This article describes the development of a software architectural framework for implementing compute-aware control systems, where the term “compute-aware” describes controllers that can modify existing low-level computing platform power managers in response to the needs of the physical system controller. This level of interaction means that high-level decisions can be made as to when to operate the computing platform in a power-savings mode or a high-performance mode in response to situation awareness of the physical system. The framework is demonstrated experimentally on a mobile robot platform. In this example, a situation-aware governor is developed that adjusts the speed of the processor based on the physical performance of the robot as it traverses a path through obstacles. The results show that the situation-aware governor results in overall power savings of up to 38.9 percent with 1.3 percent degradation in performance compared to the static high-power strategy.

This paper describes a sophomore-level ECE course that bridges the gap between the maker movement and typical engineering design courses. This course incorporates the practical skills of fabrication, assembly, and hardware integration found in makerspaces but deepens the perspective from one of getting it to work to designing it to work well and efficiently. This paper discusses mastery-based education in practical skills, an open-ended student-defined design project, and a logistics model that incorporates support staff as partners in the course. The relationship between this course and an ECE-themed makerspace is also discussed.

This paper introduces a software policy for memory management in heterogeneous memory systems in order to improve the trade-offs between performance and power consumption, while attempting to make the best use of different characteristics of the underlying memory technologies. In this policy, the operating system and the application co-schedule page management in order to make informed decisions about page allocation and migration. Software-Controlled 2- Level Memory (Soft2LM) is a hardware-agnostic approach for efficient usage of heterogeneous memory that allows for region-based allocations, migrations, and application advice. We include analysis of the access characteristics of the PARSEC 3.0 suite of benchmarks, both as motivation for software-guided intelligent page placement as well as for understanding where application advice can make a difference. Our evaluation running PARSEC on our Soft2LM Linux kernel shows an average 4.7x improvement over comparable RAMDISK- based swapping. Even when migrating 100 and 1000 pages per second between regions, Soft2LM performs comparably to a stock system with twice the available DRAM, showing a 1.8\% and 0.7\% improvement, respectively.

There are numerous ways to control power usage of modern systems. Dynamic voltage and frequency scaling (DVFS) and C- and P-states are commonly available in hardware, and operating systems have fine-grained control of these states. Unfortunately, most DVFS systems simply consider CPU utilization when determining which P-state to enter. Running average power limit (RAPL), introduced by Intel in their Sandy Bridge line of processors, allows researchers and system designer to obtain detailed estimates of energy consumption by the core, uncore and DRAM. In this paper we use Linux's perf_events subsystem to measure a number of commonly cited performance metrics while running the SPEC CPU2006 benchmarks in addition to calculating average power using RAPL registers. This paper shows initial promising results correlating DRAM memory consumption to stall-cycle ratio.

The development of a feedback-based method for evolving the blended class format for large, multiple section courses is presented with the application to two circuits courses taught at Georgia Tech: ECE 2040 is a circuits course for ECE majors and ECE 3710 is a circuits and electronics course for non-majors. These are both multi-section, high-enrollment courses, with a combined enrollment of over 600 students per term. The courses utilize two MOOCs that provide all of the online content. ECE 3710 was first offered in a flipped mode in the Summer of 2013 and has evolved over five semesters based on feedback from the student surveys, focus groups, instructor feedback, discussion forum questions, and performance on tests. ECE2040 has been offered two semesters in the blended mode using a modified version of the blended model developed for ECE3710.

This paper discusses a training program that we are developing and are using for instructors who teach a circuits and electronics course in a blended format. The course has over 450 students enrolled in nine different sections each semester and has been taught in a blended format for six semesters. A key motivation for teaching the course in this format is to improve consistency across sections, which necessitates the need for training among the instructors on how to teach blended courses.

An application is said to be soft real-time if it is able to tolerate occasional violations of its timing constraints. The overall CPU utilization of a soft real-time system can be improved by exploiting the relaxed nature of its timing constraints. One possible approach is by adapting CPU-budgets in reservation-based schedulers. Reservation-based schedulers allow parts of the system and groups of tasks to be isolated from one another. By making the reservations adaptive, resources such as CPU can be allocated to tasks based on current usage rather than worst-case usage, which can help to improve the overall utilization of CPU time committed to tasks. This paper presents an adaptive budget allocation algorithm where the allocated budget is adapted at reservation-period boundaries based on predictions of future CPU usage. This approach differs from previous algorithms where adaptations are performed on job completion. Simulation results show that adaptations at reservation-period boundaries allow for a faster response time. Results from experiments performed on a prototype demonstrate the robustness and effectiveness of the proposed system.

Pedagogical practices in electrical engineering education have been shifting away from teacher-centered learning during the past decade. An innovation that has enabled the adoption of inquiry-based and problem-based learning into the curriculum using experimentation coupled with simulation and analysis has been the development of portable oscilloscopes and other instruments that rely on tablet or laptop computers to perform some of the data processing and to act as the display. Faculty members at six institutions of higher learning have incorporated hands-on experimental activities into existing courses and/or developed new courses that take advantage of these new tools. Assessment data collected by these faculty members have demonstrated that the change towards student-centered learning facilitated by portable electronics increased student interest in electrical engineering, built student confidence in their ability to design circuits and systems, and supported the development of a deeper understanding of the theories that the students investigate or apply in the hands-on activities. A summary of the challenges that are faced in the different implementation models and a discussion of best practices are presented.

This paper presents a general fault-tolerant event detection scheme that allows nodes to detect erroneous local decisions by leveraging the local decisions reported by their neighbors. This detection scheme can handle cases where nodes have different accuracy levels. The derived fault-tolerant estimator is proven to be optimal under the maximum a posteriori (MAP) criterion. An equivalent weighted voting scheme is also derived. Further, two new error models are derived to take into account the neighbor distance and the geographical distributions of the two decision quorums. These models are particularly suitable for detection applications where the event under consideration is highly localized. The fault-tolerant estimator is simulated using a network of 1,024 nodes deployed randomly in a square region and assigned random probabilities of failure. Several estimation schemes that allow nodes to learn their error rates continuously are developed. These error rates are used in the distributed estimation schemes to assign appropriate weights to the nodes in the voting scheme.

This article explores methods to reconfigure infinite impulse response (IIR) filters in processes that utilize computer resource management. A high-performance mode uses a full-length IIR filter, while low resources or a desire to conserve power triggers a low-performance mode where a smaller length filter is used. Two alternatives for algorithms are explored: switching between filter banks of varying lengths, and anytime filters where the later stages can be skipped if computational resources are scarce. Two main challenges are addressed in this article. The first is how to configure an IIR filter as an anytime algorithm, and the second is how to initialize the states of the filters in order to mitigate transients induced by reconfiguration. The performance and feasibility of several common transient management schemes are investigated for use in this application. It is found that an anytime IIR filter outperforms the filter banks in almost every case studied due to its inherently smaller transients and the fact that the reconfiguration happens at the speed of resource management and not at a lower speed typically associated with reconfiguration due to physical system properties.

We present a sensor network simulation environment that allows users to evaluate the effects of different architectural choices and strategies on the lifetime and performance of a sensor network. Our tool can also be used to evaluate new approaches (routing protocols, cooperation algorithms), and compare them with the ones already in place. This simulator is an extension of the Georgia Tech Network Simulator (GTNetS) and leverages its design choices for maximum performance. It incorporates models for the different functional units composing a sensor node and characterizes the energy consumption of each. It also has a model for a network base station and its interactions with the rest of the network. We report our results from a set of tests that demonstrate some of the capabilities of our simulator. In particular, we were able to simulate sensor networks of several hundred thousand nodes while using less than 2 GB of memory, which exceeds the capabilities of existing simulators by an order of magnitude. The simulator is also used to study the performance of an algorithm that assigns globally unique IDs to nodes in a sensor network.

A sensor network consists of a set of battery-powered nodes, which collaborate to perform sensing tasks in a given environment. It may contain one or more base stations to collect sensed data and possibly relay it to a central processing and storage system. These networks are characterized by scarcity of resources, in particular the available energy. We present a distributed algorithm to solve the unique ID assignment problem. The proposed solution starts by assigning long unique IDs and organizing nodes in a tree structure. This tree structure is used to compute the size of the network. Then, unique IDs are assigned using the minimum number of bytes. Globally unique IDs are useful in providing many network functions, e.g. configuration, monitoring of individual nodes, and various security mechanisms. Theoretical and simulation analysis of the proposed solution have been preformed. The results demonstrate that a high percentage of nodes (more than 99%) are assigned globally unique IDs at the termination of the algorithm when the algorithm parameters are set properly. Furthermore, the algorithm terminates in a relatively short time that scales well with the network size. For example, the algorithm terminates in about 5 minutes for a network of 1,000 nodes

A partial differential equation (PDE)-based feature-level image fusion approach is proposed for multisensory image segmentation. The energy functional of the proposed fusion model is a weighted sum of several functionals, each constructed based on the characteristics of the sensor image. The weight selection decides the way that the model handles redundant, conflicting, or complementary information involved in the multisensory data. The method is implemented using level sets and is fast enough for real-time segmentation tasks. Finally the algorithm is applied to the segmentation of X-ray and visual images, and the results show that the fusion algorithm is efficient, accurate, and robust.

A partial differential equation (PDE)-based feature-level image fusion approach is proposed for multisensory image segmentation. The energy functional of the proposed fusion model is a weighted sum of several functionals, each constructed based on the characteristics of the sensor image. The weight selection decides the way that the model handles redundant, conflicting, or complementary information involved in the multisensory data. The method is implemented using level sets and is fast enough for real-time segmentation tasks. Finally the algorithm is applied to the segmentation of x-ray and visual images, and the results show that the fusion algorithm is efficient, accurate, and robust.