Adaptive low-power analog/digital converters for wireless sensor networks

Abstract

The paper addresses the problem of power consumption of sensor nodes in a wireless network. An integrated low-power analog/digital converter (ADC) is presented that is particularly suited for wireless sensor applications. The converter makes use of information theoretic redundancy in the input signal for reducing the conversion workload and performing data compression on-the-fly during conversion (entropy-coding A/D converter). Thus, energy is saved both in signal conversion and transmission. Experimental results from a prototype chip are presented. The converter is especially suitable for sensor networks that maintain a global data model. This is further illustrated on an exemplary scenario of distributed wave detection. It is shown that sensor signal detection and acquisition in this type of applications can be carried out very efficiently with entropy-coding converters used in the sensor nodes.