Efficiency issues in the implementation of finite difference time domain codes

Implementation techniques which increase the performance of time domain finite-difference codes for the solution of sparse time-domain electromagnetic problems are considered. As the basis of this analysis. Maxwell´s equations are expressed as a system of conservation laws. Practical issues such as computational efficiency and memory requirements are discussed for the implementation of the finite-difference schemes. Advanced programming techniques in the C-language are used to implement the finite-difference schemes discussed. The example of the penetration of electromagnetic energy through a shield with a thick-gap is used to check the efficiency of the methods.<>