Introduction
The technological evolution has allowed the creation of low-cost devices with high processing power and wireless communications. Examples of those devices are the Arduino platform, the Raspberry Pi, among others. Through sensors, these platforms acquire physical data from its surroundings and via actuators it can respond to variations of the values acquired. The same devices when applied in conventional areas make the so-called “smart spaces”. Those spaces can be used in several fields such as environmental monitoring, health care and in domotics
The system is designed to be integrated to any given scenario. In one hand when applied to the environmental monitoring, the device’s energy consumption is highly low. On the other hand, on the health care field, it includes a tracker inside of buildings, which are quite an advantage and a significant breakthrough.
System Architecture
The system itself is an open source, low-cost solution that can be adapted in many applications. The system has three devices all adapted to several scenarios. There is a personal device that collects vital parameters and it monitors the physical activity of the people who carry it. Also, an environmental device is set in order to gather data from the surroundings. It also works as a repeater. Another device gateway is used to connect the system to the Internet and it provides a bidirectional interaction with the outside world. The system manager can access to a web page where all data can be consulted and the modules can be enquired.
The Gateway XBee / GPRS start the network and work like the coordinator of the Digimesh network and create a connection to the web server. At this time it can receive and send data throw the web. All modules can then send sensor or position data.
Hardware Design
The environment device is an Arduino Diecimila ATmega168 with 16 KB of flash, 1 KB of SRAM and 512 bytes of EEPROM. Connected to Arduino we have an XBee shield. This shield is used to connect the XBee radios and the Arduino.
The Corporal Device is an Arduino Fio with an ATmega328P of 32KB flash, 2KB of SRAM and 1KB EEPROM. An XBee socket is available on the bottom of the board.
The gateway as an Arduino Mega with an ATmega1280 of 128KB flash, 8KB of SRAM and 4KB EEPROM and four serial ports. Connected to Arduino we have a GPRS shield. This shield is used to connect the GPRS radio and the Arduino
The wireless link between the devices and the gateway is established using the Digimesh protocol (IEEE 802.15.4). Our Choice was XBee-PRO® ZNet 2.5 OEM RF Module from DIGI
The wireless link between the gateway and the web server is established using GPRS. For that we used a GSM/GPRS Arduino ready module from LIBELIUM.
For temperature measurement we used an analogue temperature sensor from PHIDGETS that measures values between -40°C and 125°C.
Corporal device
The corporal device was an evolution of another the project Remote Patient Monitoring (2009, Paulo Gonçalves). In this case we used an Arduino Fio, an XBee radio, a 3 axis accelerometer and a lithium battery for power supply. Each person uses this device.
Environmental device
The Environmental device uses an Arduino Diecimilla, an XBee radio and temperature sensor. This device is inserted into a known location and he can perform three principal tasks: Sense temperature from the environment, work as a mesh router and show his network neighbours for device location.
Gateway
The Gateway uses an Arduino Mega, a GPRS module and an Xbee radio. This device connects to the electricity and work as a mesh coordinator and a bridge to the web server.
Web Server
The web server collects all the data and stores it to a MySQL database. It then deploys all the information in a web page. That Web page is used to watch all data from the environment or the patient and to ask all devices or a specific device for his sensor values or positioning. All this is performed in real-time.
Results
We have made location testing to figure out where the corporal devices were. For that we measure the neighbours of two environmental devices with known location. We also get the RSSI value of each neighbour.
| Mobile Device location | Neighbor Devices | RSSI |
| 1 | R01 – 404bfeda | -27 |
| 1 | R02 – 405c2c78 | -58 |
| 5 | R01 – 404bfeda | -38 |
| 5 | R02 – 405c2c78 | -66 |
| 8 | R01 – 404bfeda | -40 |
| 8 | R02 – 405c2c78 | -47 |
In this situation we can see that the higher the RSSI the closer we are from the devices. The mobile device moved between position 1 and 9. The table above shows the average of values obtained over the various measurements of visible neighbours in each position of the mobile device, it also shows the received signal strength values (RSSI) measured in each case. For example, the neighbours of the mobile device in position 1 are R01 and R02. The link to R01 has the value of -27dBm and to R02 of -58dBm. So, we can verify that we are closer from R01 and far away from R02. With those values we can have an idea of where is the mobile device.
Conclusion
Our system keeps the idea of not being necessary any present network in the deployment location. This shows that it can be used in any research area. This and the use of low-cost equipment help to highlight one of the proposed requirements of the project, the creation of a low-cost system.