Practical packet combining for use with cooperative and non-cooperative ARQ schemes in wireless sensor networks
O'Rourke, Damien (2009) Practical packet combining for use with cooperative and non-cooperative ARQ schemes in wireless sensor networks. PhD thesis, Dublin City University.
Full text available as:
Although it is envisaged that advances in technology will follow a "Moores Law" trend for many years to come, one of the aims of Wireless Sensor Networks (WSNs) is to reduce the size of the nodes as much as possible. The issue of limited resources on current devices may therefore not improve much with future designs as a result. There is a pressing need, therefore, for simple, efficient protocols and algorithms that can maximise the use of available resources in an energy efficient manner.
In this thesis an improved packet combining scheme useful on low power, resource-constrained sensor networks is developed. The algorithm is applicable in areas where
currently only more complex combining approaches are used. These include cooperative communications and hybrid-ARQ schemes which have been shown to be of major benefit for wireless communications. Using the packet combining scheme developed in this thesis more than an 85% reduction in energy costs are possible over previous, similar approaches. Both simulated and practical experiments are developed in which the algorithm is shown to offer up to approximately 2.5 dB reduction in the required Signal-to-Noise ratio (SNR) for a particular Packet Error Rate (PER). This is a welcome result as complex schemes, such as maximal-ratio combining, are not implementable on many of the resource constrained devices under consideration.
A motivational side study on the transitional region is also carried out in this thesis. This region has been shown to be somewhat of a problem for WSNs. It is characterised
by variable packet reception rate caused by a combination of fading and manufacturing variances in the radio receivers. Experiments are carried out to determine whether
or not a spread-spectrum architecture has any effect on the size of this region, as has been suggested in previous work. It is shown that, for the particular setup tested, the
transitional region still has significant extent even when employing a spread-spectrum architecture. This result further motivates the need for the packet combining scheme
developed as it is precisely in zones such as the transitional region that packet combining will be of most benefit.
Archive Staff Only: edit this record