William M. Mongan

Development and Specification of a Reference Model for Agent-Based Systems

Agent-based systems have been the object of intense research over the past decade. While great theoretical progress has been made, the software frameworks for creating agent-based systems offer considerable variability in their capabilities and functionality. This paper introduces a reference model for agent-based systems. The purpose of a reference model is to provide a common conceptual basis for comparing systems and driving the development of software architectures and other standards. The Foundation for Intelligent Physical Agents and other groups have advanced a number of agent standards, yet, to date, no comprehensive reference model has been presented for software systems composed of agents. This paper provides an overview of a reference model for agent-based systems. The agent systems reference model is the result of a multiyear effort studying software systems built with agents and software frameworks for implementing these systems. As part of this study, the team applied software reverse engineering techniques to perform static and dynamic analysis of operational agent-based systems. This analysis enabled identification of key common concepts across over one dozen different agent frameworks. To demonstrate its applicability, the reference model is then used to analyze a number of complete agent-based software systems. It is the belief of the authors that the reference model will be an essential prerequisite for future transition, deployment, and integration of agent-based systems.

On the Use of Knitted Antennas and Inductively Coupled RFID Tags for Wearable Applications

Recent advancements in conductive yarns and fabrication technologies offer exciting opportunities to design and knit seamless garments equipped with sensors for biomedical applications. In this paper, we discuss the design and application of a wearable strain sensor, which can be used for biomedical monitoring such as contraction, respiration, or limb movements. The system takes advantage of the intensity variations of the backscattered power (RSSI) from an inductively-coupled RFID tag under physical stretching. First, we describe the antenna design along with the modeling of the sheet impedance, which characterizes the conductive textile. Experimental results with custom fabricated prototypes showed good agreement with the numerical simulation of input impedance and radiation pattern. Finally, the wearable sensor has been applied for infant breathing monitoring using a medical programmable mannequin. A machine learning technique has been developed and applied to post-process the RSSI data, and the results show that breathing and non-breathing patterns can be successfully classified.

Towards a reference model for agent-based systems

The current state of the art in agent technology sees that several implementations of agent frameworks exist. However, there is little agreement on the terms and concepts used to describe such systems, which is a significant barrier towards adoption of these technologies by industry, military and commercial entities. A clear definition of terms and concepts at an appropriate level of abstraction is needed to facilitate discussion, evaluation and adoption of these emerging agent technologies. In this paper, we argue that a reference model for agent-based systems can fill this need. We discuss what a reference model is, why one is needed for agent-based systems, and our proposed methodology for creating such a reference model. While the complete model is a work in progress, we present a preliminary version to motivate further discussion from the agents community at large. It is our hope that ultimately a wider community of practice will assume responsibility for the standardization similar to the way that the well-known seven-layer Open Systems Interconnection (OSI) reference model was a driving force underlying communications standards.

A Multi-Disciplinary Framework for Continuous Biomedical Monitoring Using Low-Power Passive RFID-Based Wireless Wearable Sensors

We have applied passive Radio Frequency Identification (RFID), typically used for inventory management, to implement a novel knit fabric strain gauge assembly using conductive thread. As the fabric antenna is stretched, the strength of the received signal varies, yielding potential for wearable, wireless, powerless smart-garment devices based on small and inexpensive passive RFID technology. Knit fabric sensors and other RFID biosensors can enable comfortable, continuous monitoring of biofeedback, but requires an integrated framework consisting of antenna modeling and fabrication, signal processing and machine learning on the noisy wireless signal, secure HIPAA- compliant data storage, visualization and human factors, and integration with existing medical devices and electronic health records (EHR) systems. We present a multidisciplinary, end-to-end framework to study, model, develop, and deploy RFID-based biosensors.

A methodology for developing an agent systems reference architecture

The slow adoption of agent-oriented methodologies as a paradigm for developing industry systems is due in part to their lack of integration and general-purpose use. There exists a need to define common patterns, relationships between components, and structural qualities that a reference architecture for agent-based systems would solve. However, there is little, if any, consensus on how to create a reference architecture for agent-based systems. This paper presents a methodology for developing a reference architecture that documents agent-based systems from different system viewpoints. Rather than the traditional approach of studying existing systems, the documentation methodology relies on forensic software analysis of agent frameworks (i.e., APIs and libraries for constructing agent systems). We demonstrate the methodology by describing the process used to create the Agent System Reference Architecture.

Raspberry HadooPI: A Low-Cost, Hands-On Laboratory in Big Data and Analytics (Abstract Only)

Educating STEM students in the techniques of massively parallel computing anticipates a growing current and future need for scientists, engineers, and analysts who are facile with Big Data. Using very low cost hardware (Raspberry Pi) and free software (Hadoop) we are exposing students to distributed computing while limiting expense. We anticipate that micro-cluster labs and projects will give students hands on experience necessary so they can be prepared to use these methods in real world applications. A series of lessons and projects were developed to teach Hadoop and MapReduce, and were extended into STAR (Students Tackling Advanced Research) summer competitive research projects.

Statistical analytics of wearable passive RFID-based biomedical textile monitors for real-time state classification

Wearable smart devices have become ubiquitous, with powered devices capable of collecting real-time biometric information from its users. Typically, these devices require a powered component to be worn and maintained, such as a battery-powered sensor, Bluetooth communications device, or glasses. Pregnancy and infant monitoring devices may be uncomfortable to the mother or baby and are subject to signal loss if the patient changes position or becomes mobile because the device must remain tethered to the patient by a belt and plugged into a wall for power. Our wearable, wireless, smart garment devices are knitted into the fabric using conductive thread to which a Radio Frequency Identification (RFID) chip within the fabric is inductively coupled. Our work utilizes the Received Signal Strength Indication (RSSI), which changes as the knitted antenna is deformed due to stretching of the garment, to determine different types of motion in the inductively-coupled chip and knit antenna structure as it is moved by the wearer.

Real-time detection of apnea via signal processing of time-series properties of RFID-based smart garments

Signal processing of time-series properties of Radio Frequency Identification (RFID) tags and novel work in textile knitted antennas for garment devices have enabled real-time detection of motion-based artifacts through unobtrusive, wireless, wearable devices. Capturing the Received Signal Strength Indicator (RSSI) as a time-series signal, we classify whether the subject is breathing or not, estimate the rate at which the subject is breathing, and classify whether the tag is moving in a linear, non-stretched fashion. We improve upon previous efforts to classify subject state from RSSI signals by eliminating the need to train the classifier with both breathing and non-breathing sample data (which is biologically infeasible). To test our approach, we use a programmable breathing infant mannequin yielding accurate detection of cessation of respiratory activity within 5 seconds, and a maximum root-mean-square error of 7 per minute when computing the respiratory rate.

Development and Specification of a Reference Architecture for Agent-Based Systems

The recent growth of agent-based software systems was achieved without the development of a reference architecture. From a software engineering standpoint, a reference architecture is necessary to compare, evaluate, and integrate past, current, and future agent-based software systems. The agent systems reference architecture (ASRA) advances the agent-based system development process by providing a set of key interaction patterns for functional areas that exist between the layers and protocols of agent-based systems. Furthermore, the ASRA identifies the points for interoperability between agent-based systems and increases the level of discussion when referring to agent-based systems. This paper presents methodology, grounded in software forensics, to develop the ASRA and provides an overview of the resulting architectural representation. The methodology uses an approach based on software engineering techniques adapted to study agent frameworks-the libraries and tools for building agent systems. The resulting ASRA can serve as an abstract representation of the components necessary for facilitating comparison, integration, and interoperation of software systems composed of agents.

On the Use of Radio Frequency Identification for Continuous Biomedical Monitoring

Radio Frequency Identification (RFID) technology is often deployed for inventory management scenarios. In inventory applications, a known or unknown number of RFID tags are queried in a discrete manner and for a single, short period of time, until each tag is recognized by the interrogator device. Passive RFID provides several benefits conducive to ubiquitous deployment, including RFID tags that are energized from the wireless RF interrogation signal itself that obviates the need for a battery or wired power, and antenna assemblies that can be integrated with the chip with only a small footprint.We have utilized these benefits to enable continuous biomedical sensing devices with minimal footprint and batteryless deployment.These devices are fabric-based smart garments with an embedded RFID tag and antenna assembly. However, traditional inventory-based RFID interrogation presents several challenges due to the RFID protocols and regulations that govern their use. In this paper, we discuss the considerations necessary to utilize RFID interrogation to enabling passive, continuous sensor monitoring, and the techniques we employed in developing software to do so.