Efficient Flooding Protocols and Energy Models for Wireless Sensor Networks

Abstract: Wireless sensor networks are emerging from the mobile ad hoc network concept and as such they share many similarities. However, it is not the similarities that differentiates sensor networks from their ad hoc counterparts, it is the differences. One of the most important difference is that they should operate unattended for long periods of time. This is especially important since they usually rely on a finite energy source to function. To get this into a perspective, a sensor network constitutes of a sensor field where a number of sensor nodes are deployed. The sensor nodes relay the gathered information to a base station from which the data are forwarded either through a network or directly to the enduser. The communication between sensor nodes are conducted in an ad hoc manner, which means that paths toward the base station are dynamically constructed based on current network conditions. The network conditions changes and examples of this includes node failure, deactivated nodes, variations in the radio channel characteristics, etc.As mentioned above, the sensor nodes are energy constrained and one of the more important design criteria is the life time of a sensor node or network. To be able to evaluate this criteria an energy dissipation model is needed. Most of the energy dissipation models developed for wireless sensor networks are not based on the basic sensor node architecture and as such they where not accurate enough for our needs. Thus, an energy dissipation model was developed. This model utilises the basic sensor node architecture to obtain the operation states available and their corresponding state transitions.Communication is the most energy consuming task a sensor node can undertake. As such, the contributed energy dissipation model is used to evaluate this aspect of the proposed controlled flooding protocols. Generally, the controlled flooding protocols tries to minimise the number of forwarding nodes and by doing this they lower the energy consumed in the network. Along with this, the communication overhead of a protocol also needs to be taken into account. Our idea is to utilise the received signal strength directly to make forwarding decisions based on a cost function. This idea has a number of key features, which are: no additional overhead in the message, no neighbour knowledge and no location information are needed. The results from the proposed flooding protocols are promising as they have a lower number of forwarding nodes and a longer lifetime than theothers.