Tracking-mode performance

The tracking mode performance of the on-chip digital temperature controller of the differential mixed-signal architecture is shown in Fig. 6.11. The measurement was done at room temperature, and the digital code of the reference temperature of microhotplate 1 was increased in steps of 100 digits. Microhotplates 2 and 3 were switched off during the measurement. The resolution of the digital temperature controller was better than 2 °C. 

Fig. 5.19. Tracking mode operation performance of the temperature controller...

Fig. 6.4. Performance of the single-ended analog proportional temperatm-e controller in the tracking mode...

The performance of the temperature controller was measured in the tracking mode. Figure 6.18 shows a graph, where the temperature of one of the three microhotplates is kept at a constant temperature of 300 °C, the temperature of the second microhotplate is modulated using a sine wave of 10 mHz, while rectangular temperature steps of 150 °C, 200 °C, 250 °C, 300 °C, and 350 °C have been appHed to the third microhotplate. Temperature measurements on one of the hotplate that has been operated at constant temperature in the stabihzation mode showed a variation of less than 1 °C, even though the temperature of the neighboring hotplates was, at the same time, modulated dynamically (sine wave, ramp, steps). This is a consequence of the individual hotplate temperature control, without which thermal crosstalk between the hotplates would have been clearly detectable. The power dissipation of the chip is approximately 190 mW, when all three hotplates are simultaneously heated to 350 °C. In the power-down mode, the power consumption is reduced to 8.5 mW. 

The performance (see Table 1) was measured with the program version without screen display using 4 patches. 652 object points were matched. The patch was size 7x7 pixels. The radiometric equalisation by mean and variance was applied before each iteration and the gradients were derived from the average of template and patch. The computations were performed in the tracking mode, therefore the average number of iterations was only 4.3 iterations per point. 

If a rapid screening test is used to assay for the drug or residue that is known to be present in the animal, they key performance element is the specificity of the assay response for the test analyte. A typical use of a rapid screening test in this mode would be in a pharmacokinetic study to determine the half-life of a drug in the blood of a treated animal. The assay used should be able to track the concentration of drug with acceptable specificity and provide a kinetic assessment that reflects the effects of only the animal drug under test. 

Modem computerized analytical instruments have quantitative analysis programs that allow the analyst to specify the calibration standard concentrations, select the curve-fitting mode, and calculate the results of the samples from the calibration curve equation. Many of these programs will remn outher standards and samples automatically, flag suspect data, compute precision and recovery of spikes, track reference standards for quahty control, and perform many other functions that used to be done manually by the analyst. 

Collectively, the findings of Prinzel et al. (1997) help to validate the operation of the biocybemetic system, particularly in the negative feedback condition. Specifically, the value of index beta/(alpha + theta) is lower when individuals are actively engaged in the tracking task than when they are merely observing it under automatic mode. Further, examination of the RMSE scores showed that the system produced better performance under negative feedback. 

---