No abstract

No abstract

In Figure 5(a)-(b), the bias dependency of both drain current and S parameters are accurately captured by the neural network. In addition, a harmonic balance simulation tests the accuracy of the model for large-signal non-linear simulation. Figure 5(c) shows simulated output power for fundamental, second, and third harmonics both as a function of input power and compared with measurement beyond the 1-dB compression point. As can be seen, accurate reproduction of output power at fundamental frequency has been achieved, with second and third harmonics having a maximum error of only 2-3 dBm over the power sweep range. The power level is pushed beyond the compression point to illustrate the validity of the model under high power conditions. CONCLUSIONWe have reported a neural-network-based non-linear model for CMOS, based on capacitance, extrinsic gate, and substrate resistances parameters together with drain current. The parameters used in the model show strong dependency on bias, which are difficult to formulate accurately using empirical equation. Therefore, we have used three-layered neural network architecture to derive parameter-bias-dependent function. The developed non-linear function has then been adopted in a commercial ADS circuit simulator, using symbolically defined devices (SDD) required for implementation of the CMOS non-linear model. The non-linear mode of operation of the full model has been validated, including harmonic power, and good fits were obtained.

by improving the mechanical scan system, the hysteresis effect can be reduced considerably. A similar experiment is performed for the DR sensor and a compatible result is obtained, which indicates that such effect is related to the common mechanical part rather than the other parts. CONCLUSIONDesign and performance of two triple-fiber probe systems for the distance measurements are described and compared. Using the reflection intensity variation with the target distance, the position of the target reflecting surface can be determined precisely by such sensors. The reflection signals of the metallic and nonmetallic surfaces were investigated as a function of the target axial position. Two DR and DT arrangements are tested and the results are compared for the optimum performance. For any measurement, system sensitivity, precision, reproducibility, and hysteresis effect are important factors, and thus, reported for the sensing systems.As expected for both DR and DT designs, sensitivity is higher for the mirror surface in comparison with the A4 paper surface. One can see for both systems, the front-slope and back-slope sensitivity can be considered for the signal response curve. For both designs, the front-slope sensitivity is higher than that of the backslope. For the DR design and mirror target, the averaged frontslope sensitivity is a 0.281 mV/mm, while the back-slope sensitivity is about 0.058 mV/mm. Similarly for the DT design the frontslope sensitivity is 0.182 mV/mm in comparison with the 0.036 mV/mm for the back-slope sensitivity. Both systems can be used for the small distance measurement while the DR design shows two advantages: first, for both the mirror and paper surfaces, the overall sensitivity is higher for the DR design (35.7% higher for mirror). Second, in a case of no fiber and detector mismatch, one can investigate the surface nonuniformity, and target unparallelism from a possible difference signal. For the DR design, corresponding signals for the two receiver fibers (#1 and #2 in Fig. 1) can be measured and compared. The effect of the surface material, surface structure, and curvature can be also investigated with the reported DR probe design.The reproducibility of the measured results is investigated and the results are described. The maximum signal difference is approximately 64 mV for the full scale signal level of 562.5 mV for the DR design (Fig. 5), which results a reproducibility error of the order of 1.42% for this arrangement. A similar experiment is performed for the DT design, and a compatible result is obtained. Our experiments show that most of the hysteresis effect is due to performance of the XYZ-mechanical system. The maximum signal deviation due to the hysteresis effect is approximately 64 mV (Fig. 6), which results in a hysteresis error of about 2.17% for the DT arrangement. A similar experiment is performed for the DR probe and a compatible result is obtained, which indicates that such an error is related to the mechanical part, which is common for two designs.It is noticed that...

No abstract