### Table 1: Average latitude of non-consecutive five-year age groups [8]

### Table 2: Measured values of con gurations 1-3 The direct-connected multiprocessor systems (conf. 2) reveal a lower communication bandwidth and a slightly higher latency. We have measured an unidirectional communication bandwidth of only 27 MB/s and a latency of 2.58 s for the dual Pentium Pro systems (Tab. 2). Until up to now, only a consecutive shared-memory access takes place. If the memory is accessed in a non-consecutive

1997

Cited by 4

### Table 3. Data rates with 1-4 timeslots for transparent service.

1999

"... In PAGE 7: ...non-consecutive timeslots) is dependent on the timeslot allocation procedure. The data rates of the different coding schemes are shown in Table3 .... ..."

### TABLE 1. Sequencing primers and approximate locations in the 16S rRNA sequence of E. coli

1995

### Table 3: Values of the entropy function for the FBW(i) approximation sequence.

2002

Cited by 6

### Table 1: Best Combined LFSRs with their v;10

"... In PAGE 6: ... In Table 2, we give the estimated variance reduc- tion factors (with respect to MC) obtained by the randomly-scrambled LFSR point sets (as in Section 3.4) given in Table1 . The parameters of the option are S(0) = 100, r = ln 1:09, = 0:2 and T = 120.... In PAGE 7: ... This generator turned out to be quite bad for the asian option problem, giving sometimes esti- mators with more variance than MC. The one from Table1 de nitely gives better estimators than the ME one and this shows that looking at projections over non-consecutive indices is important for this type of application. The results obtained in this example are quite promising given the simplicity of the method and the fact that it is faster than MC.... ..."

### Table 1: Best Combined LFSRs with their v;10

"... In PAGE 6: ... In Table 2, we give the estimated variance reduc- tion factors (with respect to MC) obtained by the randomly-scrambled LFSR point sets (as in Section 3.4) given in Table1 . The parameters of the option are S(0) = 100, r = ln 1:09, = 0:2 and T = 120.... In PAGE 7: ... This generator turned out to be quite bad for the asian option problem, giving sometimes esti- mators with more variance than MC. The one from Table1 de nitely gives better estimators than the ME one and this shows that looking at projections over non-consecutive indices is important for this type of application. The results obtained in this example are quite promising given the simplicity of the method and the fact that it is faster than MC.... ..."

### Table 7: Approximated (E(

1997

"... In PAGE 14: ...IfweletJ n be a random variable given by the number of jumps in the maximum order complexity pro le of a sequence of length n,then E(J n )= J n 2 n = n;2 X m=0 S m (2 ;(m+1) ; 2 ;n ): If J k n is the random variable of the numberofjumpsofsizek in the maximum order complexity pro le of a sequence of length n then E(J k n )= J k n 2 n = n;k;1 X m=0 S m 2 m+k+1 : We can use our approximation to calculate approximations ^ J n and ^ J k n to J n and J k n respectively.Table 6 gives the expected number of jumps of size k in a sequence of length 24 as calculated by our approximation as well as the true value, and Table7 gives the expected number of jumps in a sequence of length n (1 n 24) as calculated by our approximation as well as the true value. 6 Conclusions In this paper wehave derived an approximate distribution for the maximum order complexity of random binary sequences, and wehave used this approxi- mation to showhow to construct statistical tests to identify keystreams that can be simulated by short feedback shift registers.... ..."

Cited by 2