@article {477, title = {Automated Model-based Optimization of Data-Adaptable Embedded Systems}, journal = {ACM Transactions on Embedded Computing Systems}, volume = {19}, year = {2019}, month = {02/2020}, pages = {22 pages}, abstract = {

This paper presents a modeling and optimization framework that enables developers to model an application{\textquoteright}s data sources, tasks, and exchanged data tokens; specify application requirements through high-level design metrics and fuzzy logic based optimization rules; and define an estimation framework to automatically optimize the application at runtime. We demonstrate the modeling and optimization process via an example application for video-based vehicle tracking and collision avoidance. We analyze the benefits of runtime optimization by comparing the performance of static point solutions to dynamic solutions over five distinct execution scenarios, showing improvements of up to 74\% for dynamic over static configurations.

}, doi = {10.1145/3372142}, url = {https://doi.org/10.1145/3372142}, author = {Adrian Lizarraga and Jonathan Sprinkle and Roman Lysecky} } @conference {460, title = {WiP Abstract: String stability of commercial adaptive cruise control vehicles}, booktitle = {International Conference on Cyber-Physical Systems}, year = {2019}, abstract = {In this work, we conduct a series of car-following experiments with seven different ACC vehicles and use the collected data to model the car-following behavior of each vehicle. Using a linear stability analysis, the string stability of each tested vehicle is analyzed. Addition- ally, platoon experiments with platoons of up to eight identical vehicles are conducted to validate the stability findings. Previously, only one commercial ACC system has been evaluated for string stability. }, keywords = {Adaptive Cruise Control, String Stability}, doi = {10.1145/3302509.3313325}, url = {https://dl.acm.org/citation.cfm?id=3313325}, author = {George Gunter and Y. Yang and Raphael E Stern and Daniel B. Work and Maria Laura Dell Monache and Rahul Bhadani and Matt Bunting and Roman Lysecky and Jonathan Sprinkle and Benjamin Seibold and Benedetto Piccoli} } @article {Sandoval:2017:TTS:3092956.3047498, title = {Task Transition Scheduling for Data-Adaptable Systems}, journal = {ACM Transactions on Embedded Computing Systems (TECS)}, volume = {16}, year = {2017}, pages = {105:1{\textendash}105:28}, abstract = {

Data-adaptable embedded systems operate on a variety of data streams, which requires a large degree of configurability and adaptability to support runtime changes in data stream inputs. Data-adaptable reconfigurable embedded systems, when decomposed into a series of tasks, enable a flexible runtime implementation in which a system can transition the execution of certain tasks between hardware and software while simultaneously continuing to process data during the transition. Efficient runtime scheduling of task transitions is needed to optimize system throughput and latency of the reconfiguration and transition periods. In this article, we provide an overview of a runtime framework enabling the efficient transition of tasks between software and hardware in response to changes in system inputs. We further present and analyze several runtime transition scheduling algorithms and highlight the latency and throughput tradeoffs for two data-adaptable systems. To evaluate the task transition selection algorithms, a case study was performed on an adaptable JPEG2000 implementation as well as three other synchronous dataflow systems characterized by transition latency and communication load.

}, keywords = {Data adaptability, hardware/software codesign, model-based design, runtime transition scheduling}, issn = {1539-9087}, doi = {10.1145/3047498}, url = {http://doi.acm.org/10.1145/3047498}, author = {Nathan Sandoval and Casey Mackin and Sean Whitsitt and Gopinath, Vijay Shankar and Sachidanand Mahadevan and Milakovich, Andrew and Merry, Kyle and Jonathan Sprinkle and Roman Lysecky} } @proceedings {sprinkle285, title = {Model-based Fuzzy Logic Classifier Synthesis for Optimization of Data-Adaptable Embedded Systems}, year = {2016}, pages = {293-302}, address = {Hartford, CT}, doi = {10.1109/COMPSAC.2016.156}, author = {Adrian Lizarraga and Roman Lysecky and Jonathan Sprinkle} } @conference {465, title = {Model-Driven Optimization of Data-Adaptable Embedded Systems}, booktitle = {2016 IEEE 40th Annual Computer Software and Applications Conference (COMPSAC)}, year = {2016}, publisher = {IEEE}, organization = {IEEE}, author = {Adrian Lizarraga and Roman Lysecky and Jonathan Sprinkle} } @conference {sprinkle272, title = {Model-driven Optimization of Data-Adaptable Embedded Systems}, booktitle = {COMPSAC}, year = {2016}, pages = {293-302}, publisher = {IEEE}, organization = {IEEE}, abstract = {

Complex sensing and decision applications such as object tracking and classification, video surveillance, unmanned aerial vehicle flight decisions, and others operate on vast data streams with dynamic characteristics. As the availability and quality of the sensed data changes, the underlying models and decision algorithms should continually adapt in order to meet desired high-level requirements. Due to the complexity of such dynamic data-driven systems, traditional design time techniques are often incapable of producing a solution that remains optimal in the face of dynamically changing data, algorithms, and even availability of computational resources. To assist developers of these systems, we present a modeling and optimization methodology that enables developers to capture application task flows and data sources, define associated quality metrics with data types, specify each algorithm{\textquoteright}s data and quality requirements, and define a data quality estimation framework to optimize the application at runtime. We demonstrate each facet of the modeling and optimization process via a video-based vehicle tracking and collision avoidance application, and show how such an approach results in efficient design space exploration when selecting the optimal set of algorithm modalities. When searching for an application configuration within 1\% to 5\% of optimal, our model-guided approach can achieve speedups of up to 9.3X versus a standard genetic algorithm and speedups of up to 80X relative to a brute force algorithm.

}, doi = {10.1109/COMPSAC.2016.156}, url = {http://dx.doi.org/10.1109/COMPSAC.2016.156}, author = {Adrian Lizarraga and Roman Lysecky and Jonathan Sprinkle} } @conference {sprinkle249, title = {Generating Model Transformations for Mending Dynamic Constraint Violations in Cyber Physical Systems}, booktitle = {The 14th Workshop on Domain-Specific Modeling}, year = {2014}, month = {2014}, pages = {35-40}, abstract = {

Cyber physical systems by definition involve design constraints addressing the computation and communication necessary to control physical systems. These systems have been modeled using domain specific modeling languages, but some limitations exist in the continued application of such a modeling approach to more complex, or safety-critical, systems. Specifically, it is well known how to formulate constraints in a domain-specific modeling language in order to prevent users from building invalid structures, but existing constraint-based techniques do not take into consideration design requirements that may require analysis in the physical domain (i.e. dynamic constraints). Those analysis results, when interpreted by a domain expert, can inform changes to the model: unfortunately, this process does not scale. This paper presents an approach to integrating dynamic constraints that cannot be enforced using structural model constraints. The technique uses expert blocks to analyze systems and generates model transformations specific to the system using the results of those analyses to fix constraint violations. The paper describes a Dynamic Constraint Feedback (DCF) methodology for integrating this technique into existing systems from a generic perspective. Specific examples in this paper are derived from the domain of data adaptable reconfigurable embedded systems (DARES).

}, doi = {10.1145/2688447.2688454}, url = {http://dx.doi.org/10.1145/2688447.2688454}, author = {Sean Whitsitt and Jonathan Sprinkle and Roman Lysecky} } @conference {240, title = {Efficient Reconfiguration Methods to Enable Rapid Deployment of Runtime Reconfigurable Systems}, booktitle = {Asilomar Conference on Signals, Systems and Computers}, year = {2013}, month = {11/2013}, publisher = {IEEE}, organization = {IEEE}, address = {Pacific Grove, CA}, doi = {10.1109/ACSSC.2013.6810401}, url = {http://dx.doi.org/10.1109/ACSSC.2013.6810401}, author = {Roman Lysecky and Nathan Sandoval and Sean Whitsitt and Casey Mackin and Jonathan Sprinkle} } @conference {233, title = {How You Can Learn to Stop Worrying and Love Reconfigurable Embedded Systems: A Tutorial}, booktitle = {Engineering of Computer Based Systems (ECBS), 2013 20th IEEE International Conference and Workshops on the}, year = {2013}, month = {April}, pages = {213-214}, keywords = {C++ language, C/C++ code, codesign, Computers, Conferences, data streams, embedded hardware, embedded systems, Hardware, hardware tasks, hardware-software codesign, image processing algorithms, JPEG2000 standards, middleware, middleware framework, modeling infrastructure, reconfigurable embedded systems, runtime behaviors, software tasks, software tool, software tools, Transform coding, Tutorials}, doi = {10.1109/ECBS.2013.27}, author = {Nathan Sandoval and Casey Mackin and Roman Lysecky and Jonathan Sprinkle} } @conference {235, title = {Model Based Development with the Skeleton Design Method}, booktitle = {20th IEEE International Conference and Workshops on the Engineering of Computer Based Systems}, year = {2013}, pages = {12-19}, doi = {10.1109/ECBS.2013.16}, url = {http://dx.doi.org/10.1109/ECBS.2013.16}, author = {Sean Whitsitt and Jonathan Sprinkle and Roman Lysecky} } @conference {241, title = {Runtime Hardware/Software Task Transition Scheduling for Runtime-Adaptable Embedded Systems}, booktitle = {International Conference on Field-Programmable Technology (ICFPT)}, year = {2013}, pages = {342-345}, doi = {10.1109/FPT.2013.6718382}, url = {http://dx.doi.org/10.1109/FPT.2013.6718382}, author = {Nathan Sandoval and Casey Mackin and Sean Whitsitt and Roman Lysecky and Jonathan Sprinkle} } @conference {232, title = {System Throughput Optimization and Runtime Communication Middleware Supporting Dynamic Software-Hardware Task Migration in Data Adaptable Embedded Systems}, booktitle = {Engineering of Computer Based Systems (ECBS), 2013 20th IEEE International Conference and Workshops on the}, year = {2013}, month = {April}, pages = {59-68}, keywords = {combinatorial explosion, Data adaptability, data adaptable design methodology, data adaptable embedded systems, data configurations, data handling, Data models, data profile correlation, design time optimization, dynamic software-hardware task migration, embedded systems, Field programmable gate arrays, FIFO queues, Firing, Hardware, hardware accelerators, hardware-software codesign, hardware-software communication wrapper, hardware/software codesign, hardware/software communication middleware, heuristic programming, heuristic search methodology, middleware, model-based design, Pareto optimal configurations, Pareto optimisation, Runtime, runtime communication middleware, search problems, simulation-based methodology, system throughput optimization}, doi = {10.1109/ECBS.2013.25}, author = {Nathan Sandoval and Casey Mackin and Sean Whitsitt and Roman Lysecky and Jonathan Sprinkle} } @conference { c:milakovic2012ecbs, title = {Automated Software Generation and Hardware Coprocessor Synthesis for Data-Adaptable Reconfigurable Systems}, booktitle = {Engineering of Computer Based Systems (ECBS), 2012 IEEE 19th International Conference and Workshops on}, year = {2012}, pages = {15-23}, abstract = {

We present an overview of a data-adaptable reconfigurable embedded systems design methodology. The paper presents a novel paradigm for hardware/software code sign and reconfigurable computing driven by data-adaptability. The data-adaptable approach allows designers to directly model the data configurability of the target application, thereby enabling a solution that permits dynamic reconfiguration based on the data profile of the incoming data stream. This approach permits low-power, small form-factor hardware implementations of algorithms that might otherwise consume significant resources, or perhaps exceed the available space of the reconfigurable hardware.

}, keywords = {Data adaptability, hardware/software codesign, model-based design}, doi = {10.1109/ECBS.2012.16}, url = {http://dx.doi.org/10.1109/ECBS.2012.16}, author = {Milakovich, Andrew and Vijay Gopinath and Roman Lysecky and Jonathan Sprinkle} } @conference {239, title = {An Overseer Control Methodology for Data Adaptable Embedded Systems}, booktitle = {International Workshop on Multi-Paradigm Modeling (MPM)}, year = {2012}, month = {08/2012}, pages = {1-6}, doi = {10.1145/2508443.2508448}, url = {http://dx.doi.org/10.1145/2508443.2508448}, author = {Sean Whitsitt and Jonathan Sprinkle and Roman Lysecky} } @conference { c:sachi-ecbs-2011, title = {Hardware/Software Communication Middleware for Data Adaptable Embedded Systems}, booktitle = {Proceedings of the 18th IEEE International Conference and Workshops on Engineering of Computer-Based Systems}, year = {2011}, pages = {34-43}, publisher = {IEEE Computer Society Press}, organization = {IEEE Computer Society Press}, abstract = {

Recent trends toward increased flexibility and configurability in emerging applications present demanding challenges for implementing systems that incorporate such capabilities. The resulting application configuration space is generally much larger than any one hardware implementation can support. We provide an overview of a new data-adaptive approach to the rapid design and implementation of such highly configurable applications. In support of this data-adaptable approach, we present and detail an efficient and flexible hardware/software communication middleware to support the seamless communication between hardware and software tasks at runtime. We highlight the flexibility of this interface and present an initial case study and results demonstrating the performance capabilities and area requirements.

}, doi = {10.1109/ECBS.2011.12}, url = {http://dx.doi.org/10.1109/ECBS.2011.12}, author = {Sachidanand Mahadevan and Vijay Gopinath and Roman Lysecky and Jonathan Sprinkle and Jerzy Rozenblit and Michael Marcellin} } @conference { w:vijay-ecbs-mbd-2011, title = {Modeling of Data Adaptable Reconfigurable Embedded Systems}, booktitle = {Proceedings of the 8th IEEE Workshop on Model-Based Development for Computer-Based Systems}, year = {2011}, month = {April}, pages = {276-285}, abstract = {

Many applications require high flexibility, high configurability and high processing speeds. The physical constraints of a highly flexible system{\textquoteright}s hardware implementation preclude a hardware solution that satisfies all configuration options. Similarly for pure software implementations, even if configurability is satisfied, process efficiency will be sacrificed. Thus for applications of any significant size, there can be no single hardware or software configuration that can efficiently support all the configurability options of the applications. The Data-Adaptable Reconfigurable Embedded System (DARES) approach tackles this problem through combination of the hardware-software co-design and reconfigurable computing methodologies. Data-adaptability means that as data streams change, the system is reconfigured along the baselines defined within the system{\textquoteright}s specifications. In this project we use the concepts of Model-Integrated Computing to implement a domain-specific modeling language for the DARES approach. The language captures all the configurability options of the application task(s), performs design-space exploration, and provides a template for source code generation.

}, doi = {10.1109/ECBS.2011.31}, url = {http://dx.doi.org/10.1109/ECBS.2011.31}, author = {Vijay Gopinath and Jonathan Sprinkle and Roman Lysecky} }