Abstract This dissertation aimed at developing a multi-modal sensing u-healthcare system (MSUS), which reflects the unique properties of a healthcare application in a wireless sensor network. Together with health parameters, such as ECG, SpO2 and blood pressure, the system also transfers context-aware data, including activity, position and tracking data, in a wireless sensor network environment at home or in a hospital. Since packet loss may have fatal consequences for patients, health-related data are more critical than most other types of monitoring data. Thus, compared to environmental, agricultural or industrial monitoring, healthcare monitoring in a wireless environment imposes different requirements and priorities. These include heavy data traffic with wavelike parameters in wireless sensor network and fatal data loss due to the traffic. To ensure reliable data transfer in a wireless sensor network, this research placed special emphasis on the optimization of sampling rate, packet length and transmission rate, and on the traffic reduction method. To improve the reliability and accuracy of diagnosis, the u-healthcare system also collects context-aware information on the user’s activity and location and provides real-time tracking. Waveform health parameters, such as ECG, are normally sampled in the 100 to 400 Hz range according to the monitoring purpose. This type of waveform data may incur a heavy burden in wireless communication. To reduce wireless traffic between the sensor nodes and the gateway node, the system utilizes on-site ECG analysis implemented on the sensor nodes as well as query architecture. A 3D VRML viewer was also developed for the realistic monitoring of the user’s moving path and location. Two communication methods, an 802.15.4-based wireless sensor network and a CDMA cellular network are used by sensors placed on the users’ bodies to gather medical data, which is then transmitted to a server PC at home or in the hospital, depending on whether the sensor is within or outside the range of the wireless sensor network.

Multiple-input/multiple-output (MIMO) systems with the adaptive array processing technique, also referred to as smart antennas, have received extensive attention in wireless communications due to their ability to combat multipath fading and co-channel interference, two major channel impairments that degrade system performance. However, when smart antennas are deployed in wireless networks, careful attention is required since any defective or imperfect operation of smart antennas can severely degrade the performance of the entire network. Therefore, the evaluation of network performance under ideal and imperfect conditions is critical in the process of system design and should precede deploying smart antennas on the wireless network. This work focuses on the development of an analytical framework to evaluate the performance of wireless networks based on popular DS/CDMA cellular systems equipped with antenna arrays. Spatial diversity at both the base station (BS) and the mobile station (MS) is investigated through both analytical analysis and simulation. The main contribution of this research is to provide a comprehensive analytical framework for examining the system level performance with multiple antennas at both the BS and the MS. Using the framework developed in this research, system capacity and coverage of the uplink (or reverse link) are investigated when antenna arrays are implemented at both the BS and the MS. In addition, the system capacity and soft handoff capability of the downlink (or forward link) are examined taking into account MIMO. Furthermore, various physical and upper layer parameters that can affect the system level performance are taken into account in the analytical framework and their combined impact is evaluated. Finally, to validate the analytical analysis results, a system level simulator is developed and selective results are provided. / Ph. D.