In re Mouttet, No. 2011-1451 (Fed. Cir. Jun. 26, 2012) (Judges Prost, O’Malley, and Reyna)
Mouttet’s processor is built from two sets of intersecting conductive parallel wires. The two sets intersect at right angles, like a Tic-Tac-Toe board, and the crosspoints have a thin film that connects the two wires. It looks like this:
Voltages are sent through individual wires to alter the resistance of the film, which allows each crosspoint to be programmed to a high resistance state or a low resistance state. The different resistance states represent binary values (i.e., bits), which represent 1s and 0s.
At the Patent and Trademark Office (PTO), the examiner rejected all the claims in Mouttet’s application as obvious in light of a combination of 3 references: Falk, Das, and Terepin. Falk disclosed a similar device for performing arithmetic; that device had the same Tic-Tac-Toe grid arrangement, but the grid was made from criss-crossing paths of light rather than wires. Das disclosed a wire crossbar array, with molecular switches at the intersections, that is, the same scale as Mouttet’s. And Terepin disclosed an analog-to-digital converter. Combining these three references, the examiner rejected the claims as obvious over Falk, in view of Das and Terepin. In other words, Falk had the idea for a crossbar arrangement to perform arithmetic, and someone skilled in the art would recognize that substituting wires for light beams would yield Mouttet’s device. The Board of Patent Appeals and Interferences (BPAI) agreed and affirmed the rejection, and Mouttet appealed the rejection to the Federal Circuit.
The Federal Circuit also found Mouttet’s claims obvious. Mouttet argued that substituting electronic wires for optical paths would destroy Falk’s “principle of operation”—basically, the main purpose of the device. But the Federal Circuit said both Falk’s and Mouttet’s principles of operation were the same—programming intersecting crosspoints to perform arithmetic functions. So, the court said, it doesn’t matter if the invention uses wires or light beams; the main idea of the device is taught in the prior art.
As a secondary point, the Federal Circuit also agreed with the BPAI that Falk didn’t teach away from using wires. Falk simply explained the different types of technology and the perceived advantages of using and optical circuit. The court reminded Mouttet that simply because the prior art discusses a preferred embodiment, it doesn’t necessarily teach away from a different embodiment.
Do you think the Federal Circuit got this right? Let us know!