Repeater Insertion for Carbon Nanotube Interconnects

Optimal repeater insertion techniques are usually employed in long interconnects to improve the performance. As a promising alternative interconnect material, carbon nanotube (CNT) has attracted much attention in past several years. However, it is worth noting that the implementation of CNT interconnects is hindered by large contact resistance, which would affect the number and size of repeaters. In this chapter, the repeater insertion technique is explored for CNT interconnects, with the contact effect treated appropriately. It is demonstrated that multi-walled CNT (MWCNT) is susceptible to the contact resistance, and special attention should be paid to the repeater insertion in MWCNT interconnects. Further, both the delay and power consumption are considered in the design of repeaters. The particle swarm optimization (PSO) algorithm is employed to capture the optimal number and size of repeaters, and the results are used to train an artificial neural network (ANN). The developed procedure can be applied to the optimal design of interconnect system in nano-CMOS and future nano-carbon-based ICs.