an unknown solution to the inverse problem. Recently, a new paradigm of training deep models using a single image, named untrained neural network prior (UNNP) has been proposed to solve a variety of inverse tasks, e.g., restoration and inpainting. Since then, many researchers have proposed various applications and variants of UNNP. In this paper, we present a comprehensive review of such studies and various UNNP applications for different tasks and highlight various open research problems which require further research.

Traditionally, speech emotion recognition (SER) research has relied on manually handcrafted acoustic features using feature engineering. However, the design of handcrafted features for complex SER tasks requires significant manual effort, which impedes generalisability and slows the pace of innovation. This has motivated the adoption of representation learning techniques that can automatically learn an intermediate representation of the input signal without any manual feature engineering. Representation learning has led to improved SER performance and enabled rapid innovation. Its effectiveness has further increased with advances in deep learning (DL), which has facilitateddeep representation learningwhere hierarchical representations are automatically learned in a data-driven manner. This article presents the first comprehensive survey on the important topic of deep representation learning for SER. We highlight various techniques, related challenges and identify important future areas of research. Our survey bridges the gap in the literature since existing surveys either focus on SER with hand-engineered features or representation learning in the general setting without focusing on SER.

While the technique of Deep Neural Networks (DNNs) has been instrumental in achieving state-of-the-art results for various Natural Language Processing (NLP) tasks, recent works have shown that the decisions made by DNNs cannot always be trusted. Recently Explainable Artificial Intelligence (XAI) methods have been proposed as a method for increasing DNN’s reliability and trustworthiness. These XAI methods are however open to attack and can be manipulated in both white-box (gradient-based) and black-box (perturbation-based) scenarios. Exploring novel techniques to attack and robustify these XAI methods is crucial to fully understand these vulnerabilities. In this work, we propose Tamp-X—a novel attack which tampers the activations of robust NLP classifiers forcing the state-of-the-art white-box and black-box XAI methods to generate misrepresented explanations. To the best of our knowledge, in current NLP literature, we are the first to attack both the white-box and the black-box XAI methods simultaneously. We quantify the reliability of explanations based on three different metrics—the descriptive accuracy, the cosine similarity, and the Lp norms of the explanation vectors. Through extensive experimentation, we show that the explanations generated for the tampered classifiers are not reliable, and significantly disagree with those generated for the untampered classifiers despite that the output decisions of tampered and untampered classifiers are almost always the same. Additionally, we study the adversarial robustness of the tampered NLP classifiers, and find out that the tampered classifiers which are harder to explain for the XAI methods, are also harder to attack by the adversarial attackers.

The domain of Artificial Intelligence (AI) ethics is not new, with discussions going back at least 40 years. Teaching the principles and requirements of ethical AI to students is considered an essential part of this domain, with an increasing number of technical AI courses taught at several higher-education institutions around the globe including content related to ethics. By using Latent Dirichlet Allocation (LDA), a generative probabilistic topic model, this study uncovers topics in teaching ethics in AI courses and their trends related to where the courses are taught, by whom, and at what level of cognitive complexity and specificity according to Bloom’s taxonomy. In this exploratory study based on unsupervised machine learning, we analyzed a total of 166 courses: 116 from North American universities, 11 from Asia, 36 from Europe, and 10 from other regions. Based on this analysis, we were able to synthesize a model of teaching approaches, which we call BAG (Build, Assess, and Govern), that combines specific cognitive levels, course content topics, and disciplines affiliated with the department(s) in charge of the course. We critically assess the implications of this teaching paradigm and provide suggestions about how to move away from these practices. We challenge teaching practitioners and program coordinators to reflect on their usual procedures so that they may expand their methodology beyond the confines of stereotypical thought and traditional biases regarding what disciplines should teach and how. This article appears in the AI & Society track.

The products built by engineers are enabling better economies, infrastructures, sustainable living, and longer life spans. There is a rising concern about the ethical considerations taken while designing these disruptive products. To ensure ethical products, the pedagogy and curricula of engineering ethics need to be sound. In this study, we try to understand what is being taught when teaching engineering ethics. To do so, we perform a curricular analysis of 123 engineering/ tech ethics courses from around the globe by analyzing their syllabi to shed light on the topics that are covered in curricula and how they relate to current world problems. We also perform a bibliometric analysis and look at 350 research articles on engineering ethics, published in the past 20 years in four top research venues. This enabled us to observe the evolution of the research trends in the engineering ethics research community. We in the end focus on the demographic distribution and global inclusion into the engineering ethics debate. Through this combined bibliometric and curricular analysis, we explore the state of engineering ethics and present insights that will be useful for engineering educators, researchers, and practitioners alike.

Protocol standards, defined by the Internet Engineering Task Force (IETF), are crucial to the successful operation of the Internet. This paper presents a large-scale empirical study of IETF activities, with a focus on understanding collaborative activities, and how these underpin the publication of standards documents (RFCs). Using a unique dataset of 2.4 million emails, 8,711 RFCs and 4,512 authors, we examine the shifts and trends within the standards development process, showing how protocol complexity and time to produce standards has increased. With these observations in mind, we develop statistical models to understand the factors that lead to successful uptake and deployment of protocols, deriving insights to improve the standardisation process.

In this paper, we make use of systems thinking insights to study education during and beyond COVID-19. Systems thinking is a rich discipline that studies nonlinear models of social complex adaptive systems that has many insights and tools that are relevant for modelling and understanding how interactions unfold in educational systems. An important insight of systems thinking is that the root cause of chronic complex problems often lay in the underlying systemic structure. Using insights from systems thinking to study learning/education has many benefits, including: (1) support for rigorous big-picture thinking; (2) anticipating and managing unintended consequences; (3) understanding dysfunctional learning systems using systems archetypes - which are systemic structures that, experts have noticed, typically lead to a performance rut; and finally (4) identification of high-leverage interventions that lead to long-lasting benefits without being neutralized by the system. In the paper, we have modelled COVID-19 pandemic effects on students learning in a novel way by using system thinking tools (Causal Loop Diagrams and Stock and Flow Diagrams), which help us to understand the complex interconnections of students performance, learning and management reforms. We demonstrate that successful student learning during and beyond COVID-19 requires not only a focus on lectures and curriculum reforms but also on motivating students, instilling a growth mindset, and developing strategies to track and minimize online distractions.

Machine learning (ML), and deep learning (DL) in particular, play a vital role in providing smart services to the industry. These techniques, however, suffer from privacy and security concerns since data are collected from clients and then stored and processed at a central location. Federated learning (FL), an architecture in which model parameters are exchanged instead of client data, has been proposed as a solution to these concerns. Nevertheless, FL trains a global model by communicating with clients over communication rounds, which introduces more traffic on the network and increases the convergence time to the target accuracy. In this work, we solve the problem of optimizing accuracy in stateful FL with a budgeted number of candidate clients by selecting the best candidate clients in terms of test accuracy to participate in the training process. Next, we propose an online stateful FL heuristic to find the best candidate clients. Additionally, we propose an IoT client alarm application that utilizes the proposed heuristic in training a stateful FL global model based on IoT device-type classification to alert clients about unauthorized IoT devices in their environment. To test the efficiency of the proposed online heuristic, we conduct several experiments using a real data set and compare the results against state-of-the-art algorithms. Our results indicate that the proposed heuristic outperforms the online random algorithm with up to 27% gain in accuracy. Additionally, the performance of the proposed online heuristic is comparable to the performance of the best offline algorithm.

With machine learning (ML) services now used in a number of mission-critical human-facing domains, ensuring the integrity and trustworthiness of ML models becomes all important. In this work, we consider the paradigm where cloud service providers collect big data from resource-constrained devices for building ML-based prediction models that are then sent back to be run locally on the intermittently connected resource-constrained devices. Our proposed solution comprises an intelligent polynomial-time heuristic that maximizes the level of trust of ML models by selecting and switching between a subset of the ML models from a superset of models in order to maximize the trustworthiness while respecting the given reconfiguration budget/rate and reducing the cloud communication overhead. We evaluate the performance of our proposed heuristic using two case studies. First, we consider Industrial IoT (IIoT) services, and as a proxy for this setting, we use the turbofan engine degradation simulation data set to predict the remaining useful life of an engine. Our results in this setting show that the trust level of the selected models is 0.49%-3.17% less compared to the results obtained using integer linear programming (ILP). Second, we consider smart cities services, and as a proxy of this setting, we use an experimental transportation data set to predict the number of cars. Our results show that the selected model's trust level is 0.7%-2.53% less compared to the results obtained using ILP. We also show that our proposed heuristic achieves an optimal competitive ratio in a polynomial-time approximation scheme for the problem.

The worldwide spread of COVID-19 has prompted extensive online discussions, creating an ‘infodemic’ on social media platforms such as WhatsApp and Twitter. However, the information shared on these platforms is prone to be unreliable and/or misleading. In this paper, we present the first analysis of COVID-19 discourse on public WhatsApp groups from Pakistan. Building on a large scale annotation of thousands of messages containing text and images, we identify the main categories of discussion. We focus on COVID-19 messages and understand the different types of images/text messages being propagated. By exploring user behavior related to COVID messages, we inspect how misinformation is spread. Finally, by quantifying the flow of information across WhatsApp and Twitter, we show how information spreads across platforms and how WhatsApp acts as a source for much of the information shared on Twitter.

The Internet of Things (IoT) revolution and the development of smart communities have resulted in increased demand for bandwidth due to the rise in network traffic. Instead of investing in expensive communications infrastructure, some researchers have proposed leveraging Vehicular Ad-Hoc Networks (VANETs) as the data communications infrastructure. However VANETs are not cheap since they require the deployment of expensive Road Side Units (RSU)s across smart communities. In this research, we propose an infrastructure-less system that opportunistically utilizes vehicles to serve as Local Community Brokers (LCBs) that effectively substitute RSUs for managing communications between smart devices and the cloud in support of smart community applications. We propose an opportunistic algorithm that strives to select vehicles in order to maximize the LCBs' service time. The proposed opportunistic algorithm utilizes an ensemble of online selection algorithms by running all of them together in passive mode and selecting the one that has performed the best in recent history. We evaluate our proposed algorithm using a dataset comprising real taxi traces from the city of Shanghai in China and compare our algorithm against a baseline of 9 Threshold Based Online (TBO) algorithms. A number of experiments are conducted and our results indicate that the proposed algorithm achieves up to 87 % more service time with up to 10% fewer vehicle selections compared to the best-performing existing TBO online algorithm.

As we move into the third decade of the 21st century, the 2020s, the unprecedented rate of technological disruption and the short-lived nature of the specifics of engineering state-of-the-art require us to carefully evaluate what it takes to be an effective engineer and what this entails for engineering education and their lifelong learning. While it is true that certain basics of engineering will not change, there will be an increased premium for some skills (such as lifelong learning, meta-learning, collaboration, creativity, critical thinking, communication skills, and cultural/global literacy). 21st-century skills are, as such, timeless skills: it is paradoxically the volatile nature of the modern world that has forced us from ephemeral vocational fads back to these permanently valuable skills. In this full research-to-practice paper, after reporting on the skills that policy think tanks and thought leaders deem necessary for the 21st century, we provide a synthesis in which we describe the pulls and pushes that learners and educators will face in the turbulent times of 2020 and beyond, and how they can thrive in the uncertain future through holistic well-rounded engineering education.

Most of the research on Federated Learning (FL) has focused on analyzing global optimization, privacy, and communication, with limited attention focusing on analyzing the critical matter of performing efficient local training and inference at the edge devices. One of the main challenges for successful and efficient training and inference on edge devices is the careful selection of parameters to build local Machine Learning (ML) models. To this aim, we propose a Particle Swarm optimization (PSO)-based technique to optimize the hyperparameter settings for the local ML models in an FL environment. We evaluate the performance of our proposed technique using two case studies. First, we consider smart city services, and use an experimental transportation dataset for traffic prediction as a proxy for this setting. Second, we consider Industrial IoT(IIoT) services, and use the real-time telemetry dataset to predict the probability that a machine will fail shortly due to component failures. Our experiments indicate that PSO provides an efficient approach for tuning the hyperparameters of deep Long short-term memory (LSTM) models when compared to the grid search method. Our experiments illustrate that the number of client-server communication rounds to explore the landscape of configurations to find the near-optimal parameters are greatly reduced (roughly by two orders of magnitude needing only 2%-4% of the rounds compared to state of the art non-PSO-based approaches). We also demonstrate that utilizing the proposed PSO-based technique to find the near-optimal configurations for FL and centralized learning models does not adversely affect the accuracy of the models.

The performance of densely-deployed low-power wide-area networks (LPWANs) can significantly deteriorate due to packets collisions, and one of the main reasons for that is the rule-based PHY layer transmission parameters assignment algorithms. LoRaWAN is a leading LPWAN technology where LoRa serves as the physical layer. Here, we propose and evaluate a deep reinforcement learning (DRL)-based PHY layer transmission parameter assignment algorithm for LoRaWAN. Our algorithm ensures fewer collisions and better network performance compared to the existing state-of-the-art PHY layer transmission parameter assignment algorithms for LoRaWAN. Our algorithm outperforms the state of the art learning-based technique achieving up to 500% improvement of PDR in some cases.

5G is expected to serve diverse applications and users due to the popularity of Internet of Things (IoT), big data and industrial applications. Many of these IoT and industrial applications have inherent loss tolerance that can be used to enable energy efficient uplink communication. The uplink energy efficient system will increase the battery life of devices enabling new use cases in industrial IoT. In this paper, we map the effects of application loss tolerance to the rate requirements of the user. We then mathematically model an energy minimization problem for the uplink user association and resource allocation in heterogeneous networks. We aim to provide acceptable quality of service (QoS) with improved energy efficiency by exploiting the loss tolerance and bandwidth expansion simultaneously. A distributed uplink joint user association and resource allocation strategy for uplink energy per bit minimization is presented. We conduct extensive simulation based study for a heterogeneous network to evaluate the performance of our proposed schemes. Average energy per bit consumption in the proposed scheme is -74 dB compared to -53 dB in state-of-the-art channel individual offset (CIO) scheme.

Cross-lingual speech emotion recognition (SER)is a crucial task for many real-world applications. The performance of SER systems is often degraded by the differences in the distributions of training and test data. These differences become more apparent when training and test data belong to different languages, which cause a significant performance gap between the validation and test scores. It is imperative to build more robust models that can fit in practical applications of SER systems. Therefore, in this paper, we propose a Generative Adversarial Network (GAN)-based model for multilingual SER. Our choice of using GAN is motivated by their great success in learning the underlying data distribution. The proposed model is designed in such a way that the language invariant representations can be learned without requiring target-language data labels. We evaluate our proposed model on four different language emotional datasets, including an Urdu-language dataset to also incorporate alternative languages for which labelled data is difficult to find and which have not been studied much by the mainstream community. Our results show that our proposed model can significantly improve the baseline cross-lingual SER performance for all the considered datasets including the non-mainstream Urdu language data without requiring any labels.

Deep neural networks (DNN)-based machine learning (ML) algorithms have recently emerged as the leading ML paradigm particularly for the task of classification due to their superior capability of learning efficiently from large dataseis. The discovery of a number of well-known attacks such as dataset poisoning, adversarial examples, and network manipulation (through the addition of malicious nodes) has, however, put the spotlight squarely on the lack of security in DNN-based ML systems. In particular, malicious actors can use these well-known attacks to cause random/targeted misclassification, or cause a change in the prediction confidence, by only slightly but systematically manipulating the environmental parameters, inference data, or the data acquisition block. Most of the prior adversarial attacks have, however, not accounted for the pre-processing noise filters commonly integrated with the ML-inference module. Our contribution in this work is to show that this is a major omission since these noise filters can render ineffective the majority of the existing attacks, which rely essentially on introducing adversarial noise. Apart from this, we also extend the state of the art by proposing a novel pre-processing noise Filter-aware Adversarial ML attack called FAdeML. To demonstrate the effectiveness of the proposed methodology, we generate an adversarial attack image by exploiting the "VGGNet" DNN trained for the "German Traffic Sign Recognition Benchmarks (GTSRB)" dataset, which despite having no visual noise, can cause a classifier to misclassify even in the presence of preprocessing noise filters.

This paper describes the design process by which we designed an Android application equipped with audio, textual menus and visuals components for use by farmers of diverse literacy levels looking for vital weather information after the conclusion of research-work that productivity lags due to information inadequacies. The intervention provides more timely access to accurate information to low-literate farmers and thereby help in making the agricultural ecosystem more robust. We discuss the various design and implementation features of our system and presents our findings from the field on the usability of our application. We have also openly released our source code so that other users and developers can also benefit from our work.

Intrusion detection systems (IDSs) are an essential cog of the network security suite that can defend the network from malicious intrusions and anomalous traffic. Many machine learning (ML)-based IDSs have been proposed in the literature for the detection of malicious network traffic. However, recent works have shown that ML models are vulnerable to adversarial perturbations through which an adversary can cause IDSs to malfunction by introducing a small impracticable perturbation in the network traffic. In this paper, we propose an adversarial ML attack using generative adversarial networks (GANs) that can successfully evade an ML-based IDS. We also show that GANs can be used to inoculate the IDS and make it more robust to adversarial perturbations.

Deep machine learning techniques have shown promising results in network traffic classification, however, the robustness of these techniques under adversarial threats is still in question. Deep machine learning models are found vulnerable to small carefully crafted adversarial perturbations posing a major question on the performance of deep machine learning techniques. In this paper, we propose a black-box adversarial attack on network traffic classification. The proposed attack successfully evades deep machine learning-based classifiers which highlights the potential security threat of using deep machine learning techniques to realize autonomous networks.

Interest in smart cities is rapidly rising due to the global rise in urbanization and the wide-scale instrumentation of modern cities. Due to the considerable infrastructural cost of setting up smart cities and smart communities, researchers are exploring the use of existing vehicles on the roads as "message ferries" for transporting data for smart community applications to avoid the cost of installing new communication infrastructure. In this paper, we propose an opportunistic data ferry selection algorithm that strives to select vehicles that can minimize the overall delay for data delivery from a source to a given destination. Our proposed opportunistic algorithm utilizes an ensemble of online hiring algorithms, which are run together in passive mode, to select the online hiring algorithm that has performed the best in recent history. The proposed ensemble- based algorithm is evaluated empirically using real-world traces from taxies plying routes in Shanghai, China, and its performance is compared against a baseline of four state-of-the-art online hiring algorithms. A number of experiments are conducted and our results indicate that the proposed algorithm can reduce the overall delay compared to the baseline by an impressive 13% to 258%.

The popularity of cellular internet of things (IoT) is increasing day by day and billions of IoT devices will be connected to the internet. Many of these devices have limited battery life with constraints on transmit power. High user power consumption in cellular networks restricts the deployment of many IoT devices in 5G. To enable the inclusion of these devices, 5G should be supplemented with strategies and schemes to reduce user power consumption. Therefore, we present a novel joint uplink user association and resource allocation scheme for minimizing user transmit power while meeting the quality of service. We analyze our scheme for two-tier heterogeneous network (HetNet) and show an average transmit power of -2.8 dBm and 8.2 dBm for our algorithms compared to 20 dBm in state-of-the-art Max reference signal received power (RSRP) and channel individual offset (CIO) based association schemes.

Learning the latent representation of data in unsupervised fashion is a very interesting process that provides relevant features for enhancing the performance of a classifier. For speech emotion recognition tasks, generating effective features is crucial. Currently, handcrafted features are mostly used for speech emotion recognition, however, features learned automatically using deep learning have shown strong success in many problems, especially in image processing. In particular, deep generative models such as Variational Autoencoders (VAEs) have gained enormous success for generating features for natural images. Inspired by this, we propose VAEs for deriving the latent representation of speech signals and use this representation to classify emotions. To the best of our knowledge, we are the first to propose VAEs for speech emotion classification. Evaluations on the IEMOCAP dataset demonstrate that features learned by VAEs can produce state-of-the-art results for speech emotion classification.

The majority of existing speech emotion recognition research focuses on automatic emotion detection using training and testing data from same corpus collected under the same conditions.The performance of such systems has been shown to drop significantly in cross-corpus and cross-language scenarios.To address the problem, this paper exploits a transfer learning technique to improve the performance of speech emotion recognition systems that is novel in cross-language and cross-corpus scenarios.Evaluations on five different corpora in three different languages show that Deep Belief Networks (DBNs) offer better accuracy than previous approaches on cross-corpus emotion recognition, relative to a Sparse Autoencoder and SVM baseline system.Results also suggest that using a large number of languages for training and using a small fraction of the target data in training can significantly boost accuracy compared with baseline also for the corpus with limited training examples.

Modern information technology services largely depend on cloud infrastructures to provide their services. These cloud infrastructures are built on top of Datacenter Networks (DCNs) constructed with high-speed links, fast switching gear, and redundancy to offer better flexibility and resiliency. In this environment, network traffic includes long-lived (elephant) and short-lived (mice) flows with partitioned/aggregated traffic patterns. Although SDN-based approaches can efficiently allocate networking resources for such flows, the overhead due to network reconfiguration can be significant. With limited capacity of Ternary Content-Addressable Memory (TCAM) deployed in an OpenFlow enabled switch, it is crucial to determine which forwarding rules should remain in the flow table and which rules should be processed by the SDN controller in case of a table-miss on the SDN switch. This is needed in order to obtain the flow entries that satisfy the goal of reducing the long-term control plane overhead introduced between the controller and the switches. To achieve this goal, we propose a machine learning technique that utilizes two variations of Reinforcement Learning (RL) algorithms—the first of which is a traditional RL-based algorithm, while the other is deep reinforcement learning-based. Emulation results using the RL algorithm show around 60% improvement in reducing the long-term control plane overhead and around 14% improvement in the table-hit ratio compared to the Multiple Bloom Filters (MBF) method, given a fixed size flow table of 4KB.

Pakistan's agricultural sector has been making gigantic contributions towards the nation's economy, with agriculture accounting for 22% of the gross domestic product (GDP) while engaging approximately half of the country's labor force. A significant developmental challenge in this sector is inadequacy and inaccessibility of information regarding weather forecast. In this paper, we propose an Android-based solution for farmers that can facilitate the timely, localized, and customized dissemination of granular weather forecast that shields the whole agricultural ecosystem and supply chain from weather variability by appropriate decision-making. We describe our Android mobile application that sends a customized weather forecast that is configured according to the user preferences. Information is disseminated by the cloud server through encrypted SMS to the subscribing farmers containing weather information. This information is encoded through visuals and icons in a simple to understand user-interface that is accompanied by Urdu language text in a design tailored for low-literate farmers of Pakistan. The testing, feedback, and evaluation include design understanding, the effectiveness of icons and images, usability, adaptation to touch screen is in progress which will help us to reiterate the mobile app user interface (UI) to improve the preliminary design.

Heterogeneous networks are a key solution to serving the exponential surge in data volume and higher quality expectations. Nonetheless, such networks require the ubiquitous presence of fiber-to-the-cell to address the performance demands of 5G and fast-spreading small cells. To this end, innovative ways of optimizing the usage of realistic backhaul links are being investigated. In this work, we propose a fuzzy Q-learning-based user-centric backhaul-aware user cell association scheme. The proposed scheme aims at optimizing the user-cell association process in a context-aware and backhaul-aware manner. Complementing the scheme with fuzzy-logic requires 33.3% additional storage memory. On the other hand, it increases the computational efficiency by 60% and improves the users' performance by 12%.

In this paper, we present a novel idea of multipath multicast, which is imperative to bandwidth intensive applications, in the context of multimedia streaming. In addition to congestion control, multipath TCP (MPTCP) has been proposed to establish multiple paths in a network to improve network reliability. Application-layer multicast (ALM) has been proposed to leverage end systems instead of dedicated routers to multicast that is important for an easy large-scale deployment as compared to IP-based multicast. This paper presents our novel idea of multipath multicast in the form of a simple experimental framework called MP-ALM in which we combine the multiplicity feature of MPTCP with the application-layer multicast (ALM). We extensively simulate MP-ALM using ns-3 and use iPerf to generate streaming multicast-MPTCP traffic. Simulation results show that MP-ALM can be beneficial for a better user experience and reduced overall network congestion in the perspective of multicast multimedia streaming.

In this paper, we present a game theoretic framework for spectrum allocation in distributed cognitive radio networks containing both unicast and broadcast traffic. Our proposed scheme aims to minimize broadcast latency for broadcast traffic and minimize interference and access contention for both types of traffic. We develop a utility function that ensures that both objectives are met yielding a higher network throughput. Our proposed spectrum allocation game is also formulated as a potential game and is guaranteed to converge to a Nash equilibrium if the sequential best response dynamics is followed. A proof of concept of the proposed algorithm has been implemented on the Orbit radio testbed and the results verify the convergence of the potential game. Our simulation and experimental results also reveal that the choice of utility function improves the average network throughput for a mixed traffic profile.

This paper addresses the problem of "efficientrdquo broadcast in a multiradio, multichannel, multirate wireless mesh network (MR2-MC WMN). In such an MR2-MC WMN, nodes are equipped with multiple radio interfaces, tuned to orthogonal channels, that can dynamically adjust their transmission rate by choosing a modulation scheme appropriate for the channel conditions. We choose "broadcast latency,rdquo defined as the maximum delay between a packet's network-wide broadcast at the source and its eventual reception at all network nodes, as the ldquoefficiencyrdquo metric of broadcast performance. We study in this paper how the availability of multirate transmission capability and multiple radio interfaces tuned to orthogonal channels in MR2-MC WMN nodes can be exploited, in addition to the medium's ldquowireless broadcast advantagerdquo (WBA), to improve the ldquobroadcast latencyrdquo performance. In this paper, we present four heuristic solutions to our considered problem. We present detailed simulation results for these algorithms for an idealized scheduler, as well as for a practical 802.11-based scheduler. We also study the effect of channel assignment on broadcast performance and show that channel assignment can affect the broadcast performance substantially. More importantly, we show that a channel assignment that performs well for unicast does not necessarily perform well for broadcast/multicast.

We address the problem of minimizing the worst-case broadcast delay in ldquomulti-radio multi-channel multi-rate wireless mesh networksrdquo (MR2-MC WMN) in a distributed and localized fashion. Efficient broadcasting in such networks is especially challenging due to the desirability of exploiting the ldquowireless broadcast advantagerdquo (WBA), the interface-diversity, the channel-diversity and the rate-diversity offered by these networks. We propose a framework that calculates a set of forwarding nodes and transmission rate at these forwarding nodes irrespective of the broadcast source. Thereafter, a forwarding tree is constructed taking into consideration the source of broadcast. Our broadcasting algorithms are distributed and utilize locally available information. We present a detailed performance evaluation of our distributed and localized algorithm and demonstrate that our algorithm can greatly improve broadcast performance by exploiting the rate, interface and channel diversity of MR2-MC WMNs and match the performance of centralized algorithms proposed in literature while utilizing only limited two-hop neighborhood information.

We address the problem of minimizing the worst-case broadcast delay in multi-rate wireless mesh networks (WMN) in a distributed and localized fashion. Efficient broadcasting in such networks is especially challenging due to the multi-rate transmission capability and the interference between wireless transmissions of WMN nodes. We propose connecting dominating set (CDS) based broadcast routing approach which calculates the set of forwarding nodes and the transmission rate at each forwarding node independent of the broadcast source. Thereafter, a forwarding tree is constructed taking into consideration the source of the broadcast. In this paper, we propose three distributed and localized rate-aware broadcast algorithms. We compare the performance of our distributed and localized algorithms with previously proposed centralized algorithms and observe that the performance gap is not large. We show that our algorithms greatly improve performance of rate-unaware broadcasting algorithms by incorporating rate-awareness into the broadcast tree construction algorithm process.

A multi-rate capable IEEE 802.11a/b/g node can utilize different link-layer transmission rates. Interestingly, multi-rate capability is defined by IEEE802.11 standards only for unicast transmissions. In this paper, we consider a novel type of multi-radio multi-channel wireless mesh network (WMN) where a radio can multicast at different link-layer transmission rates to its neighbors. Such link-layer multi-rate multicast capability will enable low-latency network-layer broadcast/multicast for multimedia. In our previous work, we assumed a "fully multi-rate multicast" (EMM) framework in which nodes can adjust link-layer multicast transmission rate for each link-layer frame. We propose a new framework called "single best-rate multicast" (SBM) that exploits the link-layer rate diversity by enabling each WMN to decide, depending on its topological properties, a single transmission rate for all its link-layer data multicasts. Although, EMM improves performance significantly, employing SBM is attractive since it can eliminate some undesirable features of practical multi-rate media access control (MAC) protocols. We propose methods to determine the "best" link-layer transmission rate for the SBM framework. We also propose two heuristic broadcast solutions, using SBM framework that can realize low-latency broadcast by exploiting inherent rate and interface diversity in multi-radio multi-channel WMN. Simulation results indicate that SBM broadcast heuristics give comparable performance to EMM broadcast heuristics, especially in dense networks

We address the problem of minimizing the worst-case broadcast delay in multi-radio multi-channel multi-rate (MR2-MC) wireless mesh networks (WMN). The problem of 'efficient' broadcast in such networks is especially challenging due to the numerous interrelated decisions that have to be made. The multi-rate transmission capability of WMN nodes, interference between wireless transmissions, and the hardness of optimal channel assignment adds complexity to our considered problem. We present four heuristic algorithms to solve the minimum latency broadcast problem for such settings and show that the 'best' performing algorithms usually adapt themselves to the available radio interfaces and channels. We also study the effect of channel assignment on broadcast performance and show that channel assignment can affect the broadcast performance substantially. More importantly, we show that a channel assignment that performs well for unicast does not necessarily perform well for broadcast/multicast. To the best of our knowledge, this work constitutes the first contribution in the area of broadcast routing for MR2-MC WMN

In a multirate wireless network, a node can dynamically adjust its link transmission rate by switching between different modulation schemes. In the current IEEE802.11a/b/g standards, this rate adjustment is defined for unicast traffic only. In this paper, we consider a wireless mesh network (WMN), where a node can dynamically adjust its link-layer multicast rates to its neighbors, and address the problem of realizing low-latency network-wide broadcast in such a mesh. We first show that the multirate broadcast problem is significantly different from the single-rate case. We will then present an algorithm for achieving low-latency broadcast in a multirate mesh which exploits both the wireless multicast advantage and the multirate nature of the network. Simulations based on current IEEE802.11 parameters show that multirate multicast can reduce broadcast latency by 3-5 times compared with using the lowest rate alone. In addition, we show the significance of the product of transmission rate and transmission coverage area in designing multirate WMNs for broadcast