Publications - Pervasive Computing Laboratory

Access control anotacionsemantica authentication Cloud computing coap compromise cryptography cynamon emadrid emrisco home network I-Shaper identity management inrisco inteligenciafuentesabiertas Internet of Things intrusion detection system IoT magos middleware Pervasive computing privacy Protocols Qursa raudo2 Security Smart Card Trust management Ubisec VANETs 
Show all
2019
 Simon-Sherratt, Robert;  Janko, Balazs;  Hui, Terence;  S.-Harwin, William;  Dey, Nilanjan;  Díaz-Sánchez, Daniel;  Wang, Jin;  Shi, Fuqian
Task Scheduling to Constrain Peak Current Consumption in Wearable Healthcare Sensors Journal Article 
In: Electronics, vol. 8, pp. 789, 2019, ISSN: 2079-9292.
Abstract | Links | BibTeX | Tags: embedded, health care, joint digital transformation, low-power, task scheduler, wearable
@article{diazsanchez011,

title = {Task Scheduling to Constrain Peak Current Consumption in Wearable Healthcare Sensors},

author = {Robert Simon-Sherratt and Balazs Janko and Terence Hui and William S.-Harwin and Nilanjan Dey and Daniel Díaz-Sánchez and Jin Wang and Fuqian Shi},

url = {https://doi.org/10.3390/electronics8070789

https://www.mdpi.com/2079-9292/8/7/789

https://www.mdpi.com/2079-9292/8/7/789/pdf?version=1563172692},

doi = {https://doi.org/10.3390/electronics8070789},

issn = {2079-9292},

year  = {2019},

date = {2019-07-15},

urldate = {2019-07-15},

journal = {Electronics},

volume = {8},

pages = {789},

abstract = {Small embedded systems, in our case wearable healthcare devices, have significant engineering challenges to reduce their power consumption for longer battery life, while at the same time supporting ever-increasing processing requirements for more intelligent applications. Research has primarily focused on achieving lower power operation through hardware designs and intelligent methods of scheduling software tasks, all with the objective of minimizing the overall consumed electrical power. However, such an approach inevitably creates points in time where software tasks and peripherals coincide to draw large peaks of electrical current, creating short-term electrical stress for the battery and power regulators, and adding to electromagnetic interference emissions. This position paper proposes that the power profile of an embedded device using a real-time operating system (RTOS) will significantly benefit if the task scheduler is modified to be informed of the electrical current profile required for each task. This enables the task scheduler to schedule tasks that require large amounts of current to be spread over time, thus constraining the peak current that the system will draw. We propose a solution to inform the task scheduler of a tasks’ power profile, and we discuss our application scenario, which clearly benefited from the proposal.},

keywords = {embedded, health care, joint digital transformation, low-power, task scheduler, wearable},

pubstate = {published},

tppubtype = {article}

}

Close
Small embedded systems, in our case wearable healthcare devices, have significant engineering challenges to reduce their power consumption for longer battery life, while at the same time supporting ever-increasing processing requirements for more intelligent applications. Research has primarily focused on achieving lower power operation through hardware designs and intelligent methods of scheduling software tasks, all with the objective of minimizing the overall consumed electrical power. However, such an approach inevitably creates points in time where software tasks and peripherals coincide to draw large peaks of electrical current, creating short-term electrical stress for the battery and power regulators, and adding to electromagnetic interference emissions. This position paper proposes that the power profile of an embedded device using a real-time operating system (RTOS) will significantly benefit if the task scheduler is modified to be informed of the electrical current profile required for each task. This enables the task scheduler to schedule tasks that require large amounts of current to be spread over time, thus constraining the peak current that the system will draw. We propose a solution to inform the task scheduler of a tasks’ power profile, and we discuss our application scenario, which clearly benefited from the proposal.
Close
https://doi.org/10.3390/electronics8070789
https://www.mdpi.com/2079-9292/8/7/789
https://www.mdpi.com/2079-9292/8/7/789/pdf?version=1563172692
doi:https://doi.org/10.3390/electronics8070789
Close