Best Known (96−21, 96, s)-Nets in Base 2
(96−21, 96, 152)-Net over F2 — Constructive and digital
Digital (75, 96, 152)-net over F2, using
- trace code for nets [i] based on digital (3, 24, 38)-net over F16, using
- net from sequence [i] based on digital (3, 37)-sequence over F16, using
- Niederreiter–Xing sequence construction II/III [i] based on function field F/F16 with g(F) = 3 and N(F) ≥ 38, using
- net from sequence [i] based on digital (3, 37)-sequence over F16, using
(96−21, 96, 257)-Net over F2 — Digital
Digital (75, 96, 257)-net over F2, using
- embedding of OOA with Gilbert–Varšamov bound [i] based on linear OOA(296, 257, F2, 2, 21) (dual of [(257, 2), 418, 22]-NRT-code), using
- discarding factors / shortening the dual code based on linear OOA(296, 267, F2, 2, 21) (dual of [(267, 2), 438, 22]-NRT-code), using
- 21 times duplication [i] based on linear OOA(295, 267, F2, 2, 21) (dual of [(267, 2), 439, 22]-NRT-code), using
- OOA 2-folding [i] based on linear OA(295, 534, F2, 21) (dual of [534, 439, 22]-code), using
- 1 times code embedding in larger space [i] based on linear OA(294, 533, F2, 21) (dual of [533, 439, 22]-code), using
- adding a parity check bit [i] based on linear OA(293, 532, F2, 20) (dual of [532, 439, 21]-code), using
- construction XX applied to C1 = C([509,16]), C2 = C([1,18]), C3 = C1 + C2 = C([1,16]), and C∩ = C1 ∩ C2 = C([509,18]) [i] based on
- linear OA(282, 511, F2, 19) (dual of [511, 429, 20]-code), using the primitive BCH-code C(I) with length 511 = 29−1, defining interval I = {−2,−1,…,16}, and designed minimum distance d ≥ |I|+1 = 20 [i]
- linear OA(281, 511, F2, 18) (dual of [511, 430, 19]-code), using the primitive narrow-sense BCH-code C(I) with length 511 = 29−1, defining interval I = [1,18], and designed minimum distance d ≥ |I|+1 = 19 [i]
- linear OA(291, 511, F2, 21) (dual of [511, 420, 22]-code), using the primitive BCH-code C(I) with length 511 = 29−1, defining interval I = {−2,−1,…,18}, and designed minimum distance d ≥ |I|+1 = 22 [i]
- linear OA(272, 511, F2, 16) (dual of [511, 439, 17]-code), using the primitive narrow-sense BCH-code C(I) with length 511 = 29−1, defining interval I = [1,16], and designed minimum distance d ≥ |I|+1 = 17 [i]
- linear OA(21, 11, F2, 1) (dual of [11, 10, 2]-code), using
- discarding factors / shortening the dual code based on linear OA(21, s, F2, 1) (dual of [s, s−1, 2]-code) with arbitrarily large s, using
- linear OA(21, 10, F2, 1) (dual of [10, 9, 2]-code), using
- discarding factors / shortening the dual code based on linear OA(21, s, F2, 1) (dual of [s, s−1, 2]-code) with arbitrarily large s (see above)
- construction XX applied to C1 = C([509,16]), C2 = C([1,18]), C3 = C1 + C2 = C([1,16]), and C∩ = C1 ∩ C2 = C([509,18]) [i] based on
- adding a parity check bit [i] based on linear OA(293, 532, F2, 20) (dual of [532, 439, 21]-code), using
- 1 times code embedding in larger space [i] based on linear OA(294, 533, F2, 21) (dual of [533, 439, 22]-code), using
- OOA 2-folding [i] based on linear OA(295, 534, F2, 21) (dual of [534, 439, 22]-code), using
- 21 times duplication [i] based on linear OOA(295, 267, F2, 2, 21) (dual of [(267, 2), 439, 22]-NRT-code), using
- discarding factors / shortening the dual code based on linear OOA(296, 267, F2, 2, 21) (dual of [(267, 2), 438, 22]-NRT-code), using
(96−21, 96, 3264)-Net in Base 2 — Upper bound on s
There is no (75, 96, 3265)-net in base 2, because
- 1 times m-reduction [i] would yield (75, 95, 3265)-net in base 2, but
- the generalized Rao bound for nets shows that 2m ≥ 39652 940597 057455 380818 497088 > 295 [i]