Antenna in package (AiP): disrupting wireless communication and HMI

An article written by Stephan Elisabeth, System Plus Consulting and Cédric Malaquin, Yole Développement, for Chip Scale Review – Any wireless system requires antennas to convert the electric energy into radio waves traveling through the air. Whether it is for wireless sensing or broadband communication between at least two points, multiple types of antennas have been developed depending on the targeted application. Monopole, dipole, yagi, dish, patch, etc., are some of the existing configurations. Recently, a new type of system configuration, called antenna in package (AiP), has appeared and is becoming more and more popular, at least in the consumer market. In this article we will review the market drivers and the state of the art, as well as the market potential for AiP.

Advanced packaging: innovation for consumer connectivity apps

In the consumer market, the main technology deployed for cellular and WiFi connectivity is based on relatively low frequencies, the so-called sub6GHz frequency range. Because of the sparse and poor spectrum holding from network providers, complex technologies such as carrier aggregation and multipleinput multiple-output (MIMO) have been deployed to comply with the never-ending increasing demand for data consumption. In the meantime, RF board allocation has shrunk in the mobile handset because of restricted space with a larger battery size and new features added. This strategy has put pressure on RF front-end module makers and outsourced semiconductor assembly and test (OSAT) companies to develop very complex packaging technologies. An example of dense RF component integration is shown in Figure 1.

The Apple iPhone® Xr features one or another module for dual-sourcing reasons (shown in Figure 1). Both modules are pinto-pin compatible and include all necessary RF components: acoustic wave filters (in red), RF switches (in green), a power amplifier (in purple), a low-noise amplifier (in yellow) and a power management integrated circuit (PMIC) (in blue). Because two different frequency domains (mid band and high band) are utilized in this device, complex electromagnetic interference (EMI) shielding techniques have been employed to isolate the different bands. In Qorvo’s module, EMI shielding is using the package laminate substrate with a ground trace, whereas Broadcom uses grounded palladium-coated copper (PPC) wire bonded onto critical components to be protected. As RF component integration density is still increasing, the latest RF system-in-package (SiP) modules found in the Apple iPhone® 11 now come in double-sided ball grid array (BGA) packaging with EMI shielding. For low-end smartphones, where cost is more important than the form factor, the mainstream packaging is SiP using long grid array (LGA).

In fact, the iPhone® 11 still is an advanced LTE phone. With the 5G roll out, the densification trend of RF components in a mobile handset will extend to sub-6GHz frequencies. But the story is different for 5G mmWave applications. In the sub-6GHz world, antenna integration is not possible on account of the size and location of the antenna. Instead, antennas are connected to the modules with a coaxial cable, printed circuit board (PCB) or flex PCB. A basic requirement for the consumer market is that, whenever possible, antenna size needs to be shrunk down to comply with the stringent system integrator requirements. For sub-6GHz frequencies, that means an antenna tuner is increasingly becoming a key component. At millimeter wave frequencies, however, AiP and antenna-on-package (AoP) comes into play. At millimeter-wave frequencies, antenna size reaches the RF integrated circuit’s (RFIC) form factor. In addition, RF losses are such that integrating the antenna along with the RFIC is no longer an option, it is now mandatory… Full story