An aging population is one common challenge faced by many developed countries including Australia. The Australian government has realised that the existing healthcare system must be improved to provide better support longer-term for the healthcare needs of this population. This research examines one such opportunity by suggesting a reform on how the care needs of residents living in Australian residential aged care facilities (RACF) are assessed. A recent study has shown that the current assessment system, known as the Residential Classification Scale (RCS), is subjected to high administrative procedural overhead costs and significant deviations in assessment results. This thesis documents a technical feasibility study of a novel method aimed to solve issues related to the time demands and subjectivity of the RCS through the design and implementation of a Wireless Sensor Network (WSN). This WSN is engineered to unobtrusively collect data from wireless sensor nodes either embedded in the RACF environment or attached to the resident??s body. The collected data can be potentially used to provide automatic and accurate care level assessments for the resident. The methodology of preparing and conducting the experiments to prove the hypotheses is justified and described, including the experimental instruments and procedures involved. The results show that this WSN surpasses similar research systems in terms of its application scale, the number and types of sensor nodes involved and the complexity of its hardware and firmware architectures. The major contributions of this thesis are: ?? The WSN developed satisfies certain technical requirements to be declared fit for use in a mock Australian RACF. ?? The WSN provides high sensor detection accuracies (between 88% and 100%), superior location tracking capability (94.75%) and activities of daily living inference capability over similar studies. Opportunities for further improvements of this WSN include: ?? Fine tuning the detection accuracy of Passive Infra-red (PIR) motion sensors. ?? Minimising the down time of the sensor nodes due to firmware memory leak. ?? An extra location tracking mechanism to improve location accuracy determination.

An aging population is one common challenge faced by many developed countries including Australia. The Australian government has realised that the existing healthcare system must be improved to provide better support longer-term for the healthcare needs of this population. This research examines one such opportunity by suggesting a reform on how the care needs of residents living in Australian residential aged care facilities (RACF) are assessed. A recent study has shown that the current assessment system, known as the Residential Classification Scale (RCS), is subjected to high administrative procedural overhead costs and significant deviations in assessment results. This thesis documents a technical feasibility study of a novel method aimed to solve issues related to the time demands and subjectivity of the RCS through the design and implementation of a Wireless Sensor Network (WSN). This WSN is engineered to unobtrusively collect data from wireless sensor nodes either embedded in the RACF environment or attached to the resident??s body. The collected data can be potentially used to provide automatic and accurate care level assessments for the resident. The methodology of preparing and conducting the experiments to prove the hypotheses is justified and described, including the experimental instruments and procedures involved. The results show that this WSN surpasses similar research systems in terms of its application scale, the number and types of sensor nodes involved and the complexity of its hardware and firmware architectures. The major contributions of this thesis are: ?? The WSN developed satisfies certain technical requirements to be declared fit for use in a mock Australian RACF. ?? The WSN provides high sensor detection accuracies (between 88% and 100%), superior location tracking capability (94.75%) and activities of daily living inference capability over similar studies. Opportunities for further improvements of this WSN include: ?? Fine tuning the detection accuracy of Passive Infra-red (PIR) motion sensors. ?? Minimising the down time of the sensor nodes due to firmware memory leak. ?? An extra location tracking mechanism to improve location accuracy determination.