Designing MIMO Antennas for RF Front Ends and Wireless ASICs

PhysWAVE’s flexibility enables antenna design and coupling analysis for integrated chip-frontend-antenna systems

Wireless ASICs, RF Frontends for data and voice transmission, GPS subsystems, and RFID tags and systems require integrated antenna sub-systems. These antennas need to provide multiple transmit-receive, multi-input multi-output, and possibly frequency-reuse functionality.  

The antennas need to function in the presence of non-ideal environments and a background of EMI and crosstalk emanating from the system. In addition, antennas may couple back into elements of GDS and RDL layers, packages, and sections of the board.  

While several stand-alone antenna simulation tools exist, what is required, for an effective solution, is a single-suite solution that allows for modeling of antennas in the same framework as SI-PI-EMI-SNI simulation for chip-through-system.  

PhysWAVE, through its flexibility and scalability, makes this possible today.  

Tackling SI-PI- EMI-SNI (Simultaneous Noise Integrity) issues involving RDL and GDS layers, complete packages, sections of board, and embedded passive and decoupling capacitors is non-trivial. PhysWAVE makes this easier for the designer through flexibility, scaling, accuracy, and efficiency of underlying 3d accelerated    boundary element

technology. This technology also provides sufficient flexibility to model integrated antennas within the same complex system environments. Users can therefore use one tool environment to examine both undesired (EMI / EMC) and desired (antenna patterns) crosstalk, impedance mismatches, and radiation.  Importantly, the antenna

patterns are not idealized but are those produced in the presence of the integrated system. Therefore effects such as multi-antenna crosstalk, interferences from signal lines, power noise coupling, non-ideal and finite grounds, and non-planarities are sufficiently accounted for.

In low-cost wireless ASICs, such as those that are ubiquitous in multiple tether-free applications involving sensors and accelerometers in consumer electronics, health and fitness applications, and automotive markets, there is a corresponding need for low-cost effective antennas integrated into the ASIC system. An example of how PhysWAVE can effectively design and verify such antennas is shown in Fig 1, where a single patch antenna, designed to operate at the cellular frequency of 2.4 GHz is developed and validated against measurement.

PhysWAVE can also be used to design other simple antennas such as folded dipoles for wireless ASIC and RFID applications with ease. The patch antenna design and verification, utilizing PhysWAVE, takes only about 0.82 mins per frequency and 12 mins total turnaround time including model generation and parametrization / optimization.

In complex RF applications, more directive and steerable antenna patterns are often desired. Such applications include collision avoidance radar, last mile wireless delivery systems, wireless HDMI, DTV, RF transceivers for data and voice, automotive and aircraft wireless internet, etc. As an example, Fig 2 shows a steerable 4 X 4 patch antenna array designed to function at a 20 GHz frequency with an impedance bandwidth of 625 MHz (3.1 %). The fabricated prototype is shown on the left, and the PhysWAVE model is shown next to it. The model can be excited from microstrip ports or connected directly to exciting traces from the associated sub-system.

PhysWAVE’s proprietary near-linear scaling technology enables the entire antenna, with no periodic or infinite plan assumptions, to be modeled in 125 mins. The reflecting S-parameter is depicted along with the array pattern showing a strong main lobe and sidelobes.

In conclusion, PhysWAVE has the requisite flexibility to not only enable SI-PI-EMI-SNI simulations of RDL/GDS-package-board sections but also to predict radiation from simple and complex antennas associated with such systems arising in RF frontends and wireless ASICs. To learn more about how PhysWAVE can help you today, visit www.physware.com.