Abstract-For portable applications, long battery lifetime is the ultimate design goal. Therefore, the availability of battery and voltage converter models providing accurate estimates of battery lifetime is key for system-level low-power design frameworks. In this paper, we introduce a discrete-time model for the complete power supply subsystem that closely approximates the behavior of its circuit-level continuous-time counterpart. The model is abstract and efficient enough to enable event-driven simulation of digital systems described at a very high level of abstraction and that includes, among their components, also the power supply. The model gives the designer the possibility of estimating battery lifetime during system-level design exploration, as shown by the results we have collected on meaningful case studies. In addition, it is flexible and it can thus be employed for different battery chemistries.
DESIGNING COMPLEX, low-energy embedded systems requires balanced optimization of the energy needed in computation, communication, and storage. Computation is often performed by core processors (either general-purpose or application-specific) running embedded software. This choice is dictated by the inherent flexibility of processor cores-whose computational energy can be retargeted by simply modifying the program they execute-and by the modularity of processor-based architectures that facilitates design reuse. Cores are designed to be reused in many different systems. Hence, they are carefully handcrafted to achieve minimum energy demand for the required computational capabilities. In contrast, communication and storage resources are more application-specific. Memory size is tailored to application requirements (e.g., program size and communication buffer size). Similarly, memory organization (e.g., caching and interleaving) is often driven by performance, energy, and cost (area) constraints. Communication resources, such as point-to-point or shared buses, connect core processors to application-specific memories and peripherals. Consequently, they are inherently application-specific as well.Hand-tuning application-specific components for every design can be excessively timeconsuming and violates tight time-to-market constraints. Thus, energy optimization of memory and communication resources should be pursued with the help of design methodologies that can easily originate automatic synthesis tools. The objective of memory energy optimization methodologies for embedded systems is to create low-energy memory architectures customized for a specific core and a specific system functionality (i.e., an embedded application). This is in sharp contrast with generalpurpose systems, in which the main objectives of memory architecture design are robustness and flexibility.
Embedded Systems
74This article presents a methodology for automatic memory hierarchy generation that exploits memory access locality of embedded software.The methodology is successfully applied to the design of an MP3 decoder.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.