1st International IEEE/ACM Workshop on Software for Sensor Networks

Research Article

Evaluation of power duty cycle and latency in a sensor mesh with random activity

  • @INPROCEEDINGS{10.1109/COMSWA.2006.1665222,
        author={C. Gonzalez and  J.  Sanz-Maudes},
        title={Evaluation of power duty cycle and latency in a sensor mesh with random activity},
        proceedings={1st International IEEE/ACM Workshop on Software for Sensor Networks},
        publisher={IEEE},
        proceedings_a={SENSORWARE},
        year={2006},
        month={8},
        keywords={},
        doi={10.1109/COMSWA.2006.1665222}
    }
    
  • C. Gonzalez
    J. Sanz-Maudes
    Year: 2006
    Evaluation of power duty cycle and latency in a sensor mesh with random activity
    SENSORWARE
    IEEE
    DOI: 10.1109/COMSWA.2006.1665222
C. Gonzalez1, J. Sanz-Maudes1
  • 1: Dept. Tecnologia Electron., Univ. Politecnica de Madrid

Abstract

Energy saving at the nodes of a wireless sensor network (WSN) is a primary goal as the lifetime of these autonomous sensors is highly dependent on the battery lifetime when they are battery operated. Direct solutions to spare battery use are both limiting the radio range, using a hop by hop data transmission mechanism, and turning off the radio following an activity pattern, as the radio greatly contributes to energy consumption. However, if the sensor radio is kept in this "inactive" state for a long period, latency of data propagation significantly increases. We have evaluated the ratio of latency versus energy consumption assuming a simple communication algorithm in which, in each period, each individual sensor selects a slot to listen (turns on its radio) in a random manner. If we assume that the battery lifetime is directly proportional to the duty cycle of activity, the lifetime of a WSN devoted to a task, such as surveillance, with low probability of event occurrence, is proportionally increased by the inverse of the duty cycle at the price of a greater latency. The evaluation has been done using both a simulation program and an experimental set-up with several CC2420DB from Chipcon acting as sensor nodes and the results are presented