| Multi-Input Multi-Output (MIMO) technology significantly improves the channel capacity and spectrum efficiency of wireless communication systems by equipping multiple antennas at both the transmitter and receiver, and thus has been recognized as one of the key techniques in current and future cellular systems. Since the MIMO channel comprises the wireless environment as well as the response of multiple antenna elements at the transceivers, the MIMO antenna equipped by the terminals has considerable influence on transmission performance. Generally, the multiple antenna elements in the MIMO antenna system should keep balanced gain and high isolation in order to guarantee a good MIMO channel performance. The MIMO antenna design for mobile terminals is usually more difficult and complicated, because the mobile terminal has limited size and should support a wide range of frequency bands, which requires the MIMO antenna system to implement multiple antenna elements in a limited space, and guarantee high isolation among different antenna elements in a wide frequency range.Over-the-Air (OTA) testing for MIMO antenna, i.e., MIMO-OTA testing, is an important approach to evaluate the MIMO antenna performance. MIMO-OTA testing emulates a specific spatial channel environment, and tests the MIMO transmission performance of the mobile device in the emulated environment. Depending on different channel emulating method, there are different MIMO-OTA methodologies, such as multi-probe anechoic chamber (MPAC) method, reverberation chamber (RC) method, and radiated two stage (RTS) method. Different methods have their respective pros and cons. MPAC method is widely accepted for certification testing, which provides accurate channel emulation and testing result. However, its complexity and system cost are much higher than other methods. RC method is a low-cost solution, but it can only emulate spatial isotropic channel model. RTS method is also a low-cost solution and supports various channel model emulation. However, its testing accuracy depends on the quality of the chipset used by the device under test (DUT). Therefore, RTS method is considered suitable for testing during the product research and development stage, rather than certification testing.In this thesis, we studied the MIMO antenna design and antenna evaluation testing methodologies for mobile devices. First, we studied the compact high-isolation MIMO antenna design for mobile devices. Then we focused on the most prevalent MPAC MIMO-OTA testing method. Considering its disadvantage of high system-cost, we performed low-complexity optimization for MPAC system. We also studied the extension from current two-dimensional (2D) chamber to three-dimensional (3D) chamber, in order to emulate three-dimensional spatial channel environment with elevation angle spread in MPAC system. Finally, we studied the low-cost RTS MIMO-OTA testing method, where we analyzed the influence of device chipset measurement accuracy on the MIMO-OTA testing results, and established and optimized a RTS testing system.The main contribution of this thesis includes,First, proposed a MIMO antenna system for notebook devices which covers both 2.4GHz and 5.2/5.8GHz bands. Bend structure and coupled-fed technique are used to reduce the antenna size, and realize the dual-band coverage. A protruded ground and embedded T-shape slot are employed to reduce the mutual coupling among different antenna elements in different bands. A two-antenna system prototype shows a mutual coupling lower than -18dB between two antenna elements on both 2.4GHz and 5.2/5.8GHz bands. The size of the whole MIMO antenna system is only 9mm x 40mm. Further,4-antenna and 8-antenna systems are designed based on the proposed two-antenna system. The 4-antenna and 8-antenna systems are able to cover the desired dual bands, and show a mutual coupling lower than-15dB.Second, established a MPAC MIMO-OTA testing system, and proposed a low-complexity optimization approach for MPAC system. We optimize the subset of probes used for channel emulation in order to minimize the number of utilized probes, under the constraint of channel emulation accuracy requirement inside the chamber. In this way, it reduced the number of required RF channel and the needed cost for channel emulator. Comparing with existing MPAC system without proposed optimization, proposed solution saves around 40% cost on channel emulator when emulating 3GPP SCME channel models, while the MIMO-OTA measurement result align well with those of existing system.Third, studied the optimization of power weights of each probe, and proposed a probe position configuration for 3D MPAC system. We first solve the optimization problem, which optimize the power weights for probes in 3D MPAC system to minimize the emulation error of spatial correlation coefficient on a 3D sphere at the center test zone of the chamber. Using the optimized probe power weights, the channel emulation error is analyzed considering different numbers of horizontal probe ring, the elevation angle of each probe ring, and the azimuth of probes on each probe ring. An optimized 3D probe configuration is then proposed based on the analysis and the characteristics of current standard 3D channel model.Finally, studied the RTS MIMO-OTA methodology, and performed channel capacity numerical simulation to analyze the influence from step size during antenna pattern measurement, the measurement error of antenna amplitude response and the relative phase response on final MIMO-OTA results. We established a RTS MIMO-OTA system, and designed a downlink power mechanism based on the device measurement reporting, which is able to cancel part of the antenna measurement error.We performed channel correlation validation for the RTS system, and conducted comparison testing between RTS and MPAC systems using real multi-antenna devices. Results showed that the established RTS system could emulate the required channel environment accurately, and is capable to distinguish the MIMO transmission performance of different devices.
|
PDF Full Text Request