An Empirical Square-Root Law for the Degree Discrepancy of Binary m-Sequences
Source: arxiv.org/abs/2611.15635
Let $k\ge 2$, $N=2^k$, $M=N-1$ and let $Δ(r)$ be the range, over all starting offsets, of the total-variation degree discrepancy of a length-$r$ window of a primitive binary linear feedback shift register (LFSR) of period $M$. Degree-discrepancy profiles for every $k\le 28$ support the scaling $\max_r Δ(r)=Θ(\sqrt{N})=Θ(2^{k/2})$. The tabulated peaks are maxima on a discrete grid, hence lower bounds on the exhaustive maximum. The normalized peaks $d_k=\max_{r\in R_k}Δ(r)/2^{k/2}$ oscillate between values near 1/2 and values that approach, but do not exceed, 1. The data suggest $\liminf…
Comments and reviews
No comments yet.
API
curl /api/v1/records/arxiv/2611.15635
curl -X POST /api/v1/records/arxiv/2611.15635/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2611.15635/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"body": "…", "review": true}'