Optimized network layer protocol for sensor driven autonomous real-time data fusion in a star BSN

Abstract

To achieve real time data fusion and multivariate signal processing on sensor level we developed a slim network layer protocol (iTALK) based on TIMAC®. Defining a star topology Body Sensor Network (BSN) iTALK enables node-node communication initiated by the intelligent sensor nodes. Thus real time data fusion and feedback control applications can be realized. Hereby our recently developed ‘intelligent Network operating Devices’ (iNODE) [1] serve as clients and an enhancement of them using Texas Instruments microcontroller derivate MSP430F2618 and the CC2520 RF device (IEEE 802.15.4) serves as the coordinator (cNODE ~ ‘coordinating Network Operating Devices’). This concept ensures, that the sensor nodes still cope with the ‘Ultra Small Stack’ for RF-communication [1] and can focus their resources on digital signal processing while recording up to eight channels with a sample rate up to 8kHz on local data storage. To adapt the sensor network to changing external conditions the protocol allows reconfiguration of the sensors during measurements being initiated by sensor nodes themselves, by the network coordinator, or by an external expert. A multithreaded PC application providing an intuitive Graphical User Interface (GUI) and asynchronous communication feasibility ensures comfortable external access to the network. One application we aim on is the long term monitoring of the disturbed coordination between locomotion and respiration in Parkinson Disease (PD) patients during rehabilitation. Real time tactile feedback stimulation shall support relearning this coordination to improve the rehabilitation progress.