Large Language Models (LLMs) are increasingly being adopted in a wide variety of domains, including sensitive domains such as healthcare and finance. However, persistent challenges such as unreliable data sources, privacy breaches, and hallucinated output continue to hinder their usage. We have experimented with several strategies to address these challenges. First, we developed BlockQwen, a blockchain-augmented framework that integrates decentralized trust validation, role-specific access control, and verifiable audit trails into the Qwen 2.5 LLM workflow. Second, we developed PrivAware, a multilayered privacy-enforcement framework, using a fine-tuned Flan-T5 model with self-attention masking, to safeguard data while maintaining high utility. Both systems resulted in significant improvement in mitigating privacy leaks and hallucinations. In this paper, we discuss the challenges that we faced in developing these systems and how these challenges were incrementally overcome. We briefly describe each system, and more importantly, we discuss the lessons learned throughout the development and testing process. It also includes a justification for each of the strategies employed and the benefits gained by such deployments. Finally, we provide guidelines for future development of trustworthy systems using LLMs, with special focus on preventing privacy leaks, minimizing hallucinations, improved authentication and authorization, and immutable audit trails.

AI agent-based systems are becoming increasingly integral to modern software architectures, enabling autonomous decision-making, dynamic task execution, and multimodal interactions through large language models (LLMs). However, these systems introduce novel and evolving security challenges, including prompt injection attacks, context poisoning, model manipulation, and opaque agent-to-agent communication, that are not effectively captured by traditional threat modeling frameworks. In this paper, we introduce ASTRIDE, an automated threat modeling platform purpose-built for AI agent-based systems. ASTRIDE extends the classical STRIDE framework by introducing a new threat category, A for AI Agent-Specific Attacks, which encompasses emerging vulnerabilities such as prompt injection, unsafe tool invocation, and reasoning subversion, unique to agent-based applications. To automate threat modeling, ASTRIDE combines a consortium of fine-tuned vision-language models (VLMs) with the OpenAI-gpt-oss reasoning LLM to perform end-to-end analysis directly from visual agent architecture diagrams, such as data flow diagrams(DFDs). LLM agents orchestrate the end-to-end threat modeling automation process by coordinating interactions between the VLM consortium and the reasoning LLM. Our evaluations demonstrate that ASTRIDE provides accurate, scalable, and explainable threat modeling for next-generation intelligent systems. To the best of our knowledge, ASTRIDE is the first framework to both extend STRIDE with AI-specific threats and integrate fine-tuned VLMs with a reasoning LLM to fully automate diagram-driven threat modeling in AI agent-based applications.

E-commerce platforms generate extensive multi-modal data, including product descriptions, images, and customer reviews, which significantly influence consumer purchasing. However, discrepancies between seller claims and buyer experiences often lead to mistrust, dissatisfaction, and financial loss. Traditional e-commerce analytics approaches, including text-based sentiment analysis, standalone image classification, and rudimentary summarization, often fail to capture the complex interplay between modalities and therefore overlook nuanced discrepancies across textual and visual inputs. To address these limitations, we introduce ReViewQwen, a novel multimodal discrepancy detection and summarization framework leveraging the advanced Qwen2-VL Vision-Language Model. ReViewQwen integrates textual and visual inputs (product images from both buyer and seller) into a unified embedding space to systematically detect and contextualize discrepancies. Our comprehensive evaluation demonstrates that ReViewQwen outperforms state-of-the-art models such as LLaMA 3.2, Phi-3.5, and PaLiGemma 2, achieving superior precision, recall, and F1-score. Notably, the proposed system achieves accuracy improvement over the best-performing baseline from 51.15% to 88.00%. Additionally, our method promotes fairness in e-commerce review analytics by substantially reducing model biases and hallucinated content, thereby ensuring more trustworthy and balanced explanations. To access source code, data, and Prompts used https://github.com/domsoos/reviewqwen

Modern 5G networks offer a network-sliced infrastructure where each network slice contains a dedicated 5G core software service layer. The 5G core software services in each slice shares common core network resources to meet specific customer needs. A primary challenge in 5G network slicing involves resource sharing and efficient network slice orchestration. Container-based methodologies, including tools like Docker and Kubernetes, have become popular for orchestrating 5G network slice services and managing configurations in microservices-based cloud-native service deployment. However, despite their utility, these tools present significant challenges. Their complexity often necessitates dedicated DevOps teams for effective management, while configuration management can prove arduous, and end-to-end supply chain oversight is lacking. To address these challenges, this paper introduces “Llama-Recipe,” a cloud-native 5G-core service deployment and orchestration platform integrating Generative AI, SBOM, PBOM and NFT. 5G-core service configurations across different network slices are represented as “HOCON (Human-Optimized Config Object Notation)” config objects adhering to the GitOps paradigm. Leveraging custom-trained Meta's Llama2 LLM, Llama-Recipe generates the Kubernetes manifests for network-sliced 5G-core services based on the defined HOCON configurations. The generated Kubernetes manifests of the 5G-core services are deployed in designated Kubernetes clusters utilizing GitOps tools (e.g., ArgoCD), ensuring seamless and automated deployment processes. Additionally, Llama-Recipe introduced a novel mechanism to handle end-to-end supply chain verification of 5G-core software services using Software-Bill of Materials (SBOM) and Pipeline-Bill of Materials (PBOM). SBOMs track all the dependencies and PBOMs facilitate the comprehensive tracking of end-to-end supply chain data for 5G-core software services, enhancing transparency and security. These PBOMs are also generated using the fine-tuned Meta's Llama-2 LLM and are encoded as NFT tokens with a novel NFT token schema. This schema enables easy verification and validation of supply-chain data during deployments, thus helping to prevent various supply-chain attacks. To fine-tune the Meta's Llama2 LLM, we've undertaken a meticulous training process, collaborating with Qlora to transform a 4-bit quantized pre-trained language model into Low-Rank Adapters(LoRA). The effectiveness of the Llama-Recipe is demonstrated through a real-world test-bed deployment in a sliced network scenario, utilizing multiple 5G cores (i.e., Open5GS) across Ericsson's new Radio Access Network (RAN).

Current wind energy data platforms face significant challenges in securing and managing extensive data from both offshore and onshore wind farms. These challenges include vulnerabilities to cyber-attacks, data tampering, breaches, complex data-sharing issues due to privacy concerns and regulatory compliance, and a lack of scalability and flexibility in analytical tools for real-time data processing. This paper proposes a novel multilayered data security architecture, termed "VindSec-Llama," to address these challenges. It integrates Generative AI, blockchain, federated learning, and Pipeline Bill of Materials (PBOM) to enhance data analytics, model development, and security across several layers, including Infrastructure, Data Lake, Federated Learning, MLOps, Data Provenance, and LLM. Each layer is designed to meet specific functional requirements, such as handling large datasets, facilitating secure federated learning, automating risk management, and ensuring data provenance and traceability. The platform, deployable in server environments (cloud or on-premises), complies with the Risk Management Framework (RMF) guidelines and security standards. It features a blockchain-enabled, coordinator-less federated learning system to enhance data privacy and security by enabling the development of privacy-preserving machine learning models with data from different wind farms. Automation plays a pivotal role throughout VindSec-Llama, with Meta’s custom-trained Llama-3 LLM used for generating remediation scripts in the Infrastructure Layer and for producing PPBOM in the MLOps Layer. The Llama-3 LLM has been quantized and fine-tuned using Qlora to ensure optimal performance on consumer-grade hardware. The MLOps pipeline setup, a critical functionality of VindSec-Llama, ensures seamless integration and deployment of machine learning models, embodying best practices in continuous integration and delivery. This setup is geared towards maximizing security, compliance, and operational efficiency. A prototype of the platform has been implemented within a wind-energy testbed with the collaboration of Department of Energy US, illustrating its practical applications and benefits.

Large Language Models (LLMs) are widely recognized for their applications in natural language processing tasks, but their potential extends far beyond traditional use cases. This paper introduces "Bassa-Llama," a novel platform that harnesses LLMs for predictive tasks in the realm of network security. Specifically, we propose a platform for real-time network attack detection in Wind Power Plants, leveraging a fine-tuned version of Meta’s Llama-3 LLM alongside blockchain and NFT-based data storage. Using a network PCAP dataset containing both malicious and benign packets, we fine-tune the Llama-3 LLM, with Quantized Low-Rank Adapter (QLoRA), to detect anomalies in network traffic. This approach ensures optimal performance on consumer-grade hardware while significantly enhancing the model’s ability to accurately analyze PCAP data and identify attack patterns. The end-to-end orchestration of the real-time network attack detection flow for Wind Power Plants is fully automated through blockchain smart contracts, and NFTs for storing identified attack data from the PCAP. To the best of our knowledge, this research represents the first effort to utilize a fine-tuned LLM for real-time network attack detection tasks. The results highlight the transformative potential of combining fine-tuned LLMs with blockchain and NFTs to build robust and secure network defense systems for Wind Power Plants. A prototype of the proposed platform was developed in collaboration with the U.S. Department of Energy, utilizing a simulated Wind Power Plant as a testbed.

Cyberattacks on IoT devices are accelerating at an unprecedented rate, largely driven by IoT malware activities. The IoT malware attacks are typically composed of three stages: intrusion, infection, and execution. It is essential to instantaneously detect malware at the early stage of intrusion on IoT devices before massive attacks. To build an efficient and scalable instruction detection system, a multi-client single-server privacy-preserved neural network model is proposed. Further, it is important to train the neural network model on the fly to cope with the fast-evolving malware variations. For online training and detection, we propose Lynx-Net, a distributed deep learning online training model for inferring malicious activities by analyzing power side-channel signals. To protect user private data and model parameters, and reduce prediction latency, we implement a novel privacy-preserved protocol via secret sharing and packed hybrid homomorphic encryption. Through theoretical analysis and empirical experiments, we demonstrate that Lynx-Net can detect infection activities of different IoT malware with high accuracy. Our extensive experiments demonstrate not only stable training but also a 1.13 to 2.67 times speedup compared to the state-of-the-art training model and an 8 to 500 times improvement in Lynx-Net’s inference prediction latency compared to the state-of-the-art inference model.

The main challenges in the 5G/6G network slicing are resource sharing, network slice orchestration, and network optimization in the 5G ecosystem. This paper proposes a novel architecture for a dynamic network slice broker and marketplace named “SliceGPT” that leverages Custom-Trained OpenAI GPT-3.5 LLM, blockchain and NFTs to enable collaboration between different stakeholders in the 5G ecosystem to address these challenges. The platform enables different stakeholders in 5G network slicing (e.g., cloud providers, network operators, RAN providers, and transport network providers) to share and rent their resources to create customized network slices that meet the specific requirements of 5G applications. The orchestration of network slices is managed through blockchain smart contracts, and the resulting network slices are encoded as NFT tokens and made available for purchase in a decentralized NFT marketplace. Customers can select and purchase the network slices that best meet their needs by paying either crypto or flat currency. Revenue generated through the sale of network slices is distributed among different providers, facilitating a fair and efficient marketplace. Intelligent network slice optimization is accomplished through the utilization of a custom-trained GPT-3.5 LLM(which powers the ChatGPT). This LLM can generate valuable insights and recommendations from extensive network datasets, contributing to the optimization of network slices for enhanced performance and efficiency. A prototype of SliceGPT has been implemented with FreedomFi 5G gateway, OpenAirInterface 5G core, OpenAI GPT-3.5-turbo model, LlamaIndex and Langchain. To the best of our knowledge, this is the very first research endeavor to incorporate the GPT LLMs for optimizing 5G/6G network slicing,

Traditional and indigenous medicine, deeply rooted in ancient traditions and wisdom, plays a crucial role in global healthcare and cultural identity. These practices provide treatments for illnesses such as cancer and bone injuries, which often lack effective remedies in Western medicine. However, these valuable systems face challenges like potential knowledge loss, undervaluation of practitioners’ expertise, and the risk of fraud due to the absence of credential verification mechanisms. In this research, we introduce "WedaGPT," a Generative AI-enabled platform that utilizes a custom-trained Meta’s Llama2 Large Language Model (LLM), Blockchain, self-sovereign identity (SSI), Non-Fungible Tokens (NFTs), and model cards to share traditional medical knowledge and address these issues. WedaGPT creates a collaborative ecosystem connecting doctors, medicine providers, therapists, patients, and technology experts, all committed to preserving and advancing traditional healing practices. This platform enables secure and transparent contributions from all stakeholders to patient well-being. Ancient medical recipe books are translated into English and digitized into PDF formats to enrich the platform’s knowledge base. These texts are used to fine-tune the Llama2 LLM, which has been quantized and optimized with Qlora for performance on consumer-grade hardware. Through a chat-based interface in the SSI-enabled mobile wallet, users can interact with the LLM and access detailed information on treatments, recipes, prescriptions, and healing methods. Additionally, users can consult remotely with doctors who prescribe treatments through this wallet. A key feature of WedaGPT is transforming ancient medicinal recipes into NFT tokens for sale on NFT marketplaces, giving traditional knowledge digital authenticity and economic value. Revenue from these sales is distributed among platform contributors, promoting equitable ownership and recognition. Medical recipe data, including treatment histories and physician details, are encapsulated in Model Cards and securely stored on the blockchain. This system offers mechanisms to verify doctors and treatments in a privacy-preserving way, potentially reducing fraud and medication errors.

In recent years, online marketplaces have seen a large increase in business. Managing and making sellers' reputations available to buyers is vital for the success of these markets since buyers may be dealing with unknown online sellers. Typically, prospective buyers assess sellers' performance by looking at past buyers' reviews, prior to making a purchase. In current online marketplaces, buyers' reviews are associated with their identities, raising privacy issues. Several decentralized solutions have been proposed using blockchain technologies to solve this problem. However, some of these systems are not entirely decentralized or have some weaknesses such as requiring a trusted third party. In this paper, we propose a system for decentralized reviewing that trades off cost for unlinkability. Using blockchain-based smart contracts, we propose a fully decentralized marketplace to provide reviewers with a secure and trusted platform. Here, buyers can use one-time identities to provide feedback on their transactions. The system also ensures that a review can only be submitted after a transaction is completed, and that at most one review may be submitted per transaction. The system enforces these properties through blockchain technologies and smart contracts. In addition, we propose an innovative approach to encourage buyers to submit reviews. The system has been prototyped using Remix IDE.

Very recently, in an attempt to reduce the uncertainty associated with notoriously unreliable buyer feedback, researchers proposed a blockchain-based trust and reputation management system, where a Smart Contract manages all aspects of a transaction: setting up a contract, interacting with the two parties, and finally evaluating and providing feedback on the buyer/seller performance at the end of each transaction. They have also proposed a data structure that manages seller reputation scores inside the blockchain. In this paper, we provide two novel reputation management schemes that significantly improve on these methods. Our first such scheme is adaptive; the second one is randomized, wherein decisions about managing the underlying data structure are made based on coin flips. We provide analytical performance predictions and verify, empirically, by extensive simulation, the accuracy of our predictions. Our schemes are shown to result in more efficient query execution with enhanced block structure schemes.

In recent years, the use of UAVs has expanded to various applications such as surveillance, disaster response, agriculture, and delivery. However, traditional UAV monitoring systems rely on direct communication between the UAV and the ground pilot, which has several limitations such as limited range, poor reliability, and susceptibility to interference. To overcome these limitations, there has been significant interest in integrating UAVs into cellular networks such as 5G/6G network slicing. The flexibility of network slicing allows UAVs to operate on different slices based on their communication needs, which can improve their performance and efficiency. However, integrating UAVs into network slicing also poses several challenges, such as managing communication and permissions of UAVs and base stations, access control of UAVs, and identity management of UAVs. To address these challenges, we propose a blockchain, Non-Fungible Token(NFT), Federated Learning(FL), and Zero-Trust(ZT) security-enabled UAV monitoring platform for 5G/6G network sliced environments. We propose a novel approach in which UAVs are represented as NFT tokens within the platform. This innovative representation allows for enhanced security and trust in the system, aligning with the principles of the Zero-Trust security model, which assumes no implicit trust in any network component or user. Furthermore, we propose a FL system that operates on top of the blockchain, which can analyze data from multiple UAVs across different network slices. Our proposed FL system uses coordinator-less models, which eliminates the attacks of a centralized coordinator. As a use case, we consider a scenario where our proposed system detects anomaly communications of UAVs and identifies attack surfaces via analyzing network traffic data of UAVs using FL. The 5G system testbed implemented with FreedomFi 5G gateway and Indoor Radio Cell.

Roughly a decade ago, inspired by the phenomenal success of cloud computing, a group of researchers have defined Vehicular Clouds as a group of vehicles whose sensing, communication, and computing resources can be coordinated and allocated to authorized users. While both conventional and Vehicular Clouds are instances of utility computing, a number of important characteristics set vehicular clouds apart from their conventional counterparts. These characteristics include the mobility of vehicles and the volatility of resources that fluctuate with the arrival and departure of vehicles. As in the conventional version of cloud computing, approximating job completion time is one of the key performance metrics of interest. Unfortunately, estimating job completion time with any degree of accuracy and confidence requires complete knowledge of the distributions of several relevant random variables. Typically, however, these probability distributions are not known. Luckily, in many cases of practical relevance, accumulated empirical evidence allows to estimate the first few moments of these random variables. The main contribution of this paper is to offer an accurate approximation of the expected job completion time in Dynamic Vehicular Clouds built on top of vehicles on a highway. For this purpose, we rely on empirical estimates of the first moment of the user job execution time in the absence of any overhead attributable to the Dynamic Vehicular Cloud itself. Our extensive simulations have confirmed that our approximations of the expected job execution time are very accurate.

Federated learning is a collaborative/distributed machine learning system which is designed to address the privacy issues in centralized machine learning systems. The transparency and provenance of a machine learning model are important aspects of federated learning systems since they impact peoples’ lives in various domains (e.g., from healthcare to personal finance to employment). However, most of the existing federated learning systems deal with centralized coordinators which are vulnerable to attacks and privacy breaches. Also, they do not provide any standard transparency and provenance mechanisms for the resulting models. In this paper, we propose a blockchain and Model Card-based integrated federated learning system "Bassa-ML" providing enhanced transparency and trust for the models. Model parameter sharing, local model generation, model averaging, and model sharing functions are implemented using smart contracts. The generated models, model training information, and model reports are stored in the blockchain ledger as Model Card Objects. This results in enhanced transparency and auditability to the federated learning process.

The network slicing in 5G/6G mobile networks enables billions of connected devices to transmit data at higher rates than ever before. The high number of devices and the huge data rates result in configuration complexities and complex security management. Machine learning techniques could play a key role in managing these system complexities. While feder-ated learning (FL) has recently been proposed as an emerging paradigm to build privacy-preserving machine learning models, many of the existing systems involve centralized coordinators which are known to be vulnerable to attacks and privacy breaches. In addition, current FL models have weak support for transparency and provenance mechanisms. In this paper, we propose a Blockchain-based, Zero-trust Security-enabled Federated Learning system “Skunk” to address privacy and data provenance requirements. The proposed federated learning system also supports the requirements of 5G/6G networks. The sharding-based architecture in the blockchain enables the deployment of Skunk in 5G/6G network slice environments. As a use case of Skunk, we have considered a scenario with IoT device attacks in a 5G/6G network. The proposed FL models detect such attacks in the 5G/6G network sliced environment.

5G network slicing enables IoT networks to connect billions of heterogeneous objects providing high quality of service, high network capacity, and enhanced system throughput. Despite all these advantages, there are some major challenges to be addressed including decentralization, transparency, data interoperability, network privacy and security, and network slice orchestration, data provenance, and management. Blockchain technologies have the potential to offer innovative solutions to overcome these challenges. However, in the context of 5G enabled scalable IoT applications, integrating 5G with blockchain platforms could pose challenges to 5G’s goals such as high transaction throughput, high scalability, and real-time transaction processing, sharding-based consensus, network slice management and provenance. In this paper, "Moose," a blockchain platform, to overcome these challenges is proposed. It supports sharding based consensus in the blockchain network. It integrates a network slice orchestration library with smart contracts to manage and schedule network slices. As a use-case, Moose is integrated with a 5G-supported IoT device identity monitoring system on a network sliced environment. The performance results from the implemented system indicate that the proposed system indeed overcomes the aforementioned challenges.

Cryptocurrencies are more than a decade old and several issues have been discovered since their then. One of these issues is a partial negation of the intent to “democratize” money by decentralizing control of the infrastructure that creates, transmits, and stores monetary data. The Programmatic Proof of Work (ProgPoW) algorithm is intended as a possible solution to this problem for the Ethereum cryptocurrency. This paper examines ProgPow’s claim to be Application Specific Integrated Circuit (ASIC) resistant. This is achieved by isolating the proof-of-work code from the Ethereum blockchain, inserting the ProgPoW algorithm, and measuring the performance of the new implementation as a multithread CPU program, as well as a GPU implementation. The most remarkable difference between the ProgPoW algorithm and the currently implemented Ethereum Proof-of Work is the addition of a random sequence of math operations in the main loop that require increased memory bandwidth. Analyzing and comparing the performance of the CPU and GPU implementations should provide an insight into how the ProgPoW algorithm might perform on an ASIC.

Inspired by the success of conventional cloud services and by the reality of present-day vehicles endowed with powerful on-board computers that can act as servers in a datacenter, researchers have recently introduced the concept of a vehicular cloud. Our main contribution is to offer a tight theoretical analysis of the expected job completion time in vehicular clouds characterized by short vehicular residency times, under a redundancy-based job assignment strategy. We also discuss various approximations of the expected completion time. A comprehensive set of simulations have confirmed the accuracy of our theoretical predictions.

Motivated by the success of cloud computing, vehicular clouds were introduced as a group of vehicles whose corporate computing, sensing, communication, and physical resources can be coordinated and dynamically allocated to authorized users. Our main contribution is to offer an easy-to-compute approximation of job completion time in a dynamic vehicular cloud model involving vehicles on a highway. We assume estimates of the first moment of the time it takes the job to execute without any overhead attributable to the working of the vehicular cloud. Our simulations have shown that our approximation is very accurate. To the best of our knowledge, this is the first paper dealing with estimating job completion time in dynamic vehicular clouds.

BACKGROUND: Profiling gene expression in brain structures at various spatial and temporal scales is essential to understanding how genes regulate the development of brain structures. The Allen Developing Mouse Brain Atlas provides high-resolution 3-D in situ hybridization (ISH) gene expression patterns in multiple developing stages of the mouse brain. Currently, the ISH images are annotated with anatomical terms manually. In this paper, we propose a computational approach to annotate gene expression pattern images in the mouse brain at various structural levels over the course of development. RESULTS: We applied deep convolutional neural network that was trained on a large set of natural images to extract features from the ISH images of developing mouse brain. As a baseline representation, we applied invariant image feature descriptors to capture local statistics from ISH images and used the bag-of-words approach to build image-level representations. Both types of features from multiple ISH image sections of the entire brain were then combined to build 3-D, brain-wide gene expression representations. We employed regularized learning methods for discriminating gene expression patterns in different brain structures. Results show that our approach of using convolutional model as feature extractors achieved superior performance in annotating gene expression patterns at multiple levels of brain structures throughout four developing ages. Overall, we achieved average AUC of 0.894 ± 0.014, as compared with 0.820 ± 0.046 yielded by the bag-of-words approach. CONCLUSIONS: Deep convolutional neural network model trained on natural image sets and applied to gene expression pattern annotation tasks yielded superior performance, demonstrating its transfer learning property is applicable to such biological image sets.

Data integrity is a major concern in outsourced IT services like cloud computing. Cloud computing has become popular because of cost reductions, time saving and mobility in service. However data integrity is still an unresolved issue in cloud services. We present an efficient mechanism for evaluating data integrity in cloud database-as-a-service. Our approach is based on inserting fake tuples into the database. In our model the owner of the data is the only trusted party and the server as a service provider or any other users are not trusted. We refer to distrusted party as a potentially malicious attacker. In our approach we define generating functions to create fake tuples with uniform distribution. Malicious attackers are not able to distinguish between fake tuples and real tuples. Our approach does not use encryption which makes it more efficient. We explore the strengths and limitations of these generating functions by describing our approach.

With the growing demand for data-sensitive applications employing mobile devices, such as mobile clinics in remote villages and remote sensors collecting sensitive data, there is a need for a new architectural paradigm for mobile data management. Typically, these mobile devices have limited storage and processing capabilities, and operate in unreliable environments leading to possible loss of valuable data, if not properly managed. Often, the collected data is periodically offloaded to a remote server such as a cloud. However, such offloading may lead to violation of privacy if the network/server cannot be fully trusted. While encrypting the data prior to offloading appears to be a solution for this problem, this is computationally intensive and infeasible when mobile devices are employed. In this paper, we propose a partial-encryption scheme that takes into account both the privacy (confidentiality) constraints of the data being collected and the limitations of the mobile devices. The scheme employs vertical and horizontal fragmentation to determine those parts that need to be encrypted and those that can be sent in clear. The privacy constraints are represented in terms of a constraint graph and two-coloring problem solution is applied to identify the portions of the data that need to be encrypted. Any cycles in the constraint graph are handled using heuristics. The approach is effective in integrating unsecured wireless/internet with the untrusted yet cheap cloud storage servers, using the capacity-constrained mobile devices to manage sensitive data.

With the increasing growth of cloud computing and the resulting outsourcing of data, concerns of data integrity, security, and privacy are also on the rise. Among these, evidence of data integrity - being tamper-evident and up-to-date, seem to be of immediate concern. While several integrity techniques currently exist, most result in significant storage overhead at the data owner site. For clients with large data sets, these are not viable solutions. In this paper, we propose a space-efficient alternative - data integrity using Bloom filters. We propose the basic method and discuss different alternatives to implement the scheme based on the trust/threat model and processing/storage capacity at the server and the client. For one of these alternatives, we present a detailed analysis and experimental results. These results are compared with the traditional security hash functions such as SHA-1 and MD5. The Bloom filter implementations are shown to be highly space-efficient at the expense of additional computational overhead. To overcome this bottleneck, we have implemented the schemes on multiprocessor systems. The multicore implementations have significantly reduced the execution time. Our results clearly demonstrate the feasibility and efficacy of employing Bloom filters to enforce data integrity for outsourced data in cloud environments.

One of the government missions identified by the federal enterprise architecture is to use computer and networking technologies to develop infrastructure to support information sharing within government organizations as well as with external stakeholders. Currently, considerable information is being maintained at individual organizations in the form of large repositories/digital libraries with no efficient means of sharing it with other government organizations and with other external user communities, including the general public. A major obstacle to information sharing is the lack of a framework and an infrastructure that allows government organizations to share information selectively with different user groups. Lack of such a framework creates unwillingness among government organizations to share their digital content. A mechanism needs to be in place where policy makers can specify which documents can be moved from one organization to another organization and/or who can access these transferred documents. Furthermore, a system is needed that enforces these policies in realtime when external events dictate a change of policies. In this paper, we propose a framework for specification, management and enforcement of dynamic access policies across multiple geographically distributed organizations. The framework can be instantiated to integrate with individual digital library systems and provide the necessary infrastructure to provide policy controlled access control management

In a dynamic coalition environment, organizations should be able to exercise their own local fine-grained access control policies while sharing resources with external entities. In this paper, we propose an approach that exploits the semantics associated with subject and object attributes to facilitate automatic enforcement of organizational access control policies while resource sharing occurs among coalition members. Our approach relies on identifying the necessary attributes required by external users to gain access to a specific organizational object (or service). Specifically, it consists of extracting user attribute sets that semantically match with the attributes of the objects for which a role has permissions. This relies on a closer examination of why a user is assigned a specific role. These attribute sets are first pruned based on their significance in characterizing a role, which are then checked against those submitted by an external user to decide whether to allow or deny access to the specific object. While our goal in this paper is to support coalition based access control, the proposed approach can also aid in automating the process of role engineering.

Abstract Securing sensor networks has received much attention in the last few years. In a typical field deployment, a sensor network self‐organizes, closely interacts with its physical environment, and works unattended possibly in a hostile environment such as a battlefield. In such environments, sensor networks are subject to node capture as well as a myriad of other attacks. Constrained energy, memory, and computational capabilities of sensor nodes mandate a clever design of security solutions to minimize overhead while maintaining secure communications over the lifespan of the network. In this paper, we propose a Dynamic Combinatorial Key management scheme, DCK, to provide efficient, scalable, and survivable dynamic keying in a clustered sensor network with a large number of sensor nodes. DCK employs the Exclusion‐Basis Systems (EBS) as the underlying framework for key management at both the cluster and the sensor node levels. DCK enhances network security by localizing cluster key management functions, thus, limiting the impact of sensor node capture to the attacked cluster. Within a cluster, DCK decouples key generation, assignment, and distribution, and distributes the key management tasks among cluster nodes. Simulation results show that DCK is efficient in terms of energy consumption and storage. Also, it significantly outperforms other dynamic keying schemes, in particular with regards to energy consumed in key refreshment and re‐keying after node capture. Copyright © 2006 John Wiley & Sons, Ltd.

With the increasing growth in global enterprises and collaborations among the enterprises, security and trust have become essential for information systems. For example, within an enterprise, there may be a need to maintain security within each project group so the information sharing among the groups is controlled. Similarly, there may be a need to facilitate controlled and timed sharing of data among cooperating enterprises (e.g., coalitions). In this paper, we propose a policy-based security mechanism for such sharing in an enterprise. In particular, in our system, each user (or administrator) specifies restrictions on the use of resources at a particular node (or machine) in terms of a set of policy statements (NRPS and NTPS). Similarly, the owner of each object specifies the conditions on which certain operations can be performed on the object (ORPS and OTPS). Trusted policy enforcement agents (PEA), running at each node in the enterprise (or coalition), ensure that both node and object policies are enforced in the system. We show how the proposed system facilitates dynamic control at object-level and machine-level. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Secure group-based communication services are gaining momentum in wireless ad hoc networks (WAHNs). The efficiency and scalability of these services make them suitable for supporting the inherently collaborative computing and communications activities among peer WAHN nodes as well as the growing number of group-based WAHN applications. Key management, which is comprised of dynamic key generation, assignment, and distribution, is at the heart of secure group communications. This work proposes TKGS; an efficient threshold-based key generation scheme for open WAHNs based on collaborative reconstruction of key-generation keys. To mitigate node mal-behavior, a verifiable secret sharing technique is employed, where an (authenticated) dealer node is selected to collect a set number of valid shares and generate new administrative keys. While most existing key management schemes assume reliable and trusted nodes, TKGS provides a distributed solution that is resilient to node failure and mal-behavior. A major advantage of TKGS is its small overhead; collaborative key construction is done in less than two rounds with a very high probability of success.

Computing and communications in wireless ad hoc networks (WAHNs) generally require collaboration among groups of peers. This, in addition to a growing number of group applications over WAHNs, have motivated research in secure group communication services as a means for efficient and secure communications in WAHNs. Key distribution is at the heart of secure group communications. Existing key distribution schemes, designed for infrastructure networks, tend to be inappropriate for the infrastructure-less WAHNs. Also, most of these schemes assume network-level multicast which is difficult to implement in WAHNs. We propose a new efficient and scalable combinatorial key distribution scheme (CKDS) to support secure group communications in WAHNs. CKDS partitions nodes over a virtual Cartesian key space and uses combinatorial exclusion basis systems for key distribution over application-level multicast. We employ a fully distributed unicast key distribution underlying a virtual application-level multicast infrastructure. Two variants of CKDS are proposed, namely, m-dimensional multicast and 2D multicast. Performance analysis shows that these schemes achieve lower network traffic overhead as well as lower computational overhead per node compared to other unicast key distribution schemes in WAHNs. We also show our scheme to be scalable with respect to both computational and storage needs.

Since 1990, transaction processing in multilevel secure database management systems (DBMSs) has been receiving a great deal of attention from the security community. Transaction processing in these systems requires modification of conventional scheduling algorithms and commit protocols. These modifications are necessary because preserving the usual transaction properties when transactions are executing at different security levels often conflicts with the enforcement of the security policy. Considerable effort has been devoted to the development of efficient, secure algorithms for the major types of secure DBMS architectures: kernelized, replicated, and distributed. An additional problem that arises uniquely in multilevel secure DBMSs is that of secure, correct execution when data at multiple security levels must be written within one transaction. Significant progress has been made in a number of these areas, and a few of the techniques have been incorporated into commercial trusted DBMS products. However, there are many open problems remain to be explored. This paper reviews the achievements to date in transaction processing for multilevel secure DBMSs. The paper provides an overview of transaction processing needs and solutions in conventional DBMSs as background, explains the constraints introduced by multilevel security, and then describes the results of research in multilevel secure transaction processing. Research results and limitations in concurrency control, multilevel transaction management, and secure commit protocols are summarized. Finally, important new areas are identified for secure transaction processing research.

Workflow management systems (WFMS) support the modeling and coordinated execution of processes within an organization. To coordinate the execution of the various activities (or tasks) in a workflow, task dependencies are specified among them. Often, the workflow application domains are such that the workflow is a long-running activity and the various tasks that constitute the workflow need to be executed by systems that are distributed and autonomous in nature, possibly owned by di#erent organizations. In such an environment, it is desirable to minimize the number of communications among the distributed sites and minimize the number of tasks that need to wait for the execution of tasks at other sites. In [3], Atluri et al. propose an approach that can automatically redesign a workflow in such a way that it minimizes the number of communications and interference between sites. This approach performs a semantic categorization of task dependencies, and proposes to use for each categoriza...

Database systems for real-time applications must satisfy timing constraints associated with transactions in addition to maintaining data consistency. In addition to real-time requirements, security is usually required in many applications. Multi-level security requirements introduce a new dimension to transaction processing in real-time database systems. In this paper, we argue that, due to the conflicting goals of each requirement, tradeoffs need to be made between security and timeliness. We first define mutual information, a measure of the degree to which security is being satisfied by a system. A secure two-phase locking protocol is then described and a scheme is proposed to allow partial violations of security for improved timeliness. Analytical expressions for the mutual information of the resultant covert channel are derived, and a feedback control scheme is proposed that does not allow the mutual information to exceed a specified upper bound. Results showing the efficacy of the scheme obtained through simulation experiments are also discussed.

Proposes a novel model to characterize the behavior of individual programs in the presence of cache memory. The model goes beyond the traditional hit/miss ratios often used in the current literature to describe program-cache interactions. Instead, it looks at the characteristics of the sequences of hits and misses during a program execution. The model includes several distributions that describe the hit and miss sequences. Several program characteristics which are otherwise not identified by the traditional models are discovered using our model. We then discuss how the model can be effectively used to tune the performance of a program, to allocate cache to a process, or to choose a cache architecture.

Two important aspects that any study of message communication has to address are routing and admission control. The routing problem seeks to find a route for a channel and admission control involves assessing the ability to meet the demands of a channel along the chosen route. Most efforts in the area of real-time communication have been directed, primarily towards the admission control problem, not many have been targeted rewards the routing problem. The authors show that these two problems are inter-related. They address these two problems in a general framework that can abstract many practical scenarios. They assume the use of an arbitrary dynamic/fixed priority link level scheduling, thereby increasing the utility of the derived results. Their approaches for both routing and admission control are based on extending a result we have derived in a different context, viz., task scalability. A simulation study, was performed to study the effectiveness of their approach in improving both utilization of the link and admissibility of channels.

A replicated-architecture database system uses data replication to provide multilevel security. The most critical problem associated with replicated architecture multilevel-secure database systems is mutual consistency of the replicas and the impact

With the proliferation of scheduling algorithms there is a growing need to test these schedulers for their validity not just at design time but also as the system evolves. This implies that the schedulability analysis has to be robust. In this study, we identify a few often posed questions that address the robustness of schedulability analyses. First these questions are dealt in the context of uniprocessor systems and then we handle some of their extensions in a more general context of end-to-end systems. We show that these questions are closely related to a more general problem. We present a solution to this problem. An intuitive proof of correctness and optimality of the solution technique are presented.

Database systems for real-time applications must satisfy timing constraints associated with transactions, in addition to maintaining data consistency. In addition to real-time requirements, security is usually required in many applications. Multilevel security requirements introduce a new dimension to transaction processing in real-time database systems. We argue that due to the conflicting goals of each requirement, trade-offs need to be made between security and timeliness. We first define capacity, a measure of the degree to which security is being satisfied by a system. A secure two-phase locking protocol is then described and a scheme is proposed to allow partial violations of security for improved timeliness. The capacity of the resultant covert channel is derived and a feedback control scheme is proposed that does not allow the capacity to exceed a specified upper bound.>

Since protection and assurance are the primary concerns in multilevel secure (MLS) databases, performance has often been sacrificed in some known MLS database approaches. Motivated by performance concerns, a replicated architecture approach which uses a physically distinct back-end database management system for each security level is being investigated. This is a report on the behavior and performance issues for the replicated architecture approach. Especially, we compare the performance of the SINTRA (Secure INformation Through Replicated Architecture) MLS database system to that of a typical conventional (non-secure, single-level) database system. After observing the performance bottlenecks for SINTRA, we present solutions that can alleviate them.>

Data availability and security are two important issues in a distributed database system. Existing schemes achieve high availability at the expense of higher storage cost and data security at the expense of higher processing cost. We develop an integrated methodology that combines the features of some existing schemes dealing with data fragmentation, data encoding, partial replication, and quorum consensus concepts to achieve storage efficient, highly available, and secure distributed database systems.>

Four factors in parallel token ring systems which can improve network performance are identified. An analytical model is developed to predict the performance of these systems. The predictions obtained with the analytical model are compared with simulation results. While the current model accurately predicts the performance of networks with 16 or more rings, it is not so accurate at lower numbers of rings. The short/long cycle behavior of the token interarrival times is identified as one cause of the inaccuracies. The benefits and limitations of parallel token ring networks for gigabit speeds are discussed.>

The paper by Ahad et al. [1] derives an analytical expression to estimate the cardinality of the projection of a database relation. In this note, we propose to show that this expression is in error even when all the parameters are assumed to be constant. We derive the correct formula for this expression.

Many different data compression techniques currently exist. Each has its own advantages and disadvantages. Combining (pipelining) multiple data compression techniques could achieve better compression rates than is possible with either technique individually. This paper proposes a pipelining technique and investigates the characteristics of two example pipelining algorithms. Their performance is compared with other well-known compression techniques.

Database designers and researchers often need efficient schemes to evaluate transaction performance. In this paper, we chose two important performance measures: the average number of nodes accessed and the average number of data items accessed per node by a transaction in a distributed database system. We derive analytical expressions to evaluate these metrics. For general applicability, we consider partially replicated distributed database systems. Our first set of analytic results are closed-form expressions for these two measures. These are based on some fairly restrictive simplifying assumptions. When these assumptions are relaxed, no closed-form expressions exist for these averages. Hence, we develop an efficient algorithm to compute these averages.

The effect of assumptions about data distribution and replication on the performance measures as well as the computational complexity and accuracy of performance evaluations is investigated. The size of the participating node set of a transaction is chosen as the desired performance measure. Probabilistic analysis is used to evaluate six models. It is concluded that even though some of the data distribution and replication models appear to be simplistic, the results obtained from them are very close to those from complex models.>

Recently protocols have been introduced that enable the integration of synchronous traffic (voice or video) and asynchronous traffic (data) and extend the size of local area networks without loss in speed or capacity. One of these is DRAMA, a multiband protocol based on broadband technology. It provides dynamic allocation of bandwidth among clusters of nodes in the total network. In this paper, we propose and evaluate a number of traffic placement policies for such networks. Metrics used for performance evaluation include average network access delay, degree of fairness of access among the nodes, and network throughput. The feasibility of the DRAMA protocol is established through simulation studies. DRAMA provides effective integration of synchronous and asynchronous traffic due to its ability to separate traffic types. Under the suggested traffic placement policies, the DRAMA protocol is shown to handle diverse loads, mixes of traffic types, and numbers of nodes, as well as modifications to the network structure and momentary traffic overloads.

The effect of simplistic assumptions about the data distribution and replication in a system on performance measures and the computational complexity and accuracy of evaluations of performance measures is investigated. The size of the participating node set of a transaction is chosen as the desired performance measure. A data distribution and replication model is represented by four key parameters. Probabilistic analysis is used to evaluate six of these models. It is concluded that even though some of the data distribution and replication models appear to be simplistic, the results obtained from them are very close to those from complex models. In addition, the gains due to drastically reduced execution times strongly suggest the use of simple models (at least) in the early stages of the design process.>

An O(n/sup 2/) heuristic algorithm using randomized decisions is developed for determining a near-optimal set of nodes. For small values of n, the authors can determine how close the heuristic solution is to the optimal set of nodes. They also compare their heuristic to other algorithms reported in the literature.>