Author:Hemant Kumar Rath, Rajan MA, Anantha Simha

GISFI Meeting #9GISFI_IoT_201206228

Mumbai, India, June 18-20, 2012

Agenda item:WG IoT

Source:TCS

Author:Hemant Kumar Rath, Rajan MA, Anantha Simha

Title:Frame-work document for Technical Report on Landslide DetectionUse-case

Document for:Discussion and approval to accept as baseline Technical Report document

GISFI TR ab.cde V0.0.0(2012-06)

Technical Report

Global ICT Standardisation Forum for India;

Technical Working Group IoT;

Landslide DetectionUse Case Technical Report;

(Release 1)

The present document has been developed within GISFIand may be further elaborated for the purposes of GISFI.

GISFI TR ab.cde V0.0.0 (2012-06)

1

Release 1

Keywords

Landslide detection, Sensor Networks, IoT, Monitoring, self-optimization

GISFI

GISFI office address

Global ICT Standardisation Forum for India (GISFI),

Singhad Campus, Gat 308/309, Kusgaon (Bk.)

Off. Mumbai– Pune Expressway, Lonavala, India

Tel.: +91 2114 304 353 / 401 Fax: +91 2114 278304

Internet

E-mail:

Copyright Notification

No part may be reproduced except as authorized by written permission.
The copyright and the foregoing restriction extend to reproduction in all media.

© 2011, GISFI

All rights reserved.

Contents

Contents......

Intellectual Property Rights......

Foreword......

1Scope......

2References......

3Definitions, symbols and abbreviations......

3.1Definitions......

3.3Abbreviations......

4IoT applications for Landslide Detection......

4.1General description of IoT applications for Landslide Detection......

4.2Specific examples for IoT applications for Landslide Detection......

4.2.1Landslide type......

4.2.2Rate of movement of the materials......

4.2.3Types of materials......

5Landslide Detection use cases......

5.1Overview......

5.2Detailed use cases......

5.2.1Specific Requirements......

5.2.2.Network Management......

5.3IoT Devices......

5.4Mapping into the IoT Architecture

6Conclusion......

Intellectual Property Rights

IPRs essential or potentially essential to the present document may have been declared to GISFI. The information pertaining to these essential IPRs, if any, is publicly available for GISFI members and non-members, and can be found in GISFIyyyy: "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to GISFI in respect of GISFI standards", which is available from the GISFI Secretariat. Latest updates are available on the GISFI Web server (

Pursuant to the GISFI IPR Policy, no investigation, including IPR searches, has been carried out by GISFI. No guarantee can be given as to the existence of other IPRs not referenced in GISFIyyyy (or the updates on the GISFI Web server) which are, or may be, or may become, essential to the present document.

Foreword

This Technical Report (TR) has been produced by GISFI Working Group (WG)Internet of Things (IoT).

The present document may be referenced by other TRs and Technical Standards (TS) developed by GISFI WGIoT. The present document is a TR and therefore, the content is informative, but when this TR is referenced by a TS, the referenced clauses may become normative with respect to the content of the referencing TS.

1Scope

The present document collects Use Case descriptions for Landslide detectionapplications in context of Internet of Things (IoT) communications. The described Use Cases will be used to derive service requirements and capabilities of the functional architecture specified in GISFI WGIoT.

2References

The following referenced documents are indispensable for the application of the present document. For dated references, only the edition cited applies. For non-specific references, the latest edition of the referenced document (including any amendments) applies.

[1].Technical Report on IoT Reference Architecture (Draft - GISFI_IoT_201203175)

[2]. Frame-work document for Technical Report on IoT Service Requirements (Draft GISFI_IoT_201106103)

[3].Framework document for landslide detection using sensor networks use casee (Draft - GISFI_IoT_201109119)

[4].Hemant Kumar Rath, Rajan MA and Balamuralidhar PA “Monotonic Signed Graph Approach for Cross-layer Congestion Control in Wireless Ad-hoc Networks,” 6th IEEE HeterWMN (IEEE Globecom), Houston, USA, Dec, 2011.

[5].Hemant Kumar Rath, Rajan MA, Anantha AImha, “Lifetime and Throughput Enhancement of Distributed Wireless Networks: A Multi Time Scale Self-Optimization Approach”, TACTiCS 2012.

3 Definitions, symbols and abbreviations

3.1 Definitions

Landslide Detection: Generic term for a class of applications that serves the purpose of detecting landslides and alerting the authority by means of electronic information and/or communication technology.

IoT Device: Entity capable of sensing the state/features of an object and environment and has communication and computing capability. It may run limited local applications. The IoT device connects to the IoT Core network either through a gateway or directly through embedded gateway functionality.

IoT Gateway: IoT Gateway acts as a proxy between the IoT devices and IoT Core network. It shall locally manage the IoT devices (including IoT Limited Devices) and shall run IoT applications.

IoT Network: IoT Network includes the access network that connects the IoT devices to the gateway and core network that connects the gateways to the IoT service platforms.

3.3 Abbreviations

For the purposes of the present document, the following abbreviations apply:

IoTInternet of Things

IPInternet Protocol

TCPTransmission Control Protocol

4 IoT applications for Landslide Detection

This document describes the IoT applications and use cases for landslide detection scenarios

4.1 General description of IoT applications for Landslide Detection

Real-time monitoring of the place/environment around the globe, which is very vulnerable to the natural disasters is one of the prime requirements of the world. Information related to the occurrence of natural disasters such as Landslide in highways, train lines, mountains, etc is important and is of national interest. Detection of landslides or predicting the occurrence of landslides and subsequently transmitting/communicating the location, time and severity information related to landslide to a central location is of prime importance for the society and for the country. Real-time transmission and monitoring of such incidents requires specific devices and protocols to be planned. Though there might be different methods to detect and to communicate landslide information, sensor networks can be used extensively for this purpose. In this, wireless sensors can be deployed in the landslide prone area in a determined and non-determined (sensors can be dropped randomly) manner. In most of the cases, deployment of such sensors may not be possible in a planned manner. Therefore, for accurate sensing, more and more sensors need to be deployed in the target area. Along with accuracy, the life time, efficient transmission and self-maintainability of these sensors are also important. Sensor networks of this type not only will save life, but also will be beneficial for providing necessary and reliable information for road/rail transport and other means of communication (such as telecommunication). This will also help us to save our forest lands, agricultural fields and human settlements.

Considering to India specific requirements, there are cases of major landslides followed by accidents in the Himalayan area (states like J&K, Himachal Pradesh, UP, Bihar, and the entire North-East), in the Western Ghat (Konkan area) and in highways like Mumbai-Pune Highway, Konkan Highways etc. Therefore, deployment of sensor networks to detect landslides in those areas is of national importance and necessity. However, deployment of such sensors and effective use of them are quite challenging. The major difficulties one may face while deploying such techniques are as follows:

• Scale. The deployed land in this case can be of millions of hectares and can be of difficult terrain (such as hills, jungles, river beds, etc.). It is not only difficult to access (physically) the wireless sensors of this network in these areas, but also difficult to manage such kind of networks.
• Reliability: The sensor nodes should be less complex and reliable, such that information sensing and communication to the sensor gateway should be accurate. These nodes need to be reliable in terms of accurate information and quick information sources. Any delay in sensing and communicating, may cost human life and other loss.
• Lifetime Deploying the sensors and replacing old sensors periodically or frequently in un-accessible terrains like hills, river beds, jungles, etc., are not practically viable. Therefore, these sensors should work for a longer lifetime by optimally using battery power available with them. Individual node life time as well as the network lifetime should be considered while designing the network deployment.

4.2 Specific examples for IoT applications for Landslide Detection

Major use cases for the sensor based landslide detection are:

• Landslide type
• Rate of movement of the materials
• Types of materials

We now briefly discuss the major applications/services expected out of automated landslide detecting mechanism with suitable data analytics support.

4.2.1 Landslide type

In India, landslide mainly occurs in the rainy season, mostly triggered by heavy rainfall and de-forestation. Heavy rainfall not only triggers land sliding, but also accelerates the movement of the debris. Landslide may occur in the form of slide or slump or flow or fall of debris. So, the sensors deployed in the landslide prone area must differentiate the landslide type in terms of slide/slump/flow or fall. Based on this, messages/digital information can be generated from the sensors and can be transmitted to a central point. Detection of landslide type provides necessary information about the gravity of the disaster that might occur due to this.

4.2.2 Rate of movement of the materials

Because of landslide type as discussed above, the debris move from one place to other. Depending upon the rate of and type of movement (either flow/slump/fall), soil type and slope of the terrain, the debris get transported and deposited at a place. If this deposited place happens to be the highway or railway line or human settlements, then disasters may occur. Therefore, detection of the rate of movement of the debris is necessary for taking appropriate action. The sensors deployed for this should be capable of sensing and communicating the above information to the central location. Moreover, there is high chance of destruction or loss of the sensors during the flow or movement of the debris. Therefore, there should be enough redundancy in communicating the above information.

4.2.3 Types of materials

Mostly, landslides are composed of bedrock, unconsolidated sediment and/or organic debris. Sometimes, there are chances of avalanche of huge rocks or big trees during landslide. These information need to be communicated to the central location.

5 Landslide Detection use cases

5.1 Overview

Landslide is a common Phenomenon in hilly regions. It is a geological activity which consists of various kinds of ground movement like rock-falls, shallow debris avalanchewhich can cause immense damage and threat to the people who live there. In India major incidents of landslide had occurred at Ladakh, Western Ghats, Hilly regions of Kerala, Hilly regions of Mumbai sub-urbs, etc, resulting in accidents, loss of human life and property, etc. In addition to these incidents, there are frequent cases of accidents due to landslides in Konkan and Central Railway. Because of the landslide occurred in the area near the railway tracks or inside tunnels through which the tracks are passed through, huge rock boulders used to fall on the railway track causing major train accidents due to track destructions and blockage. Hence, a solution to prevent this kind of disastrous is to detect/predict the landslides as early as possible and alarm the concerned public authorities, so that accidents can be averted.

Remote sensing techniques can be employed for landslide hazard assessment and analysis. Before and after aerial photographs and satellite imagery can be used to gather landslide characteristics. Around the world, the research is carried out applying the wireless sensor networks to detect the landslides. It would be very appropriate if we bring it under IoT framework

5.2 Detailed use cases

Implementation of wireless sensor networks in hilly regions, jungles, riverbeds has potential to detect/predict the landslides and alert the authorities to prevent disasters. Apart from that it can also predict the threat level of landslides so that the extra precaution can be taken to avoid accidents. This document discusses various ICT requirements for applications of IoT in Landslide detection system.

5.2.1 Specific Requirements

Requirements on the sensors/parameters to be monitored

Environment/Soil Monitoring: Following environment parameters are required to be measured:

• Soil Moisture
• Earth Pressure (Due to rain water absorption)
• Tilt of the area
• Strain of the soil
• GPS information of the area of concern
• Soil Type

Earth Parameters:

• Ground displacement (geophone)
• Pressure (due to height, due to movement of materials, due to other reasons)

Network Requirements

The network should have sufficient nodes / measurement density that are necessary to model the landslide detection. This will vary with the type of terrain. The terrain can range from several meters to kilometres and can be of difficult area to access.

Data Monitoring

The data sampling rate should have sufficient periodicity to support the analysis, modelling and responsive action. The periodicity required for sensors sampling is one snapshot typically per 3 minutes. The sensor data should be aggregated and available to applications typically in every half an hour during rainy season and in an hour during other season. The typical formatted data rate per node is 2/4K bytes / hour. There should be a provision to change the sampling and aggregation periodicity as and when required

5.2.2. Network Management

The network should have a very low maintenance with limited human intervention. If battery is used as an energy source then the replacement periodicity should be at least two years (in most of the cases battery replacement may not be possible at all). For this kind of scenario, solar power driven sensors should be deployed (along with battery power), such that the lifetime of the network can be maximized. The network should be partitioned into several clusters, as the area coverage runs into several kilometres. So the design issue here is how to partition the network into optimum number of clusters. Based on the cluster size and type, designated routing points need to be selected. These designated routing points or cluster heads need to aggregate the data from the cluster nodes and transfer the data to the server. For simplicity, the cluster heads must be capable of supporting both IoT features and IEEE 802.11 protocol. Flexibility must be provided in selecting new cluster heads due to cluster head failures or due to landslide. Efficient cluster based routing mechanisms are required for information exchange between the nodes and the application server. Self-optimization based routing dynamic routing protocols can be of importance for this purpose. In addition, Remote Calibration of sensors through cluster head is also required.

Upon detection of landslide by the application, the same should be broadcasted to the relevant cluster heads so that beacon alarm at the cluster head is activated. So efficient multicasting algorithm should also be designed. Network should have self healing capability so that the failure of a few nodes will not affect the network operation and application in a major way, in other words network should support self-organisation to ease the deployment.

5.3 IoT Devices

5.3.1.Sensor Device

Sensor nodes should have the mechanism to conserve their energy consumption and desirably have energy scavenging to maximise their lifetime. Use of alternate energy sources such as solar power should be incorporated. Sensor node should be robust enough to be deployed in the field. It should be compact and weather proof (rain, sun, wind, soil acidity, animals etc). It should detect and alert in case of tampering or theft attempts. It should be capable of reprogramming/recharged/reset while in the network. Sensor nodes should use practically implementable and efficient schemes such as self-optimization based power-aware routing or joint power and congestion control schemes as proposed in [4, 5]. These schemes not only improve the lifetime of the individual nodes, they improve the lifetime of the entire network with QoS constraints.

5.3.2.Cluster Head

As discussed earlier, the network is partitioned into several clusters with designated cluster heads (acts as gateway devices) which aggregate data from connected sensor nodes and connect to the Internet through suitable backhaul connectivity. The cluster head should connect to the backend server/cloud using standard Internet protocols. The cluster head should support a standard data exchange format or protocol to communicate with the backend server. The cluster head should support data compression for the efficient data transmission.

5.3.3.Other sensors / Devices on the network

The network also should desirably connect to other sensors and devices that are likely to be in the terrain such as:

• GPS Devices
• Geophones
• VSAT
• Sensing Cameras

These devices are IoT devices and are capable of transmitting sensed information to cluster-head or to IoT Gateway of Figure 1 (explained in Section 5.4)

Localisation

The network should support the automatic localization/position information of the sensor nodes. This information is required for the data modelling and analysis and for taking necessary action so as to save life from disaster and accidents.

Naming and Addressing

These devices should follow unique naming and addressing schemes as approved by the IoT WG.

Packetization and Data Transmission

The sensed information should be packetized and transmitted from/to the sensors to/from the sensor gateway/sensors. For reliable transmission any appropriate transport layer protocol approved by IoT WG should be used.

5.4 Mapping into the IoT Architecture

In this section, we discuss how the landslide use case is mapped to IoT interfaces? Since the sensors used for landslide detection are IoT devices, they can communicate with the IoT gateway or Sensor gateway directly or indirectly (through multi-hop networking) using the interface I1 of the proposed architecture. Figure 1 illustrates the connection of IoT devices used for landslide detection with that of IoT Reference architecture [2] proposed by the IoT WG. The Sensors of Figure 1 are the sensors deployed in the field to perform landslide detection. They have either limited IoT functionalities or full functionalities (IoT device/IoT-limited device [2]). The cluster nodes are either sensors/IoT devices or IoT gateway of Figure 1. Other devices or sensors as discussed in Section 5.3.3. are also sensors/IoT devices of Figure 1.