Ultrahigh-frequency bandwidths are easily blocked by objects, leading to transmission losses when walking between rooms or walls. This type of technology has become highly demanded due to its ability to carry vast amounts of data needed in the virtual reality industry.
Machine Learning System Shapes Signals to Avoid Obstacles
In response, a team of Princeton researchers have developed a machine-learning system that could allow ultrahigh-frequency transmissions to dodge those obstacles. They created a system that shapes transmissions to avoid obstacles coupled with a neural network that can rapidly adjust to a complex and dynamic environment.
The team bent transmission beams by transmitting a signal that curves around the obstruction, using an idea proposed in 1979 for a kind of radio wave called Airy beams.
Airy beams allow engineers to shape transmissions like curveballs – when correctly controlled, the beams manoeuvre through a complex and moving field of objects.
Advancing Wireless Connectivity with Sub-Terahertz Frequencies
Princeton professor Yasaman Ghasempour led a team to enable data transmission in the sub-terahertz band, at the upper end of the microwave spectrum.
Compared to static systems, the new system allows transmitters to adapt to changes in real time. By adjusting the exact curvature properties on the fly, the transmitter can steer signals around new obstacles as they appear, maintaining a strong connection even in crowded, constantly changing environments.
Haoze Chen, a graduate student at Princeton and the paper’s lead author said: “This is for complex indoor scenarios where you don’t have line of sight.”
“What we are doing is not only generating the beams but finding which beams work best in the situation,” he said. “People have shown that these beams can be created, but they have not shown how the beams can be optimised.
Opening the Door for Virtual Reality and Autonomous Vehicles
Transmissions in the sub-terahertz band have the potential to manage ten times the data of current wireless systems. This technology would provide great benefits for virtual reality systems and autonomous vehicles.
“As our world becomes more connected and data-hungry, the demand for wireless bandwidth is soaring. Sub-terahertz frequencies open the door to far greater speeds and capacity,” Ghasempour said.
In addition, the team designed a neural net, a computer system that mimics the brain which solves problems around finding the best curved beam in a cluttered environment. Finding the best transmission path does not work for bendable transmissions.
Neural Network Simulator Enables Smart Signal Adaptation
Co-author Atsutse Kludze designed a simulator that allowed the net to train virtually for different obstacles and different environments. The system included the underlying physics to almost any scenario.
The researchers used principles from physics to create and train the neural net. Once the system was trained, the neural net was able to adapt incredibly quickly.
“This work tackles a long-standing problem that has prevented the adoption of such high frequencies in dynamic wireless communications to date,” Ghasempour said. “With further advances, we envision transmitters that can intelligently navigate even the most complex environments, bringing ultra-fast, reliable wireless connectivity to applications that today seem out of reach - from immersive virtual reality to fully autonomous transportation.”
