Which of the following is true about the multiple baseline design? check all that apply.

2.3 Combining Within- and Between-series Elements: Multiple-baseline Design

Multiple-baseline designs build upon and integrate the basic logic and structure of within- and between-series elements. The fundamental premise of multiple-baseline designs is to replicate phase change effects systematically in more than one series, with each subsequent uninterrupted series serving as a control condition for the preceding interrupted series. Series can be compared and arranged across behaviors, across settings, across individuals, or some combination of these (see Hayes et al. 1999). Such designs require that the series under consideration be independent (e.g., two functionally distinct behaviors), and that intervention be administered sequentially beginning with the first series, while others are left uninterrupted as controls. For example, a client might present with three distinct behavior problems all requiring exposure therapy. All behaviors would be monitored during baseline (A phase), and after some semblance of stability is reached the treatment (B phase) would be applied to the first behavior series, while the remaining two behaviors are continuously monitored in an extended baseline. Once changes in the first behavior resulting from treatment reach stability, the second series would be interrupted and treatment applied, while the first behavior continues in the B phase and the third behavior is monitored in an extended baseline. The procedure followed for the first two series elements is then repeated for the third behavior. This logic can be similarly applied to multiple-baseline designs across individuals or settings (see Barlow and Hersen 1984, Hall et al.). More complex within-series elements (e.g., A-B-A-B, or B-C-B, or counterbalanced phases such as B-A-B and A-B-A) can be evaluated across behaviors, settings, and individuals.

Note that with multiple-baseline designs the treatment is never withdrawn, rather, it is introduced systematically across a set of dependent variables. Independent variable effects are denoted from how reliably behavior change correlates with the onset of treatment for a particular dependent variable. For such reasons, multiple-baseline designs are quite popular, owing much to their ease of use, strength in ruling out threats to internal validity, built-in replication, and fit with the demands of applied practitioners working in therapy, schools, and institutions. Indeed, such designs can be useful when therapists are working with several clients presenting with similar problems, when clients begin therapy at different times, or in cases where targets for change in therapy occur sequentially.

4.4.3 Multiple-baseline design

The third experimental design to show the effect of interventions is the multiple-baseline design. The aim of this design is to demonstrate that the change in behavior is associated with the introduction of the contingency at different time points. This design requires the use of two or more interventions. After the baseline reaches a stable rate, one of the interventions is introduced, while baseline conditions are continued for other interventions. When the rates are stable once again, the second intervention is introduced. The effects of each intervention are established only after it has been introduced.

One problem that can occur with the multiple-baseline design is that changes in behavior take place too early. If behavior changes before the intervention is introduced, it is not clear whether the intervention is responsible for the changes or not (Kazdin 1975). Another problem is that the effects of the different interventions could be confused.

3.6.1 Linear Quadratic Regulator Formulation

The linear quadratic regulator (LQR) is a classical optimal control problem used by many control engineers. Solutions to LQR are easy to compute and can typically be used to compute a baseline design useful for comparison. A formulation of the LQR problem considers the state equation of the plant:

(3.22)x˙(t)=Ax(t)+Bu(t).

The formulation also considers the following quadratic cost function of the states and control input:

(3.23)J=12xT(tf)Sx(tf)+1 2∫0tfxT(t)Qx(t )+uT(t)Ru(t)dt,

where S = ST ≥ 0, Q = QT ≥ 0 and R = RT > 0.

To minimize the cost function, consider the Hamiltonian system with state and costate (p(t)) dynamics given by:

(3.24)[x˙(t)p˙(t)]=[A −BR−1BT−Q−AT ][x(t)p(t)] .withinitialcondition:x(0)=x0;p(tf)=Sx(tf).

The optimal controller uses full-state feedback and is given by:

u(t)=−R−1BTP(t)x(t)=−K(t)x(t),

where P(t) is the solution of the matrix Riccati equation:

P˙(t)=−P(t)A−AT P(t)+P(t)BR−1BTP(t )−Q,

which is solved backward in time starting at P(tf) = S. The optimal feedback gain matrix K(t) is given by:

K=R−1BTP(t),

and the optimal cost is as follows:

Jopt=12xT(0)P(0)x(0).

1 INTRODUCTION

The (double) dielectric window is an essential and the most vulnerable component of the ICRF Vacuum Transmission Line for ITER, as it provides ultimate vacuum and tritium containment. Eight windows are present in each of the four ICRF arrays in the present baseline design. Although several models of dielectric window have been developed and are currently in use [1], the ITER stringent reliability requirements call for special design and prototyping to allow a realistic evaluation of the failure rate. The ITER window design requires specific design and careful selection of the dielectric material because of the long discharge pulse, high electric field strengths, a possible degradation of the dielectric properties due to neutron or gamma irradiation, possible changes in the mechanical, and thermal properties and in gas permeation [2]. Furthermore, the metal-ceramic joints required for the windows and the support structures need to retain reliable vacuum tightness under cyclic operation conditions.

In the ITER VTL, the window assembly will be located on the equatorial duct in the region between the back of the ICRF Array neutron shield and the exit of the cryostat. The earlier calculations of neutron radiation estimates for these positions have been based on simplified homogenized geometries for the shield/blanket, vacuum vessel, cryostat, divertor, midplane port and ICRF array neglecting the neutron streaming. In the present work, detailed neutron radiation calculations are performed with the MCNP-4 code with account for the neutron streaming through the detailed structure of the ICRF Array/Shield/VTL assembly.

BeO and Al2O3 (97.5% to 99.99% purity) are considered as dielectric materials. 97.5% Al2O3 (polycrystalline) is internationally recognized and stable standard dielectric, with probably the best database available for ICRF conditions. The 5-10 times higher thermal conductivity and the unsensitivity of the loss tangent against modest neutron radiation for BeO, and the good experience with BeO windows at higher RF frequencies, make the latter dielectric interesting for the present study.

The annular windows are assumed to be axisymmetric fitting the characteristic impedance of 30 Ohms and dimensions of the transmission line. The modeling includes the double window structure with a dynamically evacuated intermediate region, 10−5-10−2 Pa vacuum conditions on the antenna side, 300 kPa dry air or 100 kPa SF6 conditions on the pressurized size, cooling on the inner and outer conductor, coronal rings to reduce the electric fields close to the joints, and the joining structure. The design is based on the maximum operating voltage of 50 kV peak RF voltage, with arbitrary amplitude modulation. 2-dimensional finite element codes IVOFEM/IVOHEAT [3] and ANSYS are used for the evaluation of the dielectric losses, related heating of the ceramics window, temperature distribution, created stresses, electric potential distribution, and heat conduction from the ceramics through the joints to the cooling channels. Water cooling with inlet temperature 100 °C and outlet temperature 150 °C with coolant pressure 4 MPa are assumed. The results summarized in this paper are reported in details in the ITER Task 238/2 Final Report.

4.5.4 Power consumption analysis

Figure 4.18 shows the power consumption of the NoC designs. The power consumption results were obtained by executing application traffic in an 8 × 8 mesh topology. As the figure indicates, the conventional NoC design has the largest power consumption, and is used as the baseline design. The NOCHI EVC design reduces the power consumption by an average of about 14%. This is mainly due to a reduction in buffer and crossbar power consumption in bypassing routers. In a similar way, the proposed VBON design can also reduce the power consumption by about 20%. In summary, the proposed VBON design clearly outperforms the other two NoC designs across all benchmarks not only in performance but also in power consumption.

3.2 The Multiple-baseline Design

An A-B-A or A-B-A-C design assumes that the treatment (B or C) can be reversed during the subsequent A period. Sometimes it is impossible or unethical to re-institute the original baseline conditions (A). In these cases, other designs can be used. One such approach is a multiple-baseline design. In a multiple-baseline design, baseline data are gathered across several environments (or behaviors). Then a treatment is introduced in one environment. Data continue to be gathered in selected other environments. Subsequently the treatment is implemented in each of the other environments, one at a time, and changes in target behavior observed (Poling and Grossett 1986).

In the earlier example, if the child had been exhibiting disruptive behavior in math, spelling, and social studies classes, the same analysis of the problem might be applied. The teachers or the researcher might not be willing to reinstitute the baseline conditions if improvements were noted because of its disruptive effects on the learning of others and because it is not in the best interests of the child. Figure 2 shows how a multiple-baseline design might be implemented and the data presented visually. Baseline data are shown for each of the three classroom environments.

While the initial level of disruptive behavior might be slightly different in each class, it is still high. The top graph in Fig. 2 shows the baseline number of disruptive behaviors in the math class. At the point where the vertical dotted line appears, the intervention is implemented and its effects appear to the right of dotted line. The middle and bottoms graphs show the same disruptive behaviors for the spelling and social studies class. No intervention has yet been implemented and the disruptive behavior remains high and relatively stable in each of these two classrooms. This suggests that the changes seen in the math class were not due to some other cause outside of the school environment or some general policy change within the school, since the baseline conditions and frequency of disruptive behaviors are unchanged in the spelling and social studies classes. After four more weeks of baseline observation, the same shift in attention treatment is implemented in the spelling class, but still not in the social studies class. The amount of disruptive behavior decreases following the treatment implementation in the spelling class, but not the social studies class.

This second change from baseline is a replication of the effect of treatment shown in the first classroom and is further evidence that the independent variable is the cause of the change. There is no change in the social studies class behavior. This observation provides additional evidence that the independent variable rather than some extraneous variable is responsible for the change in the behavior of interest. Finally, the treatment is implemented in the social studies class with a resulting change paralleling those occurring in the other two classes when they were changed. Rather than having to reverse the salutary effects of a successful treatment as one would in an A-B-A reversal design, a multiple baseline design allows for successively extending the effects to new contexts as a means of demonstrating causal control.

Multiple-baseline designs are used frequently in clinical settings because they do not require reversal of beneficial effects to demonstrate causality. Though Fig. 2 demonstrates control of the problem behavior by implementing changes sequentially across multiple settings, one could keep the environment constant and study the effects of a hypothesized controlling variable by targeting several individual behaviors. For example, a clinician could use this design to test whether disruptive behavior in an institutionalized patient could be reduced by sequentially reinforcing more constructive alternatives. A treatment team might target aggressive behavior, autistic speech, and odd motor behavior when the patient is in the day room. After establishing a baseline for each behavior, the team could intervene first on the aggressive behavior while keeping the preexisting contingencies in place for the other two behaviors. After the aggressive behavior has changed (or a predetermined time has elapsed), the autistic speech behavior in the same day room could be targeted, and so on for the odd motor behavior. The logic of the application of a multiple-baseline design is the same whether one studies the same behavior across different contexts or different behaviors across the same context. To demonstrate causal control, the behavior should change only when it is the target of a specific change strategy and other behaviors not targeted should remain at their previous baselines.

Interpretation can be difficult in multiple-baseline designs because it is not always easy to find behaviors that are functionally independent. Again, consider the disruptive classroom behavior described earlier. Even if one's analysis of teacher attention to disruptive behavior is reinforcing, it may be difficult to show change in only the targeted behavior in a specific classroom. The child may learn that alternative behaviors can be reinforced in the other classrooms and the child may alter his or her behavior in ways that alter the teacher reactions, even though the teachers in the other baseline classes did not intend to change their behavior. Nevertheless, this design addresses many ethical and practical concerns about A-B-A (reversal) designs.

9.5.2.6 The effect of application communication: Power and scalability

Figure 9.13a illustrates the power consumption results of the proposed NoC-cache system. The power consumption results are obtained by executing the application traffic with a 64-core configuration. As the figure indicates, the conventional NoC design has the largest power consumption, which is used as the baseline design. The VBON point-to-point MPI support reduces the power consumption by an average of about 11%. This is mainly due to a reduction in the buffer and crossbar power consumption in bypassing routers. The proposed communication design can further reduce the power consumption, by about 3%. In summary, the proposed communication design clearly outperforms the other two NoC designs across all benchmarks not only in the performance but also in the power consumption.

To show the scalability of the proposed design, Figure 9.13b illustrates the computation results for the speedup over the conventional MPI support design with 4-core to 64-core configurations. The 4-core design is used as the baseline design, whose performance speedup is normalized to 1. That is, we compare the speedup of each configuration of the core count with the baseline MPI support design. The speedup of the 64-core configuration outperforms the speedup of 4-core and 16-core configurations for all benchmarks. For the 16-core system, about 25% more speedup can be obtained over the 4-core system; for the 64-core system, about 47% more speedup can be obtained over the 4-core system. This is because as the number of cores increases, reducing the MPI communication delay through hardware support becomes more important. The evaluation result successfully demonstrates that the proposed design has good scalability as the number of cores increases.

15.2.2.4.2 Uplink reference signals

LTE-M-U supports the uplink Demodulation Reference Signal (DMRS) and the Sounding Reference Signal (SRS). Their purpose and design follow LTE-M described in Section 5.2.5.3 with some small deviations. Since the uPUSCH transmission bandwidth is restricted to one narrowband, so is the DMRS bandwidth for the support of coherent uPUSCH demodulation. DMRS for uPUCCH formats 1, 1a, 2 and 2a are supported, just as for LTE-M operating as a FDD system. uPUCCH format 3 was added to the LTE-M-U design baseline to support ACK/NACK of up to 10 HARQ processes. For uPUCCH Format 3 the DMRS is mapped on symbols 1 and 5 in each slot as depicted in Fig. 15.8.

The SRS is mapped over the center four PRBs in subframes fulfilling nsfrelmod TSFC∈ΔSFC in mframes satisfying nmframemodTmframe=0 . nsfrel is the relative subframe numbering, TSFC is the SRS periodicity within the applicable mframes, ΔSFC defines the subframe for SRS transmissions within the period TSFC. Tmframe is finally the SRS mframe periodicity. 32 different SRS configurations are supported. In its most dense configuration the SRS is transmitted once every five uplink subframes. In the least dense configuration two SRS subframes are transmitted every four mframes.

10.5.2.2 Hotspot traffic pattern

For the hotspot traffic pattern, we set one or two network nodes as the hotspot nodes to which other nodes send data messages with a greater probability. To simulate the unexpected cases of receiving messages, MPI_Receive is executed later than MPI_Send (uniform random ranging from 1 to 256 cycles) with 30% possibility. Unlike for the round-trip traffic, we use the average network traffic and message delay in the hotspot and real traffic scenarios. We first evaluate the protocols through the network traffic, which is identified as the number of bytes transmitted through the underlying NoC. The synchronous protocol serves as the baseline design; that is, the protocol is normalized to one in different hotspot scenarios. Figure 10.14 illustrates the network traffic comparison results for different communication protocols. The buffered protocol entails significantly more network traffic than the synchronous protocol for the retry mechanism, which is approximately 29% for the single-hotspot scenario and approximately 43% for the double-hotspot scenario. The ADCM approach minimizes the network traffic overhead. This traffic overhead is primarily attributed to the wrong buffer usage predictions or partial data retry operations.

Figure 10.15 illustrates the average message delay comparison results for different protocols. We define the message delay as the time interval between the send and receive operations for each data unit. The ADCM approach has the least message delay compared with the two other protocols. The ADCM outperforms the synchronous protocol in terms of the elimination of the handshaking process in most cases and is better than the buffered protocol in terms of the reduction of retry delay. The message delay in the double-hotspot scenario is longer than that in the single-hotspot scenario primarily because of the increased retry delay and network traffic contention. The ADCM minimizes this negative impact, achieving the best tradeoff between the message delay and the traffic load.

2.1.6 Review Process

Reviews take place at many levels in a project, from detailed circuit board reviews and software code walkthroughs to major mission-level, project-wide reviews. As illustrated in the previous section (see Fig. 5.6), many of the major mission-level reviews form milestones or “gates” between project phases. Customer reviews during the early phases provide input to the selection process during competitive procurements. Later in the project life cycle, technical reviews benefit the project, because they can catch problems or potential problems that may have been overlooked by the development team. Reviews also provide the project with fresh insights from experienced reviewers for potential improvements and lessons learned.

Before going into the purpose and content of individual reviews, some key elements of any effective design review should be mentioned. Every review must have a review board with a chairperson. For small reviews this board may consist of only a few people, but for mission-level reviews the board may have as many as 10 or more members. The chairperson is responsible for selecting the board and conducting the review. To provide an independent viewpoint, the board members should not be involved in the design that is being reviewed. To allow time for a comprehensive review, the design review package should be provided to the review board at least 1 week prior (2 weeks is preferable) to the start of the review. The review package should contain the agenda and a summary of the requirements and the baseline design.

During the review, the chairperson is responsible for making sure that the agenda is completed, the minutes are recorded, and the action items are captured. After the review the chairperson distributes the minutes as a record of the review, including the date, the agenda, attendees, any major decisions, and the list of action items with assignments and due dates. Finally, it is the responsibility of project team to provide timely written response to each action item.

Major milestones illustrated in Fig. 5.6 are explained in Table 5.7.

Table 5.7. Major Project Reviews for Spacecraft Development

In addition to the major project-level reviews described in Table 5.7, peer reviews of components and subassembly designs are also a key part of the technical review process. In these peer reviews, component test plans are checked to ensure that all requirements are met. At these lower levels of assembly, more of the details of the designs and tests are evaluated.

Implementing good review practices help make any review more productive. Every review should have a review chair, who is responsible for selecting the review committee and conducting the review. The review chair should be experienced in the subject matter and independent from the members of the team presenting the review. For project-level reviews, the review chair is selected from individuals outside the project and often outside the organization. For some projects, the review chair of project-level review may be a member of the sponsor organization. For subsystem-level peer reviews, the chair may be part of the project, but must be independent from the team implementing the design under review. In general, the objective of the review is to evaluate whether the design meets the requirements. To achieve that objective, the requirements must be documented and the team must have selected a single design. In advance of the meeting, the agenda and presentation material must be supplied to the review committee with sufficient time to prepare for the review. This time may vary depending on the amount of material involved in the review. Typically, a week or more is required for a project-level review. During the review, meeting minutes and action items must be recorded. After the meeting, the minutes should be distributed and the action items tracked to closure.

Closing an action item requires a procedure or process; Fig. 5.7 illustrates one example for closing an action item. An action time may lead to a minor correction or it may force a major design change or it may prove intractable and force a major change in the mission or re-evaluation. A record of all action items and their disposition should be archived in the project depository or database.