### TABLE II INTERMEDIATE ON-OFF SCENARIOS.

### Table 3: Overview of the Observations performed with the HEGRA CT System. All results are based on preliminary (partly online) analysis of the data. Upper limits correspond to 0.99865 (3 ) con dence level and are calculated using the method described in (Helene, 1983). For all observed objects, except G78, exist the same amount of o source data, either due to appropriate ON-OFF pairs or due to the so called wobble mode, where the source is displaced by 0.5 in respect to the camera center and the displacement is changed every 20 minutes.

"... In PAGE 5: ... This was done partly in parallel to the commissioning of the telescope system. Table3 summarizes the observations. Table 3: Overview of the Observations performed with the HEGRA CT System.... ..."

### Table 1: Statistics for an ON-OFF #0Dow with 19 competing

"... In PAGE 8: ... The ON-OFF source sent 6758 packets over the course of the experiment. Table1 shows the number of packets from the ON-OFF source dropped at the congested link. The DRR results show what happens when the ON-OFF source is restricted to its fair share at all times.... ..."

### Table 4: ON-OFF bursty traffic-overlapped RED

2000

"... In PAGE 8: ...he third test for this type of ON-OFF traffic used RED parameters shown in fig 7. Experiment 5.1b was repeated with these new RED parameters. Table4 shows the observed results at the core device. Examining Drop Probability 1 Figure 7: Overlapped RED parameter settings... ..."

Cited by 11

### Table 1: Example of Button function according to interaction mode ON BUTTON Interaction mode Defined function description Related state

"... In PAGE 4: ... We can manipulate large object more easily by extending attached line segment. Table1 shows how can we associate finger button ON/OFF states with function for interaction. ON BUTTON shows which finger-button collides with an object.... ..."

### TABLE I REQUIRED PERFORMANCE FOR MULTICLASS ON/OFF TRAFFIC

1998

Cited by 2

### Table 9. Parameters for the On/Off model to generate data traffic

### Table 2 Prediction of frame error rate From Table 2, we observe that the Extended On/Off model accurately captures the FER for both part1 and part2 of the GSM trace. The 4-th-order MTA model, even with larger state space, fare less well. The 4-th-order Markov model and the 5-states HMM can predict FER for part1 accurately; yet for part2, neither of them performs

in Modeling frame-level errors in GSM wireless channels,” in Special Issue of Performance Evaluation

2004

"... In PAGE 12: ... In addition, we also observe that these models do not capture the autocorrelations of part1 and part2 well, as shown in Fig 7. Finally, we note that, as shown in Table2 of Section 5, the 4th-order MTA model does not predict the frame error rate for either part1 or part2 accurately, and the 5-state HMM also does not accurately predict the FER for part2 on average. These results suggest that existing models do not capture the tail behavior of the burst length distributions, the autocorrelation, or even the frame error rate very well.... In PAGE 17: ...ig. 12. Performance of the extended On/Off model with or without EM algorithm for part2 6 Evaluation Results of the Extended On/Off Model As in the previous section, we are interested in how well the performance measures predicted by the extended On/Off model match the measured observations in the empirical traces, and how the predictions of the extended On/Off model compare with those of the other models we described in the previous section. In Table2 , we compare the predicted frame-error rates of all the models presented in this paper. For HMMs, the nal HMM depends on the initial estimate of the parameters [10].... In PAGE 17: ... For HMMs, the nal HMM depends on the initial estimate of the parameters [10]. Thus, we computed ten 5-state HMMs, and their corresponding FERs, and show the mean value and the errors of the FERs in Table2 . The extended On/Off model for part1 has three Geometric phases in the Off period and four Geometric phases in the On period, and the extended On/Off model for part2 has one Geometric phase in the Off period and three Geometric phases in the On period.... In PAGE 18: ...ig. 14. Comparison of autocorrelations provided by the 4-th-order MTA model and the extended On/Off model as well as the extended On/Off model. Note that although the HMM considered in Table2 has only 5 states, the complexity of computing its parameters is signi - cantly greater than that of the extended On/Off model, since transitions and outputs in the latter are much more constrained. We also computed the CCDF and the autocorrelation of the extended On/Off model, and compared these with those of other models discussed.... ..."

Cited by 3