Heuristic doubling period trick by preprocessing pitch correlation values?

[YangWangGit]

I think the trick in this surrounding "for" code block is to suppress insignificant, possible doubling period. Is there any more in-depth motivation, explanation, or paper mentioning this? Thanks.

LPCNet/src/lpcnet_enc.c

Line 597 in 2159b4f

if (st->xc[2+sub][i] < xc_half*1.1) st->xc[2+sub][i] *= .8;