The wireless sensor network is comprised of many sensor devices, each of which communicates and operates with its neighbors by wireless transmission. Each sensor device includes a microcontroller unit, a memory unit and an RF chip. The sensor device is responsible for measuring physiological signals, processing the measured data, and then sending the measured data to a specific device wirelessly. Since the sensor device with the characteristic of low power consumption is used to perform long-term monitoring, we must survey the current wireless technology and select the best candidate for low power consumption. According to our requirements, we have chosen the ZigBee wireless transmission standard [1], which is a wireless personal area network that creates a wireless sensor network of tree topology using hierarchy routing with web-based management for usage convenience.
1.1. Motivation
As the population of elderly people has increased in recent decades, the number of persons aged 60 is projected to be almost two billion by 2050 [2]. Thus heath related issues are becoming more and more important for this population. To increase medical quality, home health-care is a useful solution to achieve long-term monitoring of chronic disease effectively. The wiring of patients to obtain physiological signals has many drawbacks, including high cost of developments and maintenance.
However, such drawbacks are lessened using wireless instead of wired connections to allow patients to move freely while carrying only a few small nodes. Now, in this study we propose a wireless sensor network solution that can support mobility and flexibility of sensor nodes in a network, thus providing many advantages while replacing wired with wireless logic links. Since advancement in wireless communication, sensor technology and information technology in general provides opportunities to in the area of home health-care service, which is very promising application to improve the quality of life.
The home-care service usually provides real-life long term monitoring of a patient’s health that is useful for the assessment of treatments at home. This study focuses on the design and implementation of a wireless sensor network that has a tree topology of three levels using hierarchical routing with
web-based management. The wireless sensor network has many advantages, including low power consumption and high addressing space. The wireless sensor network is capable of formatting a network automatically, accepting several sensor devices and collecting data from existing devices in the network.
Many studies [3] [4] [5] of home health-care issues use Bluetooth as the wireless personal area network. However, Bluetooth has many drawbacks, including limited addressing capability, higher power consumption than ZigBee and longer waking time than ZigBee.
The wireless sensor network can be used for collecting the necessary physiological signals of a patient, which are used to determine existing health problems and help predict such problems in the future. Because wireless sensor networks have a very large number of sensor nodes, their management and querying is a very troublesome problem. This study proposes a solution to this, which is to implement web-base management. This has the advantages of flexibility, ease of development and convenience of access. To determine existing health problems and help predict such problems in the future, we must measure the physiological signal of a patient. This study first implements an electrocardiogram (ECG) senor, since the ECG provides signals to monitor the most vital biological processes of the patient. Secondly, we select a commercial SpO2 sensor that provides oximeter information containing SpO2 and the heart rate.
1.2. Review of the Literature
Recently, in health-care related research Tura et al. [4] have focused on the implementation of a network for managing home-care activities. The device of this network can measure blood oxygen saturation, the heart rate, the respiration rate and the patient’s quantity of movement. Measured data are stored into a multimedia card and then transmitted to a PC through Bluetooth. At the PC end the measured data is finally transmitted to the server center through the internet. The device uses a 16-bit microcontroller of the Mitsubishi M16C/62 family. There are many reasons for choosing this micro-controller, including compact design, smaller board dimensions and lower power consumption.
For sensors they use the OEM pulse oximeter board, which is the NONIN OEM П module, to measure plenty of physiological signals of blood oxygen saturation and heart rate. The respiration rate is
measured using the Respiratory Effort Monitoring System (RESP-EZ) from the EPM systems module.
The patient’s quantity of movement is measured through a dual axis thermal accelerometer called MEMSIC MXD205.
In terms of the hardware design for the wireless sensor, some papers [6] [7] have focused on integrating the hardware and communication interface such that each sensor device can communicate with other sensor devices and enter various physiological values into a database.
For network topology some studies have focused on the constructing suitable network topology for physiological signal transmission. One study focused on the implementation of wireless sensors network using Bluetooth, which consists of sensor nodes, relay nodes and a control node. It allowed all nodes to communicate with each other with a Bluetooth module, sensor and relay nodes detecting certain events and reporting the events to the control node. Its network is a tree topology based on Piconet where each Piconet has one master and up to seven slaves.
Korhonen et al. [8] proposed methods to implement health monitoring in homes of the future home. These methods included models of remote monitoring, comparing wearable sensors with environmental sensors, system architecture that contained sensor devices, a personal gateway or home gateway that was a mobile device, such as mobile phone used to transfer data from a sensor device to a remote database through the internet. In this study the authors presented some useful models for health monitoring and discussed the technical requirements for the health monitoring system based on wearable and ambient sensors, which measured health related data in the daily environment of the patient. To make these methods more affordable, health-monitoring needed to support the use of the generic platforms such as Bluetooth or ZigBee. The gateway used generic hardware such as a mobile phone or Personal Digital Assistant (PDA) as a personal gateway or a Digital TV as a home gateway.
Since the wireless sensor network is comprised of a very large number of sensor nodes, their management and querying of the wireless sensor network was a troublesome problem. Web-based management was useful solution for management and querying of the wireless sensor network. Hwang et al. [9] implemented a wireless sensor gateway that was responsible for processing a user’s query from
a web page and transferring the user’s query into a sensor query. Finally the wireless sensor gateway sent the sensor query to the wireless sensor network that was generated by a sensor network emulator.
Tan et al. [3] focused on the design and implementation of a home health-care system using Bluetooth, web service and Global System for Mobile communication (GSM) short message service. In this study the authors implemented a mobile ECG unit that comprised a sensor, transmitter and receiver.
The ECG physiological signal was passed to a remote database of the health web server unit based on the SOAP for diagnosis through the internet. This study used the web service to enhance the flexibility of architecture and convenience of usage, which improved medical quality and reduced medical costs.
A system called Telemedicine [10] has been developed to improve the quality of home-based care and medical treatment. The system supports 24-hour real-time monitoring in a patient’s home by reliable sensors, which are responsible for measuring the physiological signals of the patient and sending this data to the patient’s computer through wireless communication. The computer analyzes and stores the data.
For medical care, a project called CodeBlue [10] was developed by the division of engineering and applied science at Harvard University. The project explored the application of a wireless sensor network to a range of medical applications including pre-hospital, in-hospital emergency care, disaster response and stroke patient rehabilitation. The project integrated medical sensors with lower power wireless networks, which used wireless ad-hoc routing protocols for critical care, security, robustness and prioritization. The hardware architecture of the project focused on many aspects, including ultra-lower-power sensing, computation and communication. The project also implemented 3D location tracking using radio signal information and adaptive resource management, congestion control and bandwidth allocation in the wireless sensor network.
Another project [10] has been initiated to integrate body sensors with the internet. Sensors were attached to the body of a patient and connected with other sensor nodes through wireless transmission.
The wireless body sensor network comprised several body sensor units and one body control unit. The body control unit was responsible for collecting data from all body sensor units and sending this data to
the internet. Body sensor units were responsible for measuring the necessary physiological signal of patients. This data could be used for disease diagnosis or disease prediction. The project was suitable for hospital applications, home care services, healthcare centers and sports medicine.
1.3. Objective
Home health-care is a useful solution to improve medical quality by using wireless technology to transmit physiological signals to a hospital. The hospital uses this data to do long-term monitoring.
Doctors can use this data to make diagnoses or predict future diseases. To implement the home health-care system for multiple users, the wireless sensor network is need. The wireless sensor network is comprised of many sensor nodes, and has lower power consumption, high addressing capability and low transmission latency.
1.4. Thesis Organization
This study is organized as follows. Chapter 2 briefly introduces the rationale of ECG and SpO2
and describes implementation methods for ECG sensor and SpO2 sensors. Chapter 3 illustrates the implementation methods of hardware and firmware for the wireless sensor network. Chapter 4 describes the design method of web-based management. In Chapter 5 we show experimental results of the whole system. These experimental results include the wireless sensor network and the web-based management system. Chapter 6 compares the wireless sensor network in this study with existing research, and describes related advanced applications. Chapter 7, the final chapter of this study, provides some conclusions and deals with another great challenge of this area of research. We also provide directions of future research for the extended applications.