Antenna Research
Advanced Antenna Design
With Sciperio's genetic, evolutionary, and neural network algorithms, it is possible to reduce the size and weight of traditional antennas, while optimizing them for each environment they'll work within.
There is no one-size-, one-shape-fits-all "super" antenna that can be used universally in any device for any purpose. Each transmitter and each receiver is unique as the material they're made of and the environment in which they operate. Each line, each curve, and each turn of an antenna has a purpose. Finding the optimal settings for each is the challenge in making an excellent antenna. With so many variations in shape, size, and RF requirements, finding the right match may seem overwhelming.
It only seems this way.
With Sciperio's genetic, evolutionary, and neural network algorithms, it is possible to reduce the size and weight of traditional antennas, while optimizing them for each environment they'll work within.
Case Study:
Stochastic Compression using Genetic Designs
Goal: Design optimized miniature antennas which may be embedded into any RF technology.
Results: Antennas with reduced size & volume
Approach 1
- Apply stochastic (random-like) design rules to genetic algorithm (GA)
- GA will optimize an antenna according to ANY quantifiable parameter (VSWR, impedance, metal obstacles, Gain, BW, conformality, etc.)
Genetically Designed Stochastic vs. Standard Linear Comparison
Crossed dipoles suspended at λ/2 over ground plane
Left: Crossed Linear Dipole,
Footprint = 19.1 in2
Right: Crossed Stochastic Dipole,
Footprint = 6.8 in2
Total Area Reduction:
65%
Approach 2
- Using stochastic approach with LC components in GA
- GA will optimize the antenna configuration as well as place the LC components
Dualband GPS Antenna