"... In PAGE 12: ...able 1. The identified groups and their functionality categories ........... 21 Table2 .... In PAGE 34: ...______________________________________________________________ _____________________________________________________________________ 22 Access messages Descrambling messages Other messages 5.2 Subscription The subscription category contains functions that allow applications to retrieve information regarding the subscription and SO data stored in the used smartcard (see Table2 in appendix). The information can be used for presenting (for the user of the receiver) an overview of subscribed services.... ..."

"... In PAGE 6: ... Classifying the reason a potential dependence is unanalyzable by the techniques discussed in this paper is useful in determining where additional e ort may prove useful. As can be seen from Table4 , the reasons for which a subscript is unanalyz- able can be divided into three similarly sized categories.... ..."

"... In PAGE 14: ... Note that the spatial approach requires calculating four inverse DCTs for the blocks Xi, i = 1; ::; 4 before performing inverse motion compensation. A naive approach in the spatial domain (referred to as spatial(2) in Table1 ) will not make any sparseness assumptions on the contents of Xi, i = 1; ::; 4. If sparseness of Xi, i = 1; ::; 4 is exploited during the inverse DCT calculations for the spatial-domain approach (referred to as spatial(1) in Table 1) the computation complexity relative to the naive approach is substantially reduced.... In PAGE 14: ... A naive approach in the spatial domain (referred to as spatial(2) in Table 1) will not make any sparseness assumptions on the contents of Xi, i = 1; ::; 4. If sparseness of Xi, i = 1; ::; 4 is exploited during the inverse DCT calculations for the spatial-domain approach (referred to as spatial(1) in Table1 ) the computation complexity relative to the naive approach is substantially reduced. In all but the DC only case, the DCT domain approach for inverse motion-compensation has lower complexity compared with the spatial domain approaches.... In PAGE 15: ... Since the image quality resulting from the 3 ? 2 ? 1 case is better than the DC + 2AC case and the computation savings using DC + 2AC is not signi cant, the 3 ? 2 ? 1 case may be preferred for most applications. The results of Table1 are depicted in Fig. 11.... ..."

"... In PAGE 50: ...or more information on Brocade Service Plans, visit www.brocade.com. Table3 -1 details the current Brocade recommended support levels for large fabrics. If a certain level of Fabric OS does not appear in Table 3-1, it does not mean that version will not work or is not supported.... In PAGE 50: ...or more information on Brocade Service Plans, visit www.brocade.com. Table 3-1 details the current Brocade recommended support levels for large fabrics. If a certain level of Fabric OS does not appear in Table3 -1, it does not mean that version will not work or is not supported. The table simply reflects testing levels as of this writing, but it is possible that a version that has not undergone this testing will work without issues.... ..."

"... In PAGE 5: ... 3.2 Analysis of estimated models Initial analysis of the estimated models using the first set of data from region 1 ( Table2 ) did not pro- vide definitive information to decide whether the estimated regressions were all similar. The differ- ence in x2 values for the estimated unrestricted and partially restricted models was 5.... In PAGE 5: ... Region 1 included plots from both interior and coastal Alaska whereas region 2 plots were from the interior and region 3 plots were from southeast coastal Alaska. A second, independent sample was drawn from region 1 ( Table2 ). It was felt that if the second esti- mated model for region 1 proved to be similar to the first estimated model,-then observer learning could ... ..."