and disaster situations for mines and tunnels (Ramulu M. 2009) where the focus was on explosions effects on the infrastructure (Peng S.J. 2012). Some of them had their area of interest focused on WSN architecture for disaster survivor’s detection (Saha S. 2007) and others on earthquake detection and disaster reduction using Wireless Sensor and Actuator Networks (WSAN) based on simulations but with certain drawbacks (Rahman M. 2016).
To the best of our knowledge, we are the first to propose an adaptive location based routing mechanism for emergency communication in disaster scenarios, especially earthquakes or snow slide, where survivors are trapped underneath ruins with sensors.
The rest of the paper is organized as follows: related work is discussed in section 2. Elements of the proposed service and tool are in section 3, including the architecture of the message system, the communication mechanism is in section 4.
Finally, conclusions and possible perspectives of this work are presented in section 5.
Related work
M. Erd et al. work, focus on time-critical events monitoring using wireless sensor networks. A system of energy-efficient wireless sensor nodes has been developed to acquire information about the state of the infrastructure in the event of a disaster. The wireless sensor nodes were positioned in strategic locations to sense possible explosions occurring in the construction. The acquired data is transmitted wirelessly to a central unit, where emergency services were triggered. Data from wireless sensor nodes offers the firefighters and the related staff accurate information about the infrastructure damages and the possibility to detect the building breakdown situation (Erd M. 2016).
N. Ahmad and al suggest an ad hoc wireless sensor network architecture for disaster survivor detection which is an enhancement of previous existing network models (Kone C.T. 2010; Silva R.I.D. 2010), with the integration of Telemedicine-based systems. Ad hoc relay stations oversee communicating the data precisely to sink node, in the case of disasters, particularly when the disaster situation happens the base stations are unreachable or destroyed, due to some shortcoming in the existing network models (Naveed A. 2011).
H. Afzaal and N.A Zafar focused on early prediction of earthquakes minimizing destruction caused by the earthquakes using Wireless Sensor and Actor Networks (WSANs). WSANs are deployed in a subnets system for more energy efficiency on the areas in which there is a more probability of damage caused by the earthquake.
The actors are used for reporting the predicted earthquake information to a gateway and performing action on the environment to save lives. The gateway communicates with the base station so that the information is circulated towards people quickly (Afzaal H. 2016).
DistressNet (Stephen M.G. 2010) is an extensible, scalable, heterogeneous network that minimizes congestion using multi-channel communications and has a distributed collaborative sensing and robust, adaptive localization and location-based services that provide accurate detailed data that enhances decision support. It is proposed as a solution for the lack of communications infrastructure faced in larger disasters scenarios. Where entire regions suffer from degraded communications, and the remaining capacity is exhausted by the demands of victims. It is being implemented using custom and commodity sensors, mobile and static gateways capable of cross-protocol routing, and a variety of servers providing network services, analysis, and decision support.
A detailed discussion of the proposed communication technologies for emergency services is given in the (Grant C. 2015). As an infrastructure-based network technology groups are: cellular, satellite, WLAN, wireless sensor network and wireless mesh network. Infrastructure-less network technologies considered for emergency advertisements are: Wi-Fi direct, Bluetooth/ZigBee, mobile ad hoc network (MANET, VANET) and multi-hop ZigBee. Cognitive radio systems having the feature to autonomously detect existing communication links makes them usable in most emergency cases. Satellite networks offer not only transmission service of the data but geographic coordinate information to the connected nodes. This extra service is essential to the users being in a catastrophic situation.
Architecture and services of the ALBEMS system
Figure 1(a) shows the ALBEMS system architecture during an emergency event, e.g., earthquake, fire, hurricane, flood etc. The data communication is hierarchically forwarded through different levels of the system. The system architecture is composed of ALBEMS Nodes (AN), these are sensors that are installed on buildings which acquire information about survivors and send the data off to the ALBEMS Sink Node (ASN). The ASN in this network architecture is in charge for forwarding the
a) Architecture b) Operation
information to ALBEMS Operator (AO). The AO center is responsible for receiving the critical disaster information after the catastrophe. The AO authority at the AO center delivers S&R teams at the point of need within a limited time. The collected sensor data is passed from the AN to the ASN using the ALBEMS Routing System (ARS), which will be described in more details in section 4.
The ALBEMS System in operation, Figure 1(b) has several functions integrated: a) detection of sounds in the neighborhood produced by the victim; b) emergency signal transmission to the operator; c) voice message playing to the victim; d) geographic coordinate determination for each AN; e) capturing physical parameters of the environment (i.e. temperature, humidity); f) transmission to the ARS; and blinking light indication to the victim.
Adaptive communication mechanism of the ALBEMS system
In our research, the nodes are placed in virtual coordinates known as Grid Points (GP). Filled Grid Points (FGP) have sensor nodes in their vicinity, defined by a sphere G (r, V), where r is the radius and V is the Descartes coordinates of the sphere’s center. Vx, Vy, Vz are the virtual coordinates of GP. Because of the relatively small value of the r, vicinity radius, at most only one node can be in any G (r, V).
Non-filled Grid Points (nFGP) have no sensor nodes in their vicinity given by radius r.
The radio interface is able to cover distance at least R. Four running modes have been defined for the sensor nodes to improve the energy efficiency of the network.
The activities in these running modes are presented in Table 1.
The GP can be classified as follow: AF, PF, AS for FGP; PS and obstacles are nFGP.
The First-Class Neighbors (FCN) are the nodes inside the intersection of the spherical sensing range [R, R+dR] and the incomplete pyramid given in Figure 3.
Because of energy constraints we want our network to have the minimum number of hops as possible. Not to forget that having a large radius consume energy and
Figure 2. Relation between Grid points and AN nodes
having a small one leads to data latency. We defined an incomplete 3D shape pyramid having hexagon bases. This 3D shape is perfect over square or triangular shapes in our architecture because we want to cover the entire area without overlapping (i.e.
we can cover the entire geographical region without any gaps), as we want to connect every existing sensor in the 3D space to reach our objective.
We divided our routing protocol into two specific phases: i) Path Setup phase (PS_Phase); ii) Data Transmission phase (DT_Phase). The PS_Phase aims to discover multiple routes where each node performing a Three-Way Handshake with it FCN. The optimum path, in number of hops and remaining energy, is selected for the DT_Phase.
In Figure 3, S0 (Source) wants to transmit data to SN (Destination). For that matter, S0 start the PS_Phase and send a signal to it FCN nodes (S3 and S4) included in the first zone (Z0). The signal frame structure contains fields: FC (Field Control), FZP (First Class Zone Parameters), SA/DA/PA (Source/ Destination/ Previous Addresses) and ST (Status: Aging Time/ Energy Level/ Number of Hops, etc.). The Data Frame has the same structure extended by DATA field. In the Field Control record are included: FT (Frame Type: Signal/Data), AM (Addressing Mode: Vertex/Center), TC (Trial Counter: First/Second), S/A (Setup/ACK), EXT (Extension), see Figure 4. S0 gets back an ACK from S3 and S4, even though S1 and S2 are closer to the source but as we said before we want our protocol to have the minimum number of hops as possible on the path. Nodes S5, S6 and S7 cannot be FCN to S0 from the point that they are far away from it sensing range. S3 and S4 redo the same process with their own FCN (S8 and S9) for S4.
Running mode Data forwarding Setup initiation Signal forwarding
Active Forwarder (AF) + + +
Passive Forwarder (PF) - + +
Active Sleeping (AS) - - +
Passive Seeping (PS) - -
-Table 1. Running modes of the ALBEMS node
In front of obstacles, nodes cannot perform to their maximum capabilities. Our routing mechanism learns and adapts from the surrounding. In this situation, S3 will extend its sight vision to three or six more zones depending on the needs. As shown in Figure 5.
Extending the sight vision will create multiple directions instead of performing another trial from S0 to find another path which will cost energy and bandwidth. S8 and S9 have their sight vision clear and convey the signaling process to S10, the last forward to S12 and S13 because they are part of S10 FCN and receive back the ACK, unfortunately for S12 the destination is out of reach and it will be the role of S13 to send the last hop of the signal to the destination to begin the DT_Phase.
With this mechanism, we dynamically set up a path from the source to the destination in the presence of obstacles and lack of sensors without performing several trails that can aggressively decrease the network lifetime. The final path can either be: a) S0, S4, S8, S10, S13, SN or b) S0, S4, S9, S10, S13, SN.
Conclusion
In this work, we present the components, architecture and the communication mechanism of a new energy efficient adaptive location based emergency message service using wireless sensor network in emergency scenarios. The outcomes of the research study focus on any disaster where rescue is necessary to save the life of
a) Signal Frame [bit] b) Data Frame [bit] c) Field Control [bit]
Figure 4. Signaling frame structure
Figure 5. Cases of the next zone addressing
a) Case 1: Vertex b) Case 2: Center
people in critical emergency circumstances. The proposed ALBEMS communication service has dynamic adaptation feature integrated making possible to communicate in harsh conditions. Possible continuation of this work can be making the algorithm more reliable, effective on a higher scale, determination of the optimum parameters of the routing algorithm based on energy consumption and connection setup time minimization constraints.
References
Afzaal H. (2016): Towards Formalism of Earthquake Detection and Disaster Reduction using WSANs, International Conference on Frontiers of Information Technology.
Benedetti M. (2010): Wireless sensor network: a pervasive technology for earth observation, IEEE J.Sel. Topics Appl. Earth Obs. Remote Sens. 3, pp. 488–496.
Erd M. (2016): Event monitoring in emergency scenarios using energy efficient wireless sensor nodes for the disaster information’s, International Journal of Disaster Risk Reduction 16, pp. 33–42.
Flammini F. (2010): Towards Wireless Sensor Networks for railway infrastructure monitoring, Proceedings of the IEEE Electrical Systems for Aircraft, Railway and Ship Propulsion (ESARS), pp. 1–6.
Gomez C. (2010): Wireless home automation networks: a survey of architectures and technologies, IEEE Commun. Mag. 48, pp. 92–101.
Grant C. (2015): Research Roadmap for Smart Fire Fighting, NIST Special Publication 1191, http://dx.doi.org/10.6028/NIST.SP.1191
Kone C.T. (2010): Cluster based Multi-Channel for improving Performance of Large-Scale Wireless Multi-Sink Sensor Networks, IEEE Trans. 2nd International Conference on Table 2. Path setup phase algorithm of ALBEMS
Liu Z. (2010): A wireless sensor network based personnel positioning scheme in coal mines with blind areas, Sensors 10 pp. 9891–9918.
Naveed A. (2011): Ad hoc wireless Sensor Network Architecture for Disaster Survivor Detection, International Journal of Advanced Science and Technology, Vol. 34.
Peng S.J. (2012): Experimental study on the influence mechanism of gas seepage on coal and gas outburst disaster, Saf. Sci. 50, pp. 816–821.
Rahman M. (2016): Implementation of ICT and Wireless Sensor Networks for Earthquake Alert and Disaster Management in Earthquake Prone Areas, Procedia Computer Science, pp. 92–99.
Ramulu M. (2009: Damage assessment of basaltic rock mass due to repeated blasting in a railway tunneling project, A case study, Tunn. Undergr. Space Technol. 24, pp. 208–
221.
Saha S. (2007): A Wireless Sensor Network Protocol for Disaster Management, IEEE Trans.
International Conference on Information, Decision and Control.
Silva R.I.D. (2010): Wireless Sensor Network for Disaster Management, IEEE Trans.
Proceedings of International conference on Networks operation and management symposium (MOMS), Osaka Japan, pp. 870–873.
Stephen M.G. (2010): DistressNet: a wireless ad hoc and sensor network architecture for situation management in disaster response, IEEE Communications Magazine 48(3)