Error protection for wireless imaging: Providing a trade-off between performance and complexity

In this paper, we propose a number of error protection schemes for wireless imaging ranging from simple but less efficient equal error protection to very complex yet optimal error protection. The main objective is to facilitate trade-offs between performance and complexity in choosing an error protection scheme. Our technique provides the system designer with a number of solutions from which the application can choose those that best suits the available resources such as processing power and data rate. For this purpose, an image codestream is split into smaller cells each having a certain number of packets while the available parity symbols from the deployed error protection are optimally allocated to these cells. Subsequently, the parity symbols allocated to these cells are distributed among the packets that constitute each of these cell. Finally, the error control codes for every packet are determined based on the number of parity symbols allocated to each packet. Larger cells provide less complex but also less efficient error protection while smaller cells result in a better error protection performance but at the cost of increased complexity. Simulation results validate the effectiveness of the considered technique in providing the desired performances-complexity trade-off. To obtain better correlation with human perception, performance is evaluated in terms of objective perceptual quality metrics.