System-Level Synthesis of Ultra Low-Power Wireless Sensor Network Node Controllers: A Complete Design-Flow

Abstract : Wireless Sensor Networks (WSN) is a new and challenging research field for embedded system design automation. Engineering a WSN node platform is a tough challenge, as the design must enforce many severe constraints among which energy consumption is often the most critical one due to the small size of a node and its difficult access after deployment. WSN nodes have until now been designed using commercial low-power microcontrollers (MCUs). These MCUs are not well-suited for WSN node design as they are based on a general purpose compute engine and consume too much power w.r.t. WSN node's power budget. In this thesis, we propose a complete system-level designflow for an alternative approach based on the concept of power-gated hardware micro-tasks. In this approach, WSN node architecture is made of several micro-tasks that are activated on an event-driven basis, each of them being dedicated to a specific task of the system (such as event-sensing, MAC, routing, etc.). These hardware micro-tasks are controlled by a hardware scheduler (called the System Monitor) that is automatically generated from a system-level description of the WSN node task graph in the form of a textual Domain Specific Language (DSL). By combining hardware specialization with power-gating, we can drastically reduce both dynamic and static energy of a WSN node controller. The results show that dynamic power savings by one to two orders of magnitude are possible w.r.t. the software implementations based on MCUs such as the MSP430. Similarly, static power savings of one order of magnitude are also obtained due to the reduction in data memory size and power-gating.
Document type :
Complete list of metadatas
Contributor : Muhammad Adeel Ahmed Pasha <>
Submitted on : Thursday, January 6, 2011 - 3:55:37 PM
Last modification on : Wednesday, December 18, 2019 - 4:56:16 PM
Long-term archiving on: Thursday, April 7, 2011 - 3:07:55 AM


  • HAL Id : tel-00553143, version 1


Muhammad Adeel Ahmed Pasha. System-Level Synthesis of Ultra Low-Power Wireless Sensor Network Node Controllers: A Complete Design-Flow. Computer Science [cs]. Université Rennes 1, 2010. English. ⟨tel-00553143⟩



Record views


Files downloads