Power-aware Sensor Networks

As the trends towards decentralization, miniaturization, and longevity of deployment continue in many domains, power management has become increasingly important. In sensing and communications networks, power management has long been a part of the design paradigm. However, an underlying assumption in most of the existing work is that the performance of the sensing devices remain the same throughout their lifetime, which is not always true.  Thus, this project is focused on developing power-aware control strategies for maximizing the lifetime of wireless sensor networks.

Power-aware Scheduling Controller

Static networks(comprising of agents with no mobility)  are typically deployed for the purpose of monitoring critical areas for long periods of time, and comprise of a large number of low-cost, low-power devices with limited sensing, processing, and communication capabilities. Owing to the low quality of the constituent devices and the harshness of the environments in which they are deployed, the batteries of these devices start to deteriorate as a result of which their available power decay with time. However, in all the existing sensor scheduling schemes, there is an inherent assumption that the performance of sensing devices remain constant throughout the lifetime of a network and this assumption is not always true.

In this work, we present power-aware scheduling schemes for sensor networks that consist of sensing devices whose sensing range model is a function of transmitted power to maintain a desired event detection probability throughout the lifetime of the network. The lifetime of a network is the maximum time beyond which the desired performance cannot be guaranteed. The footprints of the sensors comprising these networks are dynamic in nature because of the fact that variations in available power have a direct impact on the performance of sensing devices. Therefore, we select the area of a sensor footprint as a performance metric and use explicit relationship between footprint area and available power to quantify the effects of variation in available power on the performance of sensing devices. 

To compensate for this variation in sensor performance because of change in available power, we propose power-aware scheduling schemes in which sensors use their available power to determine their performance metric at each decision time and then update their control parameter accordingly such that the desired event detection is maintained while consuming minimum power. The impact of variation in available power on sensor performance was a missing link in the existing literature, and is addressed in this work for the first time.


Evolution of the probability of a sensor being on for a desired detection probability (Pdes).


Event detection probability (Pd) vs time (t) for a network with decaying footprints. Constant blue plot corresponds to the case with the proposed scheduling scheme while green decaying plot corresponds to the case with no scheduling scheme. Red plot corresponds to Pdes.

Power-efficient Scheduling Scheme

This project is related with a more traditional aspect of power-awareness, i.e., efficient utilization of available energy resources to maximize system lifetime. Most of the existing schemes that are available in the literature are designed to maintain complete coverage throughout the lifetime of a network by ensuring that switching a particular sensor off does not deteriorate the coverage profile of a network. Maintaining complete coverage is important especially for time critical events that must be detected immediately. However, this complete coverage is typically achieved at the expense of considerable control and communication overheads, and these overheads make this objective over restrictive for applications that can tolerate some delay in the detection of an event. Thus, for certain applications, power consumption can be reduced by relaxing the desired performance criterion, which in this case is coverage. This tradeoff between power consumption and desired performance criterion is exploited and a probabilistic switching scheme is proposed that can ensure a required level of partial coverage throughout the lifetime of a network while minimizing the overhead involved in making switching decisions.

In this project, a probabilistic power-efficient sensor scheduling scheme is proposed that is based on the concept of a hard-core point process form stochastic geometry to minimize communication among neighboring sensors in making switching decisions. Hard-core point processes are inhibition processes that maintain certain minimum distance, called the inhibition distance, among the constituent points, and in this way limit the number of redundant sensors covering any area. The information that is communicated between sensors for coordination only consists of randomly generated numbers, which results in minimum communication overhead. To efficiently design and analyze this scheme, we developed an explicit relationship between the inhibition distance and detection probability through extensive Monte Carlo simulations and the developed model is shown to accurately achieve desire performance with average error of less than 1%. The proposed scheme can extend the lifetime of a sensor network from 40% – 70% as compared to random switching scheme.


Curve fitting on the simulated data for detection probability Pd vs inhibition distance d for different values of footprint areas A and deployment intensities.

Percentage decrease in the number of sensors and increase in the lifetime of the network under the proposed switching scheme as compared to random switching scheme.


  • Hassan Jaleel
  • Magnus Egerstedt



  1. H. Jaleel, A. Rahmani and M. Egerstedt, “Probabilistic lifetime maximization of sensor networks,” in IEEE Transactions on Automatic Control, vol. 58, iss. 2, pp. 534–539, Feb. 2013.
  2. H. Jaleel and M. Egerstedt, “Power-aware scheduling of wireless sensor networks with dynamic footprints,”  Submitted in IEEE Transaction on Control of Network Systems.


  1. H. Jaleel, A. Rahmani and M. Egerstedt, “Duty cycle scheduling in dynamic sensor networks for controlling event detection probabilities,” in Proc. of IEEE American Control Conference, pp. 3233–3238, July 2011.
  2. H. Jaleel and M. Egerstedt, “Sleep scheduling of wireless sensor networks using hard-core point processes,” in Proc. of IEEE American Control Conference, Washington D.C., pp. 788–793, June 2013.
  3. H. Jaleel and M. Egerstedt, “Distributed and adaptive power-aware scheduling of wireless sensor networks,” in IEEE Conference on Decision and Control, Florence, Italy, Dec. 2013. To appear.


Comments are closed.