RESEARCH PROJECT (2010-2013) - Adaptive Wireless Sensor Networks (AWSN)

RESEARCH PROJECT (2010-2013) - Adaptive Wireless Sensor Networks (AWSN)

The official website of the national research project Adaptive Wireless Sensor Networks with Data Visualization for Crises Management, MPO FR-TI2/571

The main objective of the project FR-TI2/571-Adaptive Wireless Sensor Networks with Data Visualization for Crises Management (further only AWSN) launched in half of 2010 in cooperation with the ’Satturn Holesov spol. s.r.o.’ company is to develop the adaptive wireless sensor system for the urban crises management. 

The word Adaptive in the title means that the wireless sensor system should be easily applicable to various crises scenarios in the cities such as the immediate water flood monitoring, air pollution monitoring, monitoring of the snow load on the roofs etc. The adaptivness of our system also offers the extensive over-the-air configuaration tasks. The users may arbitrary in real-time and over the air perform the following configuration tasks:

  • to turn on/off connected sensor inputs
  • to change the sensing interval of all analog and digital inputs
  • to adapt the thershold limits to the immediate needs
  • to adapt the interval, in which the report messages are transmitted
  • to add or remove the gateway from the network
  • to gather statistical information about the node state and running
  • and many more

Our effort is to design the WSN system, which is as much adaptive to the individual requirements as possible.

The workflow of the project is divided into the 7 workpackages.

WP1 (04-09/2010): Design of WSN Architecture 

At the beggining of the project, the system architecture was defined. For the communication the Zigbee PRO stack from the Atmel was chosen. All wireless devices in the AWSN system are FFD (Fully Function Devices) with the routing capabilities and stable power supply. The Zigbee sensor nodes briefly referred to as "ZS" perform monitoring tasks in the regular intervals that can be adaptively configured over-the-air by the user. The SN nodes also control the threshold levels theirs exceeds is immediatelly  reported by the ALARM messages.

Some of the ZS nodes can be equipped by the warning peripheries such as light beacon, LCD diplays for the warning of the people in the ambient environment. These nodes are in system referred to as Zigbee Sensor with Alarm, further only ZSA. Data are gathered by one of the gateways.

Data are gathered by one of the gateways. We have defined the function of the primary gateway that collests all measurements and alarms and forward them into the crises management. In case of the gateway failure, its role is substitued by the one of the backup gateway. The list of the configured gateway for the AWSN system is stored at every SN/SNA and can be configured ober-the-air by the user.

The gateways are connected to the crises management center via the arbitrary communication media, such as TCP/IP, radio network, GPRS etc. This form of this backbone is in the responsibility of the city.

The basic concept of the AWSN system for the crises management of the cities is illustrated at the following figure:

Each network is formed by the Zigbee gateway, which generates an unique PAN-ID (Personall Area Network ID). The heterogeneus networks deployed through the city are thus identified by own PAN-ID.

WP2 (09-12/2010): Research on the commercial snow load and environmental sensors

in the next phases, we have investigated some of the environmental sensors, which can be applicable to the AWSN systems. The sensors that has been tested include: snow level ultrasonic sensor, water level sensor and meteorological sensor.

Within this workpackage, the team of our partner Satturn s.r.o. designed and developed the novel lightweight snow sensor that is capable to measure the level of snow and also to distinquish between variety of snow types such as dry,wet,frozen,fresh,old and others. Some illustrations of sensor testing are provided bellow. The snow sensor is coposed of the capacitance pads that changes own capacity according to the height and the type of the snow. The snow weight is estimated indirectly using also temperature and pressure information of ambient environment. More information about the snow sensor can be found here.



WP3 (01-08/2011): Development of Zigbee Sensor Node and design of application protocol

In this phase, two communication protocols were defined. First, the AWSN protocol defines principles and messages for the overall functionalities within the Zigbee network. It aslo includes the definiton of Report messages, Alarm messages, Statistic messages and Query messages. Secondly, the protocol for data exchange between the gateway and crises management was defined. Data from the Zigbee network and encapsulated by the simple header on the basis of the serial comuniocation (message type, address, lenght, seq, data, crc) and transmitted to the crises management. The data correctness is acknowledged by the ACK/NACK messages. And opposite the query messages destined to the individual nodes are acnoeledged by the gateway and also by the Zigbee nodes. The query messages perform tasks such as configuration of the report interval, threshold limits, list of gateways and many more. The detailed description of the proposed AWSN protocol will be later published. Bellow is photo taken during the protocol concept discussion.

Within this workpackage, the Zigbee Sensor was developed. The ZS node is controlled by the Zigbit chip which can be "adaptively" changed according demanded requency band. User can assembly Zigbit for 2,4 GHz or 868 MHz band according to the actual channels occupancy. The ZS node offers four analog inputs, two digital binary inputs and outputs, two USART ports and I2C port. Photographies of the  Zigbee Nodes  can be seen here:

The paper with the results of the system validation is now submitted to the research journal. Once is accepted, the link to the journal will be provided here.

Bellow you can find photo of our colleagues and their poster, which was presented at EWSN 2012 conference in Trento, Italy. At this conference, the novel packet analyser for 802.15.4 networks was presented. See this link for more details about our packet analyser. 

WP4 (08-12/2011): System Configuration and Visualization

This phase covered the development of the set of softwares allowing user friendly configuration of nodes and visualization of gathered data.

The main architecture of system management is illustrated in the Figure bellow. As one can see, the Wireless Sensor Network can be configured manually through the serial comunication or it can be also configured remotly through the internet. The local configuration is useful when the nodes are being deployed on the site, while the remote control can be used when the system is already running and user wishes to configure the individuals nodes from its office. 

Through the configuration application, the user can configure the network parameters as well as sensing parameters of the individual nodes. The configuration commands can be send directly to one node, to the set of nodes at once or can be broadcasted to all nodes in the systems. The printscreen of the configuration application is depicted bellow. 

The data visualization software allows an indoor, outdoor and topology view of measured data, see Figure bellow:

Indoor view:

Topology view  

WP5 (01-08/2012): Investigation of system performance

Through the simulations, we have covered the issue of the energy efficiency and the investigation of the fundamental communicatin metrics such as delay and message lost. The results will be published in some of the coming conferences.

In the first half of 2012, we have conducted the BitCloud experiment with the 25 IRIS nodes. The goal of this experiment was to evaluate the performance of the Zigbee implementation from the Atmel. The main metrics that were investigated were latency, packet loss, network startup and rerouting, overload of security implementation and some others. Results from this experiment are published in the journal of Wireless Personal Communications.

Comprehensive Performance Analysis of ZigBee Technology Based on Real Measurementsdownload link here.

zigbee performance evaluation, real measurements, zigbee PRO benchmark, zigbee throughput, zigbee self-healing, hidden terminal evaluation, zgbee network startup, zigbee routing latency, zigbee mesh routing

  • Hardware
  • Research project
  • WSN
24. September 2012 by Milan Šimek, Ph.D.