Testing plays an important role in the wireless protocol stack development. In order to free the testers out of the shielded chamber, allow both the developers and the testers to use the test systems remotely and maximize the expensive test system usage. An automated test framework is highly demanded. In this thesis, the design of the automated test framework is introduced. There are four main components in the test framework. They are the front end, scheduler, test engine and data storage. The architecture and the protocol among these components are described. Further, the evaluation of the scheduler is conducted based on the queueing theory. Based on the simulation result, a good scheduling algorithm is proposed. Compared with the original scheduling algorithm, the new algorithm improves the performance of the low priority users significantly when the test systems are limited. Moreover, the detail design of the test engine is presented. With the control of the intelligent test engine, the automated test framework has the capability to launch the test cases automatically, catch the commands sent by the test system and manipulate the SUT (System Under Test) without human’s interrupt. It fulfills the objective of automation. The automated test framework has been deployed and is working well.

Advances in wireless technologies in the last ten years have created considerable opportunities as well as challenges for wireless body medical systems. The foremost challenge is how to build a reliable system connecting heterogeneous body sensors and actuators in an open system environment. In this dissertation, we present our work towards this goal. The system addresses four design issues: the underlying network architecture, the network scheduling disciplines, the location determination and tracking methods, and the embedded application execution architecture. We first present the design of an adaptive wireless protocol (MBStarPlus) to provide the basic wireless platform WBAN (Wireless Body Area Network). MBStarPlus is a real-time, secure, robust and flexible wireless network architecture. It is designed to utilize any low-power wireless radio as its physical layer. The TDMA mechanism is adopted for realtime data delivery. The time-slot length is adjustable for flexibility. Multiple technologies are utilized to provide reliability and security. We next investigate how to coordinate the body sensors/actuators that can optimally select from a range (maximum and minimum) of data rates. Two bandwidth scheduling algorithms are proposed. One is a greedy algorithm that works for sensors with limited computational capability. The other is the UMinMax scheduling algorithm that has better scalability and power-saving performance but is more computationally intensive. The third issue addressed in this proposal is how to achieve location determination and tracking by a mix of high-precision but expensive and lower-precision but cost-effective sensors. This is achieved by a novel collaborative location determination scheme ColLoc that can integrate different types of distance measurements into a location estimation algorithm by weighing them according to their precision levels. Through this, a localization service can be both cost-effective and sufficiently accurate. Fourth, in order to minimize the effects of long network latency when the body network scales up, we propose ControlInGateway, an architectural feature that allows a control application to be executed inside the network gateway without the host's involvement. With ControlInGateway, a wireless system could achieve the same control quality as a wired system. / text