Enhancing video quality from multiple cameras
The ‘Cross-layer optimization for visual sensor networks’ (Clovisen) project is working to optimise resource allocation and thus enhance the quality of wireless visual sensor networks that are based on direct sequence code division multiple access (DS-CDMA). The researchers used the open systems interconnection (OSI) model with seven layers, originally designed for minimal communication between the layers such that each could be optimised individually. However, this design can cause interference among transmissions from different nodes as well as competition for network resources during simultaneous transmission. Recent research has suggested that joint optimisation of layers can significantly improve performance. The investigators on the Clovisen project set out to determine appropriate resource allocation and thus cross-layer optimisation. First, the researchers developed a computer simulation of a DS-CDMA wireless visual sensor network that minimises the spreading codes used to distinguish individual sensor node transmissions. They then created a cross-layer optimisation algorithm based on minimising either average video distortion or the maximum acceptable video distortion from the nodes. Finally, they implemented a second algorithm based on the Nash bargaining solution from game theory to solve the same resource allocation problem. Thus, the investigators to date have created a simulation tool enabling several important contributions to cross-layer optimisation of a wireless video sensor network, promising enhanced quality of data retrieved via more efficient resource allocation. Continuation of the project should enable thorough evaluation of system performance and perhaps new innovations along the same lines.
