On the processing architecture in wireless video sensor networks: node and network level performance evaluation

The architecture of a wireless video sensor network node has an impact on the performance of a wireless sensor network. Due to nature of WVSN applications, the node should be capable of potentially performing image processing algorithms while simultaneously has high networking capabilities. Providing an optimum architecture in which both parameters are addressed is the main challenge in WVSN. This paper reviews suggested architectures for the node and based on that three base architectures are proposed and implemented. The performance of the nodes is then evaluated at node and network level and the results are included in the paper. Based on the results, an optimized architecture is introduced. This architecture separates the node to a network and an application part. This brings the flexibility of implementing the optimized solution for each part, without considering the other. The network part is implemented using a microcontroller. It is capable of adopting operating systems that are widely used in Wireless Sensor Networks. On the other hand, the application part is implemented using FPGA, that provides a power-optimized and stand-alone solution for the node with high computation capabilities.