DARPA-funded UCSD research yields world's most complex phased array

Published 1 November 2007

Tritons researchers develop world’s most complex phased array; the 16-element chip is just 3.2 by 2.6 square millimeters, can send at 30-50 GHz

University of California-San Diego (UCSD) electrical engineers have developed the world’s most complex phased array, that is, radio frequency integrated circuit. This DARPA-funded breakthrough will likely find its way into U.S. defense satellite communication and radar systems. The innovations in this chip design will likely also spill over into commercial applications such as automotive satellite systems for direct broadcast TV, and new methods for high speed wireless data transfer. “This is the first 16 element phased array chip that can send at 30-50 GHz. The uniformity and low coupling between the elements, the low current consumption and the small size — it is just 3.2 by 2.6 square millimeters — are all unprecedented. As a whole system, there are many many firsts,” said Gabriel Rebeiz, electrical engineering professor from the UCSD Jacobs School of Engineering and leader of the project. The work was done by two graduate students, Kwang-Jin Koh and Jason May, both at the Electrical and Computer Engineering Department (ECE) at UCSD. Rebeiz presented the new chip at a 28-29 August DARPA TEAM meeting in Chicago, Illinois. This chip — the UCSD DARPA Smart Q-Band 4x4 Array Transmitter — is strictly a transmitter. “We are working on a chip that can do a transmit and receive function,” said Rebeiz. “This compact beamforming chip will enable a breakthrough in size, weight, performance and cost in next-generation phased arrays for millimeter-wave military sensor and communication systems,” DARPA officials said.

The chip contains all the CMOS digital circuits necessary for complete digital control of the phased array, and was done using the commercial Jazz SBC18HX process. This is a first and it greatly reduces the fabrication complexity of the phased array. The chip has been designed for use at the defense satellite communications frequency - the Q-band — which goes from 40 to 50 GHz. “If you take the same design and move it to the 24 or 60 GHz range, you can use it for commercial terrestrial communications,” said Rebeiz who is also a lead on a separate project, funded by Intel and a UC-Discovery Grant, to create silicon CMOS phased array chips that could be embedded into laptops and serve as high speed data transfer tools. The Intel project is a collaboration between Rebeiz, Larry Larson and Ian Galton — all electrical engineering professors at the UCSD Jacobs School of Engineering. Larson also serves as the chair of the Department of Electrical and Computer Engineering.

Phased arrays

Phased arrays have been around since the middle of the last century. They are groups of antennas in which the relative phases of the signals which feed them are varied so that the effective radiation pattern of the array is reinforced in a particular direction and suppressed in undesired directions. This property, combined with the fact that radio waves can pass through clouds and most other materials that stymie optical communication systems, has led engineers to use phased arrays for satellite communications and for detecting incoming airplanes, ships, and missiles. Some phased arrays are larger than highway billboards. The most powerful among them, used as sophisticated radar, surveillance, and communications systems for military aircraft and ships, can cost hundreds of millions of dollars. The high cost has prevented significant spread beyond military and high-end satellite communication applications. Engineers are now working to miniaturize phased arrays and fully integrate them into silicon-based electronic systems for both military and commercial applications.

The new UCSD chip packs sixteen channels into a 3.2 by 2.6 mm² chip. The input signal is divided on-chip into sixteen different paths with equal amplitude and phase using an innovative design, and the phase and gain of each of the sixteen channels is controlled electronically to direct the antenna pattern (beam) into a specific direction. By manipulating the phase, it is possible to steer the beam electronically in nanoseconds. With the amplitude, one can control the width of the beam, which is critical, for example, when information is sent to or from one satellite to another, but when the sender does not want the signal to reach any nearby satellites. With amplitude and phase control, you can synthesize deep nulls in the antenna pattern so as greatly to reduce the effect of interfering signals from neighboring transmitters.