The report summarizes the various techniques and feature engineering processes that we have applied for the Far Eastern Memorial Hospital (FEMH) Voice Data Challenge. We have used Mel scaled spectrograms and MFCC components as audio features to train various Neural Network Architectures. We have trained a 5-layer plain network, 5-layer CNN and RNN. We discuss the challenges faced and solutions to improve model performance, model parameter tuning and model evaluation.

We demonstrate interoperable mobile agents for language- and platform-independent wireless sensor network programming, with low-power resource-constrained embedded devices as static nodes and Android-based smartphones as mobile nodes, over disparate networks: 6LoWPAN and Wi-Fi. Representational state transfer architectural principles are applied in agent composition, control and migration and exposing the system resources to the Web: the devices, agents, sensor data, tasks and data processing results. The adaptable agent composition includes the task code in any programming language, the agent migrates according to a resource list and the state, i.e. the intermediate task result, represents the agent as a system resource. Mobile agents are then utilized in two tasks: first to collect light sensor data cooperatively in location, saving the mobile node battery whenever possible, and secondly the magnetometer sensor data with the scanned Wi-Fi access points' signal strength is used to detect groups of mobile nodes moving in the same direction in real-time.

We demonstrate interoperable mobile agents to integrate Internet of Things and wireless sensor networks with resource-constrained low-power embedded networked devices. We introduce adaptable composition for the mobile agent, complying with the Representational State Transfer principles, which are then used for agent migration, controlling the agent and exposing the data, system resources, tasks and services, to the Web. We gather requirements for the system and heterogeneous networked devices and present an application programming interface to enable mobile agents in these systems. The agents are demonstrated in a real-world prototype with smartphones and embedded networked devices, where we utilize a proxy component to expose system resources to the Web for human-machine interactions. Spanning over disparate networks and protocols, the proxy translates messages including the agent composition, between HTTP and Constrained Application Protocol. Lastly, we suggest an evaluation method for the agent communication and migration costs, considering the different types of system resources and utilization.

Machine-to-Machine (M2M) communications emerge to achieve ubiquitous communications among all the networked devices. One challenging problem in M2M applications is to discover the resources provided by the devices efficiently. This challenge arises from two perspectives: (1) a large number of heterogeneous devices co-exist and most of them are constrained devices (e.g., limited processing capability), (2) different communication protocols are utilized to get access to different types of resources provided by the devices (e.g., humidity information provided by a sensor). This paper proposes distributed resource directory architecture for M2M applications, referred to as DRD4M. The DRD4M supports heterogeneous devices using HTTP and CoAP protocols for resource registration and lookup. This enables the interoperability among heterogeneous devices and resource access to constrained devices from disparate network including the Internet. The DRD4M introduces two components: a resource registration component for registering resources and a resource lookup component with caching functionality for handling resource lookup with filtering. A peer-to-peer (P2P) overlay is introduced in DRD4M to connect resource peers to avoid single point of failure. A real-world prototype is implemented and is verified with a demo application. Preliminary performance evaluation in terms of response time of resource lookup is provided.