A new step in the context of increasingly efficient silicon-based computers seems to have been taken by researcher Avi Zadok of the Faculty of Engineering Sciences at Bar-Ilan University. One of the biggest problems with modern computers whose functions have been extended to include photonics is that optical signals, like electrical signals, move too fast.
Sometimes slower motion may be better, as explained in the press release: “Important signal processing activities, such as accurate selection of frequency channels, require data to be delayed at timescales of tens of nano-seconds. Given the high speed of light, optical waves spread over many meters at these time intervals. It is not possible to accept these path lengths in a silicon chip. It is not realistic. In this race, fasting does not necessarily win.”
One way of overcoming a similar problem is to use acoustic waves: the signal in question can be converted from the electrical domain into an acoustic wave. Because the speed of sound is lower, the new converted signal can have the necessary delay at tens of micrometers instead of meters. After propagation, the same delayed signal can then be converted back into an electrical signal.
This is what Zadok is trying to do with photonic computers, especially a system that combines light and sound in standard silicon, even if the same concept can be applied to any type of substrate, not just silicon.
Zadok himself adds in the press release: “Acoustics is a missing dimension in silicon chips because the acoustics can perform specific tasks that are difficult to perform only with electronics and optics. For the first time, we have added this dimension to the standard silicon photonics platform. The concept combines the communication and bandwidth of light with the selective processing of sound waves.”
This progress can be useful in applications related to 5G.