cations in the complex wave propagation environment. Similar radiation patterns for other frequencies over the operating band were also seen, and the results are not shown for brevity. Measured peak antenna gain and simulated (Ansoft HFSS) radiation efficiency versus frequency is shown in Figure 4. The measured antenna gain varies in a small range of about 1.4 -2.2 dBi for frequencies across the bandwidth. The radiation efficiency for frequencies in the 2.4/2.5 GHz bands varies from about 57 to 67% from the simulated results.
Effects of the system ground size (L and W) on the impedance matching were studied with the aid of Ansoft HFSS. Figure 5 shows the simulated return loss of the system ground-plane length L varied from 70 to 100 mm. Small variations in the achievable impedance bandwidth are seen. For the cases of the system ground-plane width W varied from 40 to 70 mm, the results are shown in Figure 6. Again, small variations in the simulated return loss are obtained. These results indicate that the proposed AiSiP can be applied in mobile communication devices with various possible system ground-plane or PCB dimensions.
The simulated surface current distributions in the proposed design obtained from SEMCAD (simulation platform for electromagnetic compatibility, antenna design and dosimetry) [15] are shown in Figure 7. The major excited surface current distributions at 2545 MHz are found to be in the integrated antenna and two adjacent side surfaces of the metal case next to the antenna. Relatively small surface current distributions are seen to be in the system ground plane. This phenomenon suggests that the proposed AiSiP will be less prone to the effects of the system ground-plane size on the antenna impedance bandwidth. That is, the results confirm that the achievable impedance bandwidth is almost not affected by the use of various system ground-plane dimensions as studied in Figures 5 and6.