Introduction
State-of-the-art
Baseline of investigation
Circuits
Meta-heuristic algorithms
-
w: Inertia. typ.range: [0, 1].
-
c1: Cognitive scaling factor. typ. range: [0, 2].
-
c2: Social scaling factor. typ. range: [0, 2].
-
r1, r2: Random values between 0 and 1.
-
Velocity: Particle’s velocity.
-
Local: Particle’s local best known position.
-
Global: Swarm’s best known position.
-
Current: Current position of the particle.
-
HMS (harmony memory size): problem dependent.
-
HMCR (harmony memory considering rate): typ.range: [0.7, 0.99].
-
PAR (pitch adjusting rate): typ range: [0.1, 0.5].
Multi-objective optimization approach
Simulator | Real execution time (s) | User \(+\) system execution time (s) |
---|---|---|
OCN | 240 | 45 |
NGS | 8 | 6 |
SVR | 0.5 | 0.4 |
Simulator | Real execution time (min) | User \(+\) system execution time (min) |
---|---|---|
OCN | 395 | 93 |
NGS | 15 | 9 |
SVR | 20 | 5 |
Time measurement
ABSYNTH concept and architecture
Hybrid multi-objective optimization approach
Incrementally evolving self-learning capability
Status of the ABSYNTH architecture’s implementation
TRAVISOS: optimization space visualization
Experiments and results
Circuit sizing methods results
PSO | SVR | NGS | OCN |
---|---|---|---|
Num. Par. | 20 | 10 | 10 |
Max. Gen. | 1000 | 1000 | 10 |
Target fitness | Mean < 2 | Mean < 0.2 | Best \(\le \) 0 |
C1 | 2 | 2 | 2 |
C2 | 2 | 2 | 2 |
Inertia | 0.5 | 0.5 | 0.5 |
Min | 1 | 1 | 1 |
Max | 100 | 100 | 100 |
HS | SVR | NGS | OCN |
Harmony size | 20 | 10 | – |
HCMR | 0.9 | 0.9 | – |
PAR | 0.3 | 0.3 | – |
Max. Iter. | 10,000 | 10,000 | – |
Target fitness | Mean < 2 | Mean < 0.2 | – |
Min | 1 | 1 | – |
Max | 100 | 100 | – |
Spec. | Miller amplifier | Folded-cascode amplifier | Buffer amplifier | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Target | Manual | PSO | PPP | HHP | Target | Manual | PSO | PPP | HHP | Target | Manual | PSO | PPP | HHP | |
Gain (dB) | 70 | 72.6 | 73.62 | 72.15 | 75.31 | 60 | 56.92 | 67.10 | 68.17 | 63.40 | 50 | 54 | 51.10 | 51.81 | 50.83 |
Std. (dB) | – | – | 2.31 | 2.0 | 3.16 | – | – | 2.97 | 6.78 | 3.28 | – | – | 0.72 | 0.27 | 0.4 |
BW (MHz) | 10 | 18.97 | 37.60 | 43.52 | 43.78 | 10 | 52.13 | 58.45 | 36.92 | 39.32 | 10 | 27 | 32.45 | 19.64 | 27.11 |
Std. (MHz) | – | – | 4.75 | 6.34 | 8.29 | – | – | 15 | 18.4 | 13.13 | – | – | 3.6 | 7.4 | 11.4 |
SR (V/s) | 10 | 60.85 | 11.26 | 17.56 | 18.50 | 10 | 54.99 | 49.53 | 37.61 | 24.74 | 10 | 14 | 27.21 | 18.14 | 23.36 |
Std. (V/s) | – | – | 0.74 | 3.7 | 6.06 | – | – | 14.64 | 1.28 | 12.6 | – | – | 3.55 | 3.73 | 6.99 |
CMRR (dB) | 80 | 77.1 | 93.39 | 92.46 | 92.00 | 80 | 90.89 | 88.25 | 92.63 | 90.49 | 80 | 80 | 109.69 | 109.72 | 116.52 |
Std. (dB) | – | – | 0.84 | 2.63 | 3.88 | – | – | 1.29 | 0.91 | 4.57 | – | – | 12.09 | 10.54 | 19.49 |
Offset (\(\upmu \)V) | 1000 | 0.0114 |
\(-\)92.1 |
\(-\)58.9 |
\(-\)24.7 | 1000 | 308 | 17.5 |
\(-\)215 | 16.9 | 1000 |
\(-\)100 |
\(-\)735 |
\(-\)358 |
\(-\)738 |
Std. (\(\upmu \)V) | – | – | 145 | 138.34 | 126.3 | – | – | 31.9 | 70.3 | 257 | – | – | 200 | 600 | 214 |
ICMR+ (V) | 1 | 1.42 | 1.32 | 1.17 | 1.27 | 0.75 | 0.75 | 0.75 | 0.76 | 0.81 | 0.45 | 0.6 | 0.9 | 1.05 | 0.96 |
Std. (V) | – | – | 0.075 | 0.106 | 0.075 | – | – | 0.001 | 0.07 | 0.073 | – | – | 0.055 | 0.06 | 0.12 |
ICMR− (V) | –1 |
\(-\)1.54 |
\(-\)1.65 |
\(-\)1.65 |
\(-\)1.65 |
\(-\)0.75 |
\(-\)1.5 |
\(-\)1.50 |
\(-\)1.42 |
\(-\)1.50 |
\(-\)0.45 |
\(-\)0.5 |
\(-\)0.45 |
\(-\)0.45 |
\(-\)0.45 |
Std. (V) | – | – | 0 | 0.001 | 0.001 | – | – | 0.0 | 0.0 | 0.001 | – | – | 0 | 0.001 | 0.0 |
OS+ (V) | 1 | 1.52 | 1.63 | 1.61 | 1.62 | 1 | 1.64 | 1.64 | 1.64 | 1.64 | 0.5 | 0.6 | 0.63 | 0.62 | 0.63 |
Std. (V) | – | – | 0.007 | 0.018 | 0.011 | – | – | 0.0005 | 0.003 | 0.001 | – | – | 0.005 | 0.004 | 0 |
OS− (V) |
\(-\)1 |
\(-\)1.58 |
\(-\)1.63 |
\(-\)1.63 |
\(-\)1.63 |
\(-\)1 |
\(-\)1.64 |
\(-\)1.63 |
\(-\)1.59 |
\(-\)1.60 |
\(-\)0.5 |
\(-\)0.6 |
\(-\)0.53 |
\(-\)0.51 |
\(-\)0.52 |
Std. (V) | – | – | 0.004 | 0.012 | 0.009 | – | – | 0.0003 | 0.0001 | 0.019 | – | – | 0.009 | 0.008 | 0.004 |
PD (mW) | 2 | 2.13 | 1.33 | 1.51 | 1.32 | 1 | 0.59 | 0.754 | 0.702 | 0.647 | 2 | 0.5 | 1.83 | 1.42 | 1.44 |
Std. (mW) | – | – | 0.44 | 0.2 | 0.21 | – | – | 0.060 | 0.027 | 0.008 | – | – | 0.32 | 0.3 | 0.358 |
Time (min.) | NA | Days | 80 | 35 | 21 | NA | Days | 85 | 46 | 30 | NA | Days | 70 | 38 | 27 |
Std. (min.) | – | – | 25 | 12.3 | 4.2 | – | – | 33 | 25.1 | 13.6 | – | – | 20.2 | 17.1 | 7.38 |
Visualization methods results
Methods | Equation based | Function approximator based | Simulation based | ABSYNTH |
---|---|---|---|---|
Speed | High | High | Low | Low, improves with repetitions |
Accuracy | High | Low, improves with examples | High | High |
Preparation effort | High | High | Low | Low |
Human influence for new circuits | Design of equations | 100\(+\) examples for learning | Almost nothing | Almost nothing |
Evolution over repetitions | Not possible | Not found | Not possible | Present |
Human computer interface | No research found | No research found | No research found | TRAVISOS, GoalPost view |