Last month I discussed the FPGA Trap (here), which is the ‘high-power and high-cost FPGA solution for open RAN RU signal processing’ trap, baited by the lure of “programmability” — i.e., not locked into a solution that may need upgrades later. In that post I pointed out competing solutions that are in fact programmable, but don’t rely on a classic FPGA programmable logic fabric.

Barely two weeks after that post, came a fairly shocking announcement from Analog Devices. (Here.)

Analog Devices, Inc. (Nasdaq: ADI) today announced an ASIC-based radio platform for O-RAN compliant 5G radio units that is designed to shorten time to market and meet the evolving needs of 5G networks…

ADI’s radio platform includes all the core functionality needed in an O-RAN compliant 5G radio unit, including a baseband ASIC, software defined transceivers, signal processing, and power…

Bringing an ASIC low PHY baseband to market is a big step forward for Open RAN.

It is indeed.

Analog claims the ASIC will implement a radio baseband (low-PHY and fronthaul I/F) compatible with the ORAN Alliance specification and fronthaul protocol for a 7.2x functional split.

Analog’s Low PHY Baseband

• Complete low PHY baseband to O-RAN compliant 7.2 split
• 7.2x O-RAN compliant Cat A
• 8T8R, 2x4T4R support, 4T4R, multiband FDD/TDD
• LTE/5G/NB-IoT support
• (i)FFT, PRACH, eCPRI
• C/U/S/M plane processing

As Joe Biden might say, this is a BFD.

Why? I will count the ways.

1. Low power compared with an FPGA. Further, when used with a corresponding ZIF (zero-IF) RFIC, which uses an analog mixer, the DACs and ADCs will sample at a much lower rate than the DRF (direct-sampled RF) architecture implemented by a Xilinx RFSoC, further reducing power consumption.
2. No software development needed (assuming this solution handles the m-plane messages for configuring radio functions).
3. (This is implied, but not explicitly stated.) No need to negotiate Standard-essential Patents from OEMs like Nokia and Samsung.

If these assumptions hold up, and if Analog’s timeline and pricing are reasonable, ORAN-compliant 4G/5G macro radios will take a giant step toward commoditization, and service providers will find a much more competitive supplier landscape available to them.

Side note: If Analog really means to support 8T8R, I suspect they will have to release a version of this chip that supports Category B, i.e. precoding and beam-forming in the chip along with modulation compression/decompression.

I for one will be keeping a close eye on this.