5G Infrastructure: Need for ASICs

In my previous blog on ASICs, I talked about the diverse types of them along with the Design Flow and Technological Characteristics of them in which a brief idea was discussed about them.

Now, moving on with the needs of them in the modern era. It’s highly required in the upsurging demand for the need of building 5G infrastructure. Speaking of which, the technology will require three times as many base stations to deliver the same coverage as LTE, will require three times as much power as LTE, and will cost four times as much as LTE. 5G networking infrastructure market revenue will almost double in 2020 to reach $8.1 billion, according to the latest forecast by Gartner.

Taking into account, the cost & power requirements the traditional approach to cellular infrastructure, we need something more than just FPGAs and DSPs. ASICs for AI and autonomous vehicles have all made recent headlines in the national press, with announcements from GAFAM. Using the “more than Moore” processes allow ASICs to provide cost-effective processes that balances the power-consumption and die size which makes it possible to include features such as RF or MEMS sensors.

ASICs optimize the number of transistors, clock cycles, production costs, and power consumption versus FPGAs/DSPs, with ASICs enabling the same performance in the soft-logic design as an FPGA.

Conclusion

5G creates several challenges in terms of power, cost, and range, thus inducing a shift for the cellular infrastructure away from FPGAs/DSPs used in 3G/4G systems and back to ASICs, which are better suited. The big companies like Tesla, Facebook and Google have all made announcement for multi-billion-dollar ASIC developments. But these are for AI and autonomous-driving equipment, where the most advanced technologies are essential. 5G equipment doesn’t need the same bleeding-edge technologies. As a result, costs can be lowered significantly using an ASIC approach.

Thanks for reading.