### Table 3: Minimum distance for square S-random inter- leavers.

2003

"... In PAGE 4: ... To reinforce this comparison, we have also evaluated the minimum distances of the five different code configurations using the exact algorithm in [10]. The results are reported in Table3 , and show that our technique yields the same minimum distance as the ad-hoc one, while the technique in [8] leads to a significant reduction. Also in terms of free distance, the interleavers generated with our technique seem also to outperform the ones described in the UMTS stan- dard.... ..."

Cited by 1

### Table 2: Correlation (r) of 1) the mean distance to the nearest optimal solution and 2) backbone robustness with log10(local search cost) for general JSPs.

"... In PAGE 10: ... Our next-descent algorithm evaluates N1 neighbors in a random order, selecting the first solution which improves on the makespan of the current solution; the algorithm terminates when no such improvements are possible. In Table2 , we report the correlation between the mean Hamming distance to the nearest global optimum and log10(local search cost). For backbone sizes of 0:1 through 0:5, the correlation is extremely high, and only moderately degrades at the larger backbone sizes.... In PAGE 10: ... Thus, we define backbone robustness for the JSP as the minimum percentage above the optimal makespan at which the backbone size is reduced by at least half (subject to the constraint that makespans must be integral). In the lower half of Table2 we report the correlation between the backbone robustness and log10(local search cost) for our general JSPs. The results are very similar to that reported by Singer et al.... ..."

### Table 2: Correlation (r) of 1) the mean distance to the nearest solution and 2) backbone robustness with log 10 (local search

2001

"... In PAGE 5: ... Our next-descent al- gorithm evaluates N1 neighbors in a random order, selecting the first improvement in makespan; the algorithm terminates when no such improvements are possible. In Table2 , we report the correlation between the mean Hamming distance to the nearest global optimum and log 10 (local search cost). For backbone sizes of #19 0:1 through #19 0:5, the correlation is extremely high, and only moderately degrades at the larger backbone sizes.... In PAGE 5: ... Thus, we define backbone robustness for the JSP as the minimum per- centage above the optimal makespan at which the backbone size is reduced by at least half (subject to integral makespan constraints). In the lower half of Table2 we report the correlation be- tween the backbone robustness and log 10 (local search cost) for our general JSPs. The results are very similar to that re- ported by Singer et al.... ..."

Cited by 3

### Table 2: Correlation (r) of 1) the mean distance to the nearest solution and 2) backbone robustness with log 10 (local search

2001

"... In PAGE 5: ... Our next-descent al- gorithm evaluates N1 neighbors in a random order, selecting the first improvement in makespan; the algorithm terminates when no such improvements are possible. In Table2 , we report the correlation between the mean Hamming distance to the nearest global optimum and log 10 (local search cost). For backbone sizes of #19 0:1 through #19 0:5, the correlation is extremely high, and only moderately degrades at the larger backbone sizes.... In PAGE 5: ... Thus, we define backbone robustness for the JSP as the minimum per- centage above the optimal makespan at which the backbone size is reduced by at least half (subject to integral makespan constraints). In the lower half of Table2 we report the correlation be- tween the backbone robustness and log 10 (local search cost) for our general JSPs. The results are very similar to that re- ported by Singer et al.... ..."

Cited by 3

### Table 2: Correlation (r) of 1) the mean distance to the nearest solution and 2) backbone robustness with log10(local search cost) for general JSPs.

2001

"... In PAGE 5: ... Our next-descent al- gorithm evaluates N1 neighbors in a random order, selecting the first improvement in makespan; the algorithm terminates when no such improvements are possible. In Table2 , we report the correlation between the mean Hamming distance to the nearest global optimum and log10(local search cost). For backbone sizes of 0:1 through 0:5, the correlation is extremely high, and only moderately degrades at the larger backbone sizes.... In PAGE 5: ... Thus, we define backbone robustness for the JSP as the minimum per- centage above the optimal makespan at which the backbone size is reduced by at least half (subject to integral makespan constraints). In the lower half of Table2 we report the correlation be- tween the backbone robustness and log10(local search cost) for our general JSPs. The results are very similar to that re- ported by Singer et al.... ..."

Cited by 3

### Table 2: Minimum distances and CPU times

2000

"... In PAGE 16: ...AM, 70MHz, 204.7 mips, 44.4 M ops. We ran both codes using 50 initial random points for each problem. The results are summarized in Table2 . This table lists the eigtheen problems with the number of variables and constraints and the statistic information related to the minimum distance between two points (minimum, maximum, average) and CPU time (minimum,maximum, average) using the Inexact{Restoration algorithm ( rst row of each set) and the ones using LANCELOT (second row).... In PAGE 16: ...roblem. The results are summarized in Table 2. This table lists the eigtheen problems with the number of variables and constraints and the statistic information related to the minimum distance between two points (minimum, maximum, average) and CPU time (minimum,maximum, average) using the Inexact{Restoration algorithm ( rst row of each set) and the ones using LANCELOT (second row). The information contained in Table2 is depicted graphically below. The intervals (min, max) of distances/log(CPU times) are represented by vertical segments, the averages are indicated with... ..."

Cited by 11

### Table 1. Comparison of results between grids with and without diagonals. New results

1994

"... In PAGE 2: ... For two-dimensional n n meshes without diagonals 1-1 problems have been studied for more than twenty years. The so far fastest solutions for 1-1 problems and for h-h problems with small h 9 are summarized in Table1 . In that table we also present our new results on grids with diagonals and compare them with those for grids without diagonals.... ..."

Cited by 11

### Table 4: Minimum L2 distances of random SLHDs and random LHDs

2000

Cited by 3

### Table 5: Minimum L2 distances of random SLHs and random LHs

### Table 4: Minimum distance for triangular S-random inter- leavers.

2003

Cited by 1