Neural networks have demonstrated superior performance over rule-based and model-based approaches in processing noisy sensing data. However, their substantial computational and energy demands hinder deployment in battery-powered embedded systems. Computing-in-Memory (CIM) devices offer a promising alternative by significantly reducing energy consumption. Prior work [2] achieves this by leveraging a non-von Neumann architecture, which minimizes data movement between memory and compute units, thereby mitigating the memory wall bottleneck. Despite these advantages, analog CIM (ACIM) systems face several key challenges, including analog noise, limited numerical precision, and increased hardware complexity. While cloud-based neural networks are still dominant, emerging applications increasingly demand real-time, privacy-preserving inference on-device. For instance, facial authentication requires local execution on edge devices to ensure low-latency responsiveness and to protect user privacy. CIM architectures are particularly well-suited to these scenarios due to their tightly integrated memory-compute structure, offering low-latency and energy-efficient inference capabilities.

We present PathoOpenGait, a cloud-based platform for comprehensive gait analysis. Gait assessment is crucial in neurodegenerative diseases such as Parkinson's and multiple system atrophy, yet current techniques are neither affordable nor efficient. PathoOpenGait utilizes 2D and 3D data from a binocular 3D camera for monitoring and analyzing gait parameters. Our algorithms, including a semi-supervised learning-boosted neural network model for turn time estimation and deterministic algorithms to estimate gait parameters, were rigorously validated on annotated gait records, demonstrating high precision and consistency. We further demonstrate PathoOpenGait's applicability in clinical settings by analyzing gait trials from Parkinson's patients and healthy controls. PathoOpenGait is the first open-source, cloud-based system for gait analysis, providing a user-friendly tool for continuous patient care and monitoring. It offers a cost-effective and accessible solution for both clinicians and patients, revolutionizing the field of gait assessment. PathoOpenGait is available at https://pathoopengait.cmdm.tw.

3D object detection using point clouds has been widely used for self-driving vehicles, roadside vehicle detection, and tracking. However, the detection accuracy using single-LiDAR-scanned point clouds suffers from occlusion to have accurate bounding boxes. Several previous works proposed compensating for the point cloud to improve 3D detection. This work designed a self-labeling method that exploits the characteristics of multiple LiDAR-scanned point clouds collected on roadside units. This labeling method allows the point completion model to be trained without human-labeled bounding boxes. The point completion model uses the single-LiDAR-scanned point cloud as input and uses the extracted object points from the fused point cloud as the training target. The voxel IoU and recall of the point-completed point clouds increased by 17% and 22% compared to that of the raw point clouds. Additionally, the detection performance of DBSCAN, i.e., 3D AP@IoU=0.25 increased by 20% after applying the point completion algorithm.

The enhanced computing platforms, which are on edge networks and used for embedding computing services, enable the processing of sensing data on edge and sharing the data with the peers of interest. Smart factories and autonomous vehicles are two typical examples. However, existing real-time communication protocols have limits on cross-network domains and complex data structures. To resolve the problems above, this work investigates a new distributed data exchange framework, Zenoh. As a unified and transparent middle-ware for devices over heterogeneous networks, Zenoh enables peers in different network domains to exchange messages in real-time by pub/sub data exchange pattern. Furthermore, it provides high flexibility and scalability by configurable end-to-end and hop-to-hop reliability and congestion control QoS profiles. This work proposes a bounded-time decision algorithm by Reliable Broadcast and CRDT based on Zenoh. This work also benchmarks the protocol’s capability over cross-domain networks and investigate the overhead of our algorithms implemented on Zenoh. The throughput and latency performance shows the potential for practical communication between robots in factories or autonomous vehicles on roads.

A low resolution infrared camera can use to collect the data of our daily activity, and protect privacy. However, pose detection on an extremely low resolution image will be a big challenge to deal with. Many researches show that Super-resolution may improve the accuracy of object detection on low resolution images. In this work, we detect the turning over while sleeping with low-resolution thermal sensors. To improve the accuracy of detection, we fuse the data from multiple sensors and use SRCNN to enhance the resolution. We also propose a method to reduce the noise caused by the residual body temperature on bed after turn over while detect the pose and turning over. The accuracy of pose detection is 65%, and the turning over detection has 50% recall rate and 83% precision.

Many virtual machines exist for sensor nodes with only a few kB RAM and tens- to a few hundred kB Flash memory. They pack an impressive set of features but suffer from a slowdown of one to two orders of magnitude compared to optimised native code, reducing throughput and increasing power consumption. Compiling bytecode to native code to improve performance has been studied extensively for larger devices, but the restricted resources on sensor nodes mean most modern techniques cannot be applied. Simply replacing bytecode instructions with predefined sequences of native instructions is known to improve performance but produces code several times larger than the optimised C equivalent, limiting the size of programmes that can fit onto a device. This article identifies the major sources of overhead resulting from this basic approach and presents optimisations to remove most of the remaining performance overhead, and over half the size overhead, reducing them to 67% and 77%, respectively. While this increases the size of the VM, the break-even point at which this fixed cost is compensated for by the smaller code it generates, is well within the range of memory available on a sensor device, allowing us to both improve performance and load more code on a device.

Optical images provide rich features but require extensive computation resources to process for SLAM. When there are limited computation resources on the robots, it becomes a heavy burden to process the images in real-time. This paper presents the design and implementation of key-frame selection algorithm for multiple robots simultaneous localization and tracking on the multi-robot soccer games which have pre-defined field and objects. Compared to traditional key-frame selection algorithms, this work makes use of the temporal and spatial relationship among objects on the pre-defined field to compute the information entropy. The selection ratio can be adjusted by two parameters: entropy threshold and the maximum moving distance. The experimental results show that the developed method can effectively detect the change of scene using selected key-frames. And comparing with the localization results using all the images, using less than 20% of all images after walking 11,203mm it only increase up to 0.87% trajectory errors.

In IoT systems, many sensing devices will periodically transmit data. However, most of the sensing data are either similar or redundant and the sensing device itself may not have effective rules standard to decide to send or not. Static rules maybe effective on specific scenario, and become ineffective when the use scenario of the sensor changes. Hence, we design an algorithm to solve the problem of data redundant for IoT devices. The algorithm iteratively separates a data region into smaller regions. In each round, it chooses a region with highest variability, and separates it into smaller regions. Finally, each region has different sizes and uses its average value to represent itself. If an area has more dynamical, diverse data, more data points will be used to present the data. In this paper, we present a method to reduce the file size of thermal sensor which can sense the temperature of a surface and outputs a two dimension gray scale image. In our evaluation result, we can reduce the file size to be 50% of JPEG format when 0:5% of distortion is allowed, and up to 93% less when 2% of distortion is allowed.

This paper presents CapeVM, a sensor node virtual machine aimed at delivering both high performance and a sandboxed execution environment that ensures malicious code cannot corrupt the VM's internal state or perform actions not allowed by the VM. CapeVM uses Ahead-of-Time compilation and introduces a range of optimisations to eliminate most of the overhead present in previous work on sensor node AOT compilers. A sandboxed execution environment is guaranteed by a set of checks. The structured nature of the VM's instruction set allows the VM to perform most checks at load time, reducing the need for expensive run-time checks compared to native code approaches.

Many IoT applications require public information and private information to make correct decision and trigger appropriate services in efficient manner. Many existing IoT services use cloud-based service model. Cloud-based service model conducts logic decision on remote servers, which store public and private information, and leads to security and privacy breakage. Moreover, it limits the users on controlling the services. The tradeoff between data privacy and intelligent service is an important issue on developing IoT services. In this paper, we present a public/private federation framework, which combines public and private intelligence to develop IoT applications. It allows the IoT applications to take advantage of the cloud-based public intelligence and local private data to provide correct services without transmitting private data to public servers. The framework is built on top of an intelligent virtual middleware, named WuKong, to federate with nodeRED, which is a web-enabled flow-based programming environment. Two case examples are shown in the paper to demonstrate how the private and public intelligence are inter-woven to compose intelligent IoT applications.

How to tolerate fault is a fundamental requirement to the designs of many cyber-physical systems. Devices or sensors might have different requirements on their levels of reliability and/or timely services in the composition of a cyber-physical system. In this work, a system framework is explored to virtualize devices/sensors and service migration is considered during run time, so that faults are masked and the timeliness in services is enhanced. In particular, a disaster messaging system supporting seamlessly service recovery in small and large scale network is developed and evaluated, where an acceptable level of connectivity in the face of numerous faults for responsive deliveries of information critical to the success of emergency response and rescue operations. The framework also takes into account the energy consumption and reliability of sensing services, while using different types of memory components. Last but not least, augmented sensing using smart phones allows users to receive the sensed information nearby; this is critical when communication infrastructures are damaged.

This work is motivated by the needs to synchronize task executions where tasks might use semaphores to protect their critical sections and run over accelerators. In particular, the Priority Ceiling Protocol is extended to manage priority inversion caused by accelerator usages. By recognizing the difference between an accelerator and a semaphore, higher-priority tasks are less likely blocked by lower-priority tasks, due to their requests for an accelerator. In particular, blocking that will not contribute to any deadlock and/or chained blocking is allowed in a managed way, with an objective to maximize the utilization of accelerators. A series of experiments is then conducted to derive insights to task synchronization when accelerators might be used.

The number and variety of applications for mobile devices continue to grow. However, the resources on mobile devices including computation and storage do not keep pace with the growth. Incorporating the computation capacity on cloud servers into mobile computing has been desired. However, there are numbers of challenging issues to realize it. In this work, we design and implement an elastic computation framework to take advantage the computation capacity on connected devices, to meet the computation demands of ever growing mobile applications. The computation framework enables resource-aware workload migration between mobile devices and connected computers. The framework was experimented and evaluated on cuckoos framework with Android applications, where communication and computation resources are taken into account. The experiments show that the application can be executed with dynamic allocated resources, either local or remote, to enhance its performance. With the elastic framework, the runtime environment can allocate local and remote resources to mobile applications with resource-aware policies.

How to effectively exchange information between parties in a disaster management system is one of the fundamental challenges to support timely and efficient disaster response and relief. Specifically, the timeliness, scalability, and availability are three desirable features for information exchange. We call the framework to support information exchange with the three features an Open Information Gateway (OIGY). In this paper, we present and experiment the mechanisms to recover communicaiton service during and after disasters for Open Information Gateway. The designed mechanisms adopt long distance radio to support tele-communication and computer networks, dynamic power adjustment for radio stations to maximize the radio recovery with minimal interference, and dynamic MESH routers to connect devices located in different networks. We experimented the proposed mechanims on a physical communication testbeds to measure the delay for service recovery and reliability of the proposed mechanism.

The number and variety of applications for mobile devices continue to grow. However, the resources on mobile devices including computation and storage do not keep pace with the growth. How to incorporate the computation capacity on cloud servers into mobile computing has been desired and challenge issues to resolve. In this work, we design an elastic computation framework to take advantage the heterogeneous computation capacity on cloud servers, which consist of CPUs and GPGPUs, to meet the computation demands of ever growing mobile applications. The computation framework extends OpenCL framework to link remote processors with local mobile applications. The framework is flexible in the sense that the computation can be stopped at any time and gains results, which is called imprecise computation in real-time computing literature. The framework has been evaluated against OpenCL benchmark and physical computation engine for gaming. The results show that the framework supports OpenCL benchmark, RODINIA, without modifying the codes with few exceptions. The elastic computation framework allows the cloud servers to support more mobile clients without sacrificing their QoS requirements. The experiment results also show that IO intensive applications do not perform well when the network capacity is insufficient or unreliable.

Simulation is a common approach for assisting system design and optimization. For system-wide optimization, energy and computational resources are often the two most critical limitations. Modeling energy-states of each hardware component and time spent in each state is needed for accurate energy and performance prediction. Tracking software execution in a realistic operating environment with properly modeled input/output is key to accurate prediction. However, the conventional approaches can have difficulties in practice. First, for a complex system such as an Android smartphone, building a cycle-accurate simulation environment is no easy task. Secondly, for I/O-intensive applications, a slow simulation would significantly alter the application behavior and change its performance profile. Thirdly, conventional software profiling tools generally do not work on simulators, which makes it difficult for performance analysis of complicated software, e.g., Java applications executed by the Dalvik virtual machine. Recently, virtual machine technologies are widely used to emulate a variety of computer systems. While virtual machines do not model the hardware components in the emulated system, we can ease the effort of building a simulation environment by leveraging the infrastructure of virtual machines and adding performance and power models. Moreover, multiple sets of the performance and energy models can be selectively used to verify if the speed of the simulated system impacts the software behavior. Finally, performance monitoring facilities can be integrated to work with profiling tools. We believe this approach should help overcome the aforementioned difficulties. We have prototyped a framework and our case studies showed that the information provided by our tools are useful for software optimization and system design for Android smartphones.

How to exchange information between parties in a mega-scale disaster management system is one of the fundamental challenges to support timely and efficient disaster response and relief. Specifically, the timeliness, scalability, and availability are three desirable features for information exchange. We call the framework to support information exchange with the three features an open information gateway, OIGY in short. In this paper, we present the challenges of information gateway, and the design of the communication protocols and the fundamental components and algorithms to support the aforementioned features. The efforts of this work will be divided into two major components: one is the distributed Truthful Real-time Information Publishing and Subscribing (TRIPS), and the other one is Heterogeneous And Plug-n-PlaY networks (HAPPY). The two components in OIGY collaborate to provide reliable and timely information publish and subscription service. TRIPS is responsible for logical information exchange management. Compared to modern real-time publish and subscription services, TRIPS is aimed at information responsiveness and distributed content-based filtering in an un-reliable network. To achieve better responsiveness, TRIPS will take advantage of the run-time service composition of SOA framework to select information sources. To enhance the success rate, TRIPS relies the information routing information provided by HAPPY. HAPPY will integrate heterogeneous communication networks including 3G/WiMAX telecommunication network and mesh mobile network into a coherent communication network and discovers the routes with probabilistic bandwidth guarantee.

Internet services are now the major service platforms to provide services for all sectors of business and industries. To assure the correctness for concurrent data access, transaction-based Internet service is a well-accepted computation model. Because of the proliferation of networked devices and globalization, internet services can barely undertake any service interruption during any time of the day. However, transaction-base service model does not guarantee service continuity and could lower the quality of service in many circumstances. In this paper, we describe the design and implementation for service continuity technology to provide seamless online service upgrade for transaction-based Internet services. In particular, we report the implementation details on enterprise testbed. The developed mechanism is verified against on-line telecommunication web-services. The experimental results show that the recovery overhead is negligible and the mechanism is practical for deployment.

The service providers introduce new telecom services, which are managed by different Operations Support Systems (OSSs), to retain existing and attract new customers. The working group, TeleManagement Forum (TM Forum), proposed a unified network infrastructure, Next Generation Operations Systems and Software (NGOSS), as the standard interface between different OSSs to reduce the overhead for integration of sophisticated operating environments, and to shorten the time for launching new telecom services. For most of service providers, existing legacy OSSs are generally not compatible to the NGOSS framework. Due to the fact that those legacy systems have been working for a long time, modification or replacement of all systems incurs extra overhead for implementation and management. This paper proposes a systematic approach Middleware Architecture (MA) for migrating the existing legacy systems to the NGOSS compliant systems. We use the OSS through Java Initiative (OSS/J) Service Provisioning as an example to show how MA exercises for provision of new telecom service. Through the proposed approach, service providers can perform the migration in an incremental manner.

Heterogeneous multi-core platforms are widely accepted for high performance multimedia embedded systems. Although pipeline techniques can enhance performance for the multi-core platforms, data dependencies for processing compressed multimedia data makes it difficult, if not impossible, to automate pipelined design. In this paper, we target on multimedia streaming applications on heterogeneous multi-core platforms and develop the "Tile Piecing Algorithm" for pipelined schedule synthesis within the targeted applications and platforms. The algorithm gives an efficient way to construct a pipelined schedule. The performance evaluation result shows that the algorithm performs as well as the optimal algorithm to utilize the computation resource. On the other hand, the algorithm only takes hundreds of milliseconds to complete, which is less than one tenth of the running time for optimal algorithm. Last, the synthesized schedule is well packed. The short execution time and schedule make-span makes the algorithm more practical to be used during the run-time.

Many modern embedded systems, such as cyber-physical systems, feature close integration of computation and physical components. Configurability, efficiency, adaptability, reliability, and usability are essential features for such systems. A workflow engine is a software application that manages workflows. It helps developers separate control flows from activities of the system, and thus is able to enhance configurability and development efficiency. This work aims at the design and implementation of a workflow engine for cyber-physical systems so as to configure workflows with less efforts. To meet timing requirements, the engine is designed to schedule activities to meet their timing requirements, and provides admission control so that these requirements of a set of workflows are guaranteed as long as it is admitted. A humanoid robot is used as the test-bed for the workflow engine. We model robot applications as workflows, and show that how the workflow engine provides real time guarantee and enhances its configurability.

Virtualization provides an excellent solution to resolve the portability, maintainability, development, and utilization problems in many system designs. In this paper, we are interested in energy-efficient designs for platform virtualization. In particular, we explore the computing resource mapping and the energy consumption relationship between virtual cores and physical cores when timing constraints in task executions are considered. Real-time and non-real-time task workloads are both considered in the study, where the computing needs of each virtual core are modeled with a computing server. A prototype with dynamic-voltage-scaling support is implemented based on a μ-kernel architecture. The capability and overheads of the proposed design was evaluated, for which we have encouraging results.

Multi-dimensional temporal data set is the common format in sensor network applications to store sampled temporal data. As time goes on, the size of the core tables in the data set may increase to enormous size and the tables become not managable. In order to reduce storage space and allow on-line query, how to trade off data compression effectiveness for on-line query performance is a challenge issue. In this paper, we are concerned with an effective framework for temporal data set that does not scarify on-line query performance and is specifically designed for very large sensor network database. The sampled data are compressed using several candidate approaches including dictionary-base compress and lossless vector quantization. In the mean time, on-line queries are conducted without decompressing the compressed data set so as to enhance the query performance. Experiments are conducted on a power meter database and sonoma database to evaluate the proposed methodologies in terms of data compression rate and data query speed. The results show that the compression rate ranges from 70% for numerical data to 20% for character data. In the mean time, the increased overhead for online query is limited up to 2%.

Internet services are now the major service platforms to provide services for all sectors of business and industries. To assure the correctness for concurrent data access, transaction-based Internet service is a well-accepted computation model. Because of the proliferation of networked devices and globalization, internet services can barely undertake any service interruption during any time of the day. Service interruption includes scheduled service upgrade, bug fixes, power outage, and service attacks. While considering the QoS (Quality of Service) for the transaction-based Internet service, service continuity is not negligible. However, transaction-base service model does not guarantee service continuity and could lessen the quality of service in many circum-stances. Although several service continuity technologies are developed, many of them focus on providing backup service when failure occurs and do not guarantee that the clients can continue their requested services step when the system is back to service. Existing technologies that support on-line service upgrade are usually too complicated to implement or have tremendous runtime overhead. In this paper, we describe the design, implementation and performance evaluation for service continuity technology to provide seamless online service upgrade for transaction-based Internet services. This approach requires minimal amount of code rewriting and has least overhead, comparing with existing approaches. We present the performance evaluation results and show that the overhead of our design is negligible.

This paper describes an embedded workflow framework (EMWF) that enables flexible personal and home automation and assistive devices and service and social robots (collectively referred to as SISARL) to be built on workflow architecture The process definition language supported by EMWF is called SISARL-XPDL. It consists of a subset of the WfMC standard XML Process Definition Language (XPDL) 2.0, together with elements that implement common mechanisms for robot behavior coordination. SISARL-XPDL definitions of workflows are first translated into standard XPDL and execution directives and then are parsed either directly into binary workflow scripts for execution or into intermediate scripts in C. EMWF provides workflow engines for Linux and Windows CE platforms. The engines are written in C in order to keep their memory footprint and runtime overhead small. Performance data show that the overheads introduced by the engine and workflow data are tolerable for most SISARL devices.

Executing functional components in pipeline on heterogeneous multi-core platforms can greatly improve the parallelism but require great amount of data communication among processes and threads. Our studies showed that existing inter-process/thread communication protocols consist of many unnecessary memory copies and prolong the execution of the applications on heterogeneous multi-core platforms. NTU ICPC uses polling-base mail notification to unnecessary context switches, and designs a memory subsystem to manage the input and output data between the senders and receivers. The protocol was implemented and evaluated on heterogeneous multi-core platform for several use scenario including H.264 encoding process. The evaluation results show that the communication overhead on sender side is independent of the data size and that on receiver side is greatly shortened, compared to several inter-process Communication (IPC) protocols including mailbox, message queue, and shared memory. When encoding H.264 video clips, the encoding frame rates increase for more than 30%. Our experiments also showed that the communication overhead accounts 40% to 50% of total execution time in average for H.264 video decoding applications. In this paper, we present the design and implementation of zero-buffer inter-core process communication protocol, named NTU ICPC, to shorten communication overhead for pipeline executed applications on heterogeneous multi-core platforms.

User-centered assistive and automation devices (UCAAD) must be flexible (i.e., configurable, customizable and adaptive). We are concerned with the interface design so as to design and build flexible UCAAD in an effective manner. We design the component interfaces for UCAAD to describe the essential information for component integration during design time and run-time. We propose to make use of the workflow model to integrate hardware and software components into a complete embedded systems. A unified component interface is required for the workflow-based integration process. In particular, we are concerned with resource management for UCAAD systems. This paper presents our approaches, progress to date and future plans.

This paper describes SISARL simulation environment. The environment is built on Microsoft Windows Workflow Foundation to support the design and development of smart devices and systems that aim to enhance life quality and well being of elderly individuals. To evaluate the usability, design, and implementation of a device, the developer provides as an input for each simulation, an operational view specification of the device and a model of the user (or models of multiple users). The specification describes the operations of the device, intended actions of the user(s) and user-machine interactions in terms activities and workflows. The device and the user are simulated when the specification is executed by the workflow engine. The simulation environment provides extensible libraries of reusable components, including device activities, user activities, and workflow definitions, to ease the efforts in writing operational view specifications of new and modified devices. It also provides a library of user models that can be used to simulate user behaviors, as well as tools for the capture, analysis and display of simulation data. In addition to these components of the simulation environment, the paper also describes ways to use simulation results in the design and development process.

This paper describes a component model and component-based architecture of smart devices and systems that are designed to enhance life quality and well being of elderly individuals. In addition to providing the traditional view of hardware, firmware and software components, the model also provides developers with an operational view. The view enables the developer to specify device-user interactions asexecutable workflows and allows the device operations and user actions to be experimented with and their correctness ascertained throughout the design and development process. The paper also presents a simulation environment for this purpose.

Dynamic voltage scaling (DVS) circuits have been widely adopted in many computing systems to provide tradeoff between performance and power consumption. The effective use of energy could not only extend operation duration for hand-held devices but also cut down power bills of server systems. Moreover, while many chip makers are releasing multi-core chips and multiprocessor system-on-a-chips (SoCs), multiprocessor platforms for different applications become even more popular. Multiprocessor platforms could improve the system performance and accommodate the growing demand of computing power and the variety of application functionality. This paper summarizes our work on several important issues in energy-efficient scheduling for real-time tasks in multiprocessor DVS systems. Distinct from most previous work based on heuristics, we aim at the provision of approximated solutions with worst-case guarantees. The proposed algorithms are evaluated by a series of experiments to provide insights in system designs.

In this demonstration, the core processing engines of two video applications, 3D video and intelligent video surveillance, are demonstrated. The developed algorithms and its VLSI design results are shown with hardware prototypes processing input video on-the-fly. In addition to the processing engine design, the development tools for efficiently designing these chips are also demonstrated.

Presents the introductory welcome message from the conference proceedings. May include the conference officers' congratulations to all involved with the conference event and publication of the proceedings record.

This work is motivated by the rapid increasing of the design complexity of many embedded systems. It aims at the proposing of solutions to resolve the hardware contention issues of non-preemptive processing elements shared among tasks and the cost optimization. A software solution based on the starting time management is proposed to interleave task executions on processing elements. Algorithms are proposed to determine the required processing elements of selected types, when there is no knowledge on the releasing time of any task: When task release orders are known a priori, an optimal algorithm is presented if processing elements have the same cost; otherwise, a pseudo-polynomial-time algorithm based on dynamic programming is presented for optimal solutions. The performance of the algorithms is also evaluated for general cases

Reconfigurable hardware provides multiple functions without greatly increasing the die size of SoCs. Field programmable gate array (FPGA), one type of reconfigurable hardware, is developing rapidly to handle high speed and complex applications. In particular, SRAM-based FPGA can be reconfigured during run-time to provide functionality as needed, thus reducing cost and power consumption. However, the reconfiguration delay time and limited resource on FPGA pose new challenges to many real-time scheduling applications. In order to optimize hardware usage and eliminate the impacts of reconfiguration delay, the scheduling algorithms and resource management mechanisms for reconfigurable SoC have to be re-designed. In this paper, we develop a template-based approach to reuse hardware resources without compromising the reconfiguration and performance constraint. The developed solution makes use of offline generated templates to generate near-optimal schedules during run-time.

We develop a data alignment mechanism for sensor networks. The mechanism allows the sensor data streams to be aligned without real-time clocks or virtual clocks. The developed mechanism makes use of the built-in counters on sensors and external synchronization signals to align data streams. The data server broadcasts out-of-channel synchronization signals with constant or variable intervals. The sensor data streams are aligned when received on the server. Only one way communication is used so as to reduce the communication overhead and clock synchronization overhead on sensors nodes. In addition, the developed mechanism is scalable thanks to one way communication. The synchronization error is bounded by the maximum sampling period of the sensors, and is independent of the number of the nodes in the network. Our analysis also shows the required awake time for sensors to tolerate different clock drifts and signal lost.

Medication errors harm at least 1.5 million people every year in United States, according to the report issued by Committee on Identifying and Preventing Medication Errors in 2007 [1]. Medication use system consists of many sub-systems including prescription entry systems and drug libraries. Integrating medication sub-systems and automating the medication process enhance medication compliance and prevent medication errors, caused by communication mistakes between care-takers and medical institutes. In this paper, we present the workflow-based integration framework for medication use. In the framework, new sub-systems including medicaiton authoring tools and intelligent medication dispensers are designed to automate the process. To integrate the complex and heterogeneous sub-systems, we develop an integration framework and make use of workflow engines. The integration framework defines how each subsystem describes itself and how the each sub-system triggers succeeding actions. Wrokflow-based approach is used to dispatch the works among the distributed sub-systems and provides the flexibility to the system.

Medication dispensers treated in this paper are designed to help improve compliance by users who live at homes and take medications over long periods of time. The paper first presents an overview of medication specifications that define constraints for dispensers and dispenser components that administer medications as specified. When given a specification and constraints defined by it, the dispenser scheduler checks for consistency and feasibility of constraints and schedules medications to meet the constraints. Several basic algorithms needed for these purposes are described and evaluated.

A smart pantry holds non-perishable household supplies and automates the purchasing and delivery of their replenishments. By relieving its user from the chore of keeping the home stocked of essentials, it provides convenience and peace of mind not only to elderly individuals but also busy people of all ages. This paper describes two alternative smart pantry designs and tradeoffs. Underlying methods and technologies used for their implementation are also described.

We describe here the prescription algebra and its implementation for medication authoring tools. The tools are parts of medication use process, which consists of prescription entry systems, medication authoring tool, medication scheduling specification, medication scheduler, and programmable pill dispenser. A medication authoring tool aids the pharmacists to collect and integrate prescriptions, to verify drug-drug interactions amongst prescriptions one took for the prescribed duration, and to generate the scheduling specifications for pill dispensers. We design a prescription algebra for medication authoring tool to correctly complete its work. We have implemented the platform-independent medication authoring tool using JAVA. The screen snapshot are shown in the paper.

This paper proposes novel techniques for scheduling radar dwells in phased array radar systems. In order to handle complex physical characteristics such as dwell interleaving, transmitting duty cycle constraint, and energy constraint, we propose a notion of schedulability envelope. The schedulability envelope designed offline hides the details of complex radar dwell scheduling and provides a simple measure for the schedulability check. Using the schedulability envelope, the proposed technique can efficiently perform the admission control for dynamic target tracking tasks. The simulation results show that the proposed approach can significantly improve the system utilization by taking advantage of dwell interleaving while guaranteeing the schedulability and physical constraints.

Energy-efficient scheduling is an effective way to balance the system performance and the energy consumption. We design a polynomial-time (1+ε)-approximation algorithm to minimize the energy consumption for periodic real-time tasks over such processors, where ε is the tolerable error given by users (1 ≥ ε > 0). It provides trade-offs between the user's tolerable error and the runtime complexity including the time complexity and the memory space complexity. System engineers could trade performance with implementation constraints.

Time-critical jobs in many real-time applications have multiple feasible intervals. Such a job is constrained to execute from start to completion in one of its feasible intervals. A job fails if the job remains incomplete at the end of the last feasible interval. This paper is concerned with how to find a schedule in which the number of jobs completed in one of their feasible intervals is maximized. We show that the maximization problem is NP-hard for both nonpreemptible and preemptible jobs. This paper develops two approximation algorithms for nonpreemptible and preemptible jobs. When jobs are nonpreemptible, algorithm LECF is with a 2-approximation factor; when jobs are preemptible, algorithm LEF is proved being a 3-approximation algorithm. We also show that our analysis on the two algorithms is tight by providing a set of input instances. Simulation results demonstrate that algorithms LECF and LEF not only guarantee the approximation factors but also outperform other multiple feasible interval scheduling algorithms.

The College of Electrical Engineering and Computer Science at the National Taiwan University has identified the area of content science for media-rich life, broadly construed, as one of core areas for the college's future directions. One major aspect of this project is to develop the enabling technology for reconfigurable platforms for multimedia applications. Specifically, the goal is to develop reconfigurable platforms of system-on-a-chip (SoC) components for the applications to support the needs of multi-modal multimedia contents and to provide rapid system prototyping. The faculties in the College of Electrical Engineering and Computer Science have formed a multi-discipline team to develop such technology. Our team includes seven faculties and more than thirty students from the college. This short report describes the reconfigurable platform for content science research activities currently underway by our team. Our current activities include to develop the technology to analyze the critical path for avoiding hardware contention, to minimize the use of logic components, to design the multimedia IPs, to optimally route the bus and place the logic units, to design energy efficient cache, to evaluate the performance and power consumption, to design the algorithm for temporal floor-planning/placement.

Timing constraints for radar tasks are usually specified in terms of the minimum and maximum temporal distance between successive radar dwells. We utilize the idea of feasible intervals for dealing with the temporal distance constraints. In order to increase the freedom that the scheduler can offer a high-level resource manager, we introduce a technique for nesting and interleaving dwells online while accounting for the energy constraint that radar systems need to satisfy. Further, in radar systems, the task set changes frequently and we advocate the use of finite horizon scheduling in order to avoid the pessimism that is inherent in schedulers that assume a task executes forever. We also develop the notion of modular schedule update which allows portions of a schedule to be altered without affecting the entire schedule, thereby simplifying the scheduler. Through extensive simulations, we validate our claims of providing greater scheduling flexibility without compromising on performance when compared with earlier work based on templates constructed offline.

Some real-time systems are designed to deliver services to objects that are controlled by external sources. Their services must be delivered on a timely basis, and the system fails when some services are delivered too late. Such a system may fail if the timing requirements, which it is designed to meet are erroneous. It may under-utilize resources and, consequently, be costly or unreliable if the requirements are too stringent. In general, the timing requirements of the system may change when the states of the objects monitored by the system change. Hence, one must identify how changes in object states call for changes in system requirements and how these changes should be incorporated in the design and implementation of the system. We first describe a methodology to determine timing requirements and to take into account of requirement changes at runtime. The method is based on several timing requirement determination schemes. Simulation data show that these schemes are effective for applications such as mobile IP hand-offs. We then discuss how to incorporate this methodology in the design of such systems and in the development process.

This paper proposes novel techniques for scheduling radar dwells in phased array radar systems. In order to handle complex physical characteristics such as dwell interleaving, transmitting duty cycle constraint, and energy constraint, we propose a notion of schedulability envelope. The schedulability envelope designed offline hides the details of complex radar dwell scheduling and provides a simple measure for the schedulability check. Using the schedulability envelope, the proposed technique can efficiently perform the admission control for dynamic target tracking tasks. The simulation results show that the proposed approach can significantly improve the system utilization by taking advantage of dwell interleaving while guaranteeing the schedulability and physical constraints.

This paper addresses the problem of scheduling real-time dwells in multi-function phase array radar systems. To keep track of targets, a radar system must meet its timing and energy constraints. We propose a new task model for radar dwells to accurately characterize their timing parameters. We develop an algorithm of transforming every dwell task as a semi-period task so the dwell task can meet its timing constraint and the interarrival times of the task will not be a constant. We also develop an enhanced template-based scheduling algorithm to schedule such tasks to meet the timing and energy constraints. Simulation results show that this algorithm can significantly improve the resource utilization.

This paper presents a new workload model, called the state-dependent deadline model, for applications whose high-level timing requirements may change with time. The problem is how to choose target deadlines for jobs from their time-varying deadlines so that the system can meet its requirements with a high probability. In the clairvoyant case, the deadline of every job is given by a known function of time for all times after the job is released. There is an optimal deadline determination algorithm. When the future values of job deadlines are unknown, it is not possible to make optimal choices of job deadlines. Several heuristics can be used to determine deadlines in this realistic case. Simulation results show that these heuristics allow the system to achieve good processor utilization and high probability of meeting its time-varying requirements.

Many application level qualities are functions of available computation resources. Recent studies have handled the computation resource allocation problem to maximize the overall application quality. However such QoS problem is fundamentally a multi-dimensional optimization problem that requires extensive computation. Therefore, online usage of optimization procedures may significantly reduce the computation resource available for applications. This raises the question of how to best use the optimization procedures for dynamic real-time task sets. This paper proposes a method called service classes configuration to address the QoS problem with dynamic arrival and departure of tasks. A simplified radar application is used as an illustrative example.

Although many query tree optimization strategies have been proposed in the literature, there still is a lack of a formal and complete representation of all possible permutations of query operations (i.e., execution plans) in a uniform manner. A graph-theoretic approach presented in the paper provides a sound mathematical basis for representing a query and searching for an execution plan. In this graph model, a node represents an operation and a directed edge between two nodes indicates the older of executing these two operations in an execution plan. Each node is associated with a weight and so is an edge. The weight is an expression containing optimization required parameters, such as relation size, tuple size, join selectivity factors. All possible execution plans are representable in this graph and each spanning tree of the graph becomes an execution plan. It is a general model which can be used in the optimizer of a DBMS for internal query representation. On the basis of this model, we devise an algorithm that finds a near optimal execution plan using only polynomial time. The algorithm is compared with a few other popular optimization methods. Experiments show that the proposed algorithm is superior to the others under most circumstances.

In this paper, we present a system that extracts and generalizes terms from Internet documents to represent classification knowledge of a given class hierarchy.We propose a measurement to evaluate the importance of a term with respect to a class in the class hierarchy, and denote it as support.With a given threshold, terms with high supports are sifted as keywords of a class, and terms with low supports are filtered out.To further enhance the recall of this approach, Mining Association Rules technique is applied to mine the association between terms.An inference model is composed of these association relations and the previously computed supports of the terms in the class.To increase the recall rate of the keyword selection process.we then present a polynomialtime inference algorithm to promote a term, strongly associated to a known keyword, to a keyword.According to our experiment results on the collected Internet documents from Yam search engine, we show that the proposed methods In the paper contribute to refine the classification knowledge and increase the recall of keyword selection. 'In the paper, we use "term" as word or phrase, which is extracted from the Internet documents, rather than use "keyword"."Keyword" is representative to the concept, while "term" is probably nonrepresentatlve.