International Journal

on Marine Navigation

and Safety of Sea Transportation

Volume 6

Number 3

September 2012

307

1 INTRODUCTION

Wireless sensor networks (WSN) have attracted

considerable amount of attention in recent years.

There is a sort of sensors in WSN, including seismic

sensors, temperature sensors, and humidity sensors

etc. It is a new way of acquiring information plat-

form that can supervise and collect various monitor

objects’ state. Its sensors can be positioned to many

geographical areas that human being can hardly ar-

rive even can’t approach, for instance, volcano, the

arctic pole and hostile battle fields and so on. Small

size and minimum requirements from existing infra-

structure make WSN one of revolutionary technolo-

gies in 21st century that will have a significant im-

pact on our future life.

On the other hand, internet-of-things (IOT) is de-

signed as a world-wide network in which everything

can be identified by a unique address, every comput-

er, each desk so much so that even a stone can ac-

quire an exclusive address if we need. Every object

can join the network dynamically and each terminal

can collaborate and cooperate efficiently to achieve

different tasks. IOT can realize real-time data ob-

taining, information exchange, remote control mak-

ing use of traditional internet. IOT finds a range of

applications in daily life including intelligent trans-

portation system, smart home furniture, intelligent

fireproof and so on.

Nowadays vessels sailing in inland waterways are

increasing dramatically while the existing inland

navigation management information system cannot

provide more efficient service for inland rivers’

management. Shortages of the management system

have manifested themselves in following spheres.

Firstly, fewer applications of advanced techniques

are employed in inland shipping administration. Ac-

cording to the survey statistics, ship-borne supervis-

ing devices are mostly deployed in ocean transporta-

tion vessels, namely AIS, GPS, RADAR, and other

modern technologies are scarcely mounted in inland

ships. Secondly, time-delay in current monitoring

system is larger, in other words information or data

broadcasted by the system are not latest. What’s

more it is difficult for such system to realize super-

vise and exchange information across the areas. As a

result more efficient real-time shipping management

system and more convenient data information shar-

ing system are required. Development of WSNs and

IOT make it possible to implement the very water-

craft management system. The newly system can

achieve low latency, inter-regional organizing net-

work, easy realization and other advantages with the

help of WSN and IOT for the aforementioned supe-

Novel Design of Inland Shipping Management

Information System Based on WSN and

Internet-of-things

H. Wu , X. Chen , Q. Hu & C. Shi

Merchant Marine College, Shanghai Maritime University, Shanghai, China

J. Mo

Science and Technology Division, Shanghai Maritime University, Shanghai, China

ABSTRACT: Currently there are more and more ships sailing in inland waterways so that the traditional in-

land shipping management information system (ISMIS) becomes relatively backward. Based on the rapidly

developed new information technologies, such as wireless sensor network, Internet-of-things, cloud-

computing and so on, we propose a novel design of ISMIS, which is featured by low cost, environment-

friendly, cross platform, high scalability and integrity and thus can efficiently improve the inland shipping

management and inland water environment.

308

riorities are the unique properties of these burgeon-

ing IT technologies.

The remainder of the paper is organized as fol-

lows: In section 2 we present related work. In section

3, we propose whole design of inland shipping man-

agement system, including introduction of system

functionality, design of network structure. Mean-

while system’s data processing based on cloud com-

puting is discussed in section 4 and middleware sys-

tem of inland shipping management system is

expounded in section 5. Then the paper is concluded

in section 6.

2 RELATED WORK

I.F. Akyildiz et al. [1] provided a review of factors

influencing the design of sensor networks and the

communication architecture for sensor networks was

outlined. In addition the algorithms, protocols for

each layer and open research issues for the realiza-

tion of sensor networks were also explored. He gives

us a better understanding of applications of sensor

network and the current research issues in the field,

like node localization, designing energy efficient ra-

dio circuits, sensor network topology, fulfillment of

its adaptivity to environment, protocols for sensor

network’s different layers and so on. It will be better

if the authors talk about difference between tradi-

tional sensor networks, wireless sensor network as

former can provide some assistance for the later.

The machine learning techniques applied in WSN

from both networking and application perspectives

were surveyed by Ma Di et al. [2]. Machine learn-

ing techniques have been applied in solving prob-

lems such as energy-aware communication, optimal

sensor deployment and localization, resource alloca-

tion and ask scheduling in WSNs. In application

domain, machine earning methods are mainly used

in information processing such as data conditioning,

machine inference and etc. The paper proposes a

novel approach that first extends rigorously pub-

lished mathematical constructs that merely approxi-

mate the long-term or stationary behavior of infor-

mation flows in WSNs to simplify computation and

simulation. Moreover, the approach, based on multi-

variate point processes, is shown to represent inter-

action of the parameters of the protocol layers by

William S. Hortos [3]. Both Ma Di [2] and William

S. Hortos [3] are talking about machine learning

techniques while Ma focuses on the application as-

pects and William mainly on theoretic field, and

their conclusions may be more stringency if they

present some simulations.

When deploying several applications over the

same WSN, One of the remaining problems resides

in the data aggregation solutions, which are pro-

posed generally for one application and may drain

the WSN power in a multi-application context.

Therefore, Ahmad Sardouk et al. [4] propose a data

aggregation scheme based on a multi-agent system

to aggregate the WSN information in an energy-

efficient manner even if we are deploying several

applications over this network. This proposal has

proved its performance in the context of one and

several applications through successive simulations

in different network scales. Ahmad Sardouk gives us

a concrete data priority processing scheme, and it

would be better if it provides some mathematical

model for the data aggregation scheme and discusses

the scheme’s node localization.

As ZigBee becomes a standard for WSN (Wire-

less Sensors Networks), and ZigBee and 802.15.4

had been proving they can achieve the mission

splendid, Xavier Carcelle et al. [5] will emphasize

on the past, present and future features for ZigBee,

taking a look on the feedback from previous imple-

mentations to finally design the next generations of

WSN based on ZigBee. Xavier Carcelle introduces

the past, present and future features for ZigBee in

detail, taking a look on the feedback from previous

implementations to finally design the next genera-

tions of WSN applications based on ZigBee. If he

connects WSN s’ applications with some new devel-

oping techniques, such as internet-of-things, cloud

computing and so forth, it will be perfect.

The research presented by Fuxing Yang et al. [6]

has been focused on wireless gateway based on

ZigBee and TD network. According to problems

caused by limited bandwidth of traditional wireless

sensory gateway, such as inferior network perfor-

mance and low efficiency of communication, this

paper comes up with a solution---a wireless sensory

gateway based on ZigBee/TD, closely combining

ZigBee net with TD-SCDMA net. In the plan of de-

sign, the network nodes transport data to gateway by

ZigBee short-range communication technique after

collecting them and the gateway sends them to con-

trol center by TD network from long distance, realiz-

ing the highly efficient long-range transport of data.

The design satisfies the efficiency and transparency

needed for inter-networks data exchange and it can

be scalable easily. Fuxing Yang gives an excellent

wireless gateway scheme and it may be more per-

suasive supposing that Fuxing Yang introduces

some related work.

Miaomiao Wang et al.[7] provide a comprehen-

sive review of the existing works on WSN middle-

ware, seeking for a better understanding of the cur-

rent issues and future directions in this field. They

also propose a reference framework to analyze the

functionalities of WSN middleware in terms of the

system abstractions and the services provided and

309

review the approaches and techniques for imple-

menting the services. Based on the analysis and us-

ing a feature tree, the paper provides taxonomy of

the features of WSN middleware and their relation-

ships, and uses the taxonomy to classify and evalu-

ate existing works. Open problems in this important

area of research are also discussed in [7]. Miaomiao

Wang gives us an overall detailed existing research

on WSN middleware. The paper can be a more co-

gency survey as if he tells some transport protocols

in WSN middleware.

Study on Internet-of-things is either in full swing.

Jianhua Liu et al. [8] propose a formal IOT context

model to perform self-adaptive dynamic service.

They provide a general context aware service based

on IOT communication and their context model is

used for service match and service composition to

reduce the consumption of devices resources and

cost. Stephan Haller et al. [9] survey that how the In-

ternet of Things is put in a wider context: how it re-

lates to the Future Internet overall and where the

business value lies so that it will become interesting

for enterprises to invest in it. Stephan Haller also

proposes the major application domains where the

Internet of Things will play an important role and po-

tential concrete business opportunities. Aitor Gomez-

Goiri et al. [10] address the progress towards a se-

mantic middleware which allows the communication

between a wide range of embedded devices in a dis-

tributed, decoupled, and very expressive manner.

This solution has been tested in a stereotypical de-

ployment scenario showing the promising potential

of this approach for local environments. Welbourne

E. et al. [11] design a suite of Web-based, user-level

tools and applications to empower users by facilitat-

ing their understanding, management, and control of

personal Radio Frequency Identification (RFID) data

and privacy settings. These applications are deployed

in the RFID Ecosystem and a four-week user study is

conducted to measure trends in adoption and utiliza-

tion of the tools and applications as well as users’

qualitative reactions.

Jianhua Liu [8] provides a simple enhanced dy-

namic service selection model and a formal internet-

of-things (IOT) context model; both of the models

are self-adaptive dynamic service model, and it

would be better as if the paper provides some math-

ematical models. Stephan Haller et al. ([9]-[11]) tell

us something more about IOT’s concrete enterprises

applications, combined with its specific applying

techniques, such as RFID, middleware and so on.

From personal point of view, the three papers need

some simulations. Stephan Haller [9] need to sup-

plement some content in chapter of the major tech-

nique issues, such as privacy, virtual and physical

world fusion and so on, as these are important for

the spread of IOT in future. Aitor Gomez-Goiri et al.

[10] can introduce something concerned with IOT

middleware and distributed computing so that the

paper may be more consummate. Analogously,

Welbourne E. [11] may replenish some information

about RFID middleware for building the internet-of-

things era.

3 INLAND SHIPPING MANAGEMNET

INFORMATION SYSTEM

3.1 System functionality

Currently inland shipping management is in com-

paratively chaotic state. Vessel control and supervise

departments are distributed in different regions such

that management efficiency is lower for information

barrier and information island. Evolution of IOT and

WSN provides novel approach for internal naviga-

tion administration. Accordingly it is high time to

establish a more efficient integrated management in-

formation system, which we refer to as Inland Ship-

ping Management Information System (ISMIS) ca-

tering to the aforementioned claim.

ISMIS takes advantage of wireless sensor net-

work technology, RFID technique, and IOT to pro-

vide real-time information for inland ships, vessel

management department, and other correlative ad-

ministrative departments. RFID tags embedded in

ship can transform ships’ dynamic and static infor-

mation to base station, auxiliary sensors can de-

ployed in inland waterways, bridges, ships which

haven’t been equipped with RFID equipments or

other identification facilities and other places if nec-

essary. Wireless Sensor Networks can acquire and

survey ships’ state (including ship’s name, vessel’s

number, tonnage of ship, vessel course, and naviga-

tional speed etc.) accurately such that upper comput-

ers and central control monitoring system can raise

the management level tremendously.

WSN that used to detect environment and device

status can avoid some unwanted accidents. For in-

stance, sensors deployed in bridge will detect

bridge’s intensity of pier and bridge girder by the

minute so that its hidden danger can be hustled out

of the way as early as possible. Sensors in ISMIS

can also aid navigation as they can broadcast fair-

way information ahead such as traffic condition,

traffic density, navigational danger, water area state,

and so forth.

What’s more ISMIS provides a good basis for the

standard of RFID technique, information system, in-

formation encoding, measurement, management

mode, manipulation, and other technique touchstone.

3.2 Design of network framework

Network structure of ISMIS includes wireless sen-

sor networks, cable network, terminal users and,

310

management system server, and so forth. Cable net-

work mainly presides over the communication

among terminal customers, principal computers, and

center controlling system. Data transmitted from

base station or principal computers can be sent to

center controlling system by cable network. Users’

requests and administrators’ management activities

are executed through internet. Spot data and infor-

mation delivered by RFID data acquisition unit can

be gained by upper computers in real-time. The ar-

chitecture of network frame is displayed in Fig. 1.

Figure 1. Architechture of network framework

Wireless sensor network in ISMIS primary im-

plements function of detecting vessel information,

environment state, bridge information, traffic status,

and other factors which affect vessel operation. Sen-

sor nodes can be divided into four categories: vessel

nodes, bridge nodes, shore-based nodes, underwater

nodes, and other nodes. With nodes distributed in a

variety of monitoring areas, we can coverage most

of supervising areas at the cost of lower costing such

that all monitoring targets will transfer their status

messages, at regular time, to base station or other da-

ta acquisition devices. WSN is ambient environment

self-adaptively ad hoc network; it consists of physi-

cal layer, MAC layer, topology layer, network layer,

transfer layer, and application layer. On one hand,

RFID data collecting facilities sweep ship-borne

RFID device to get watercraft’s relative information

at a fixed time, and data acquired in real-time are

transmitted to local vessel management department

and vessel center control monitoring system by

RFID middleware system. On the other hand, vessel

sensor nodes, which have been equipped with mi-

crochips that can realize information exchange au-

tomatically, underwater sensor nodes, shore-based

nodes, and other sensor nodes shape ad hoc network.

The ad hoc network is able to get and deliver super-

vising range’s dynamic information and static state

in real time which enhance remote management effi-

ciency. Certainly WSN’s information and data

should be transferred to center management moni-

toring system with the help of WSN information

middleware system. Partial pseudo-codes of WSN

information middleware system are as follows:

Implementation {components vessel-node;

Class create-node-information

{Public:

Transfer (vessel-node information);

Private:

Char* vessel-node location;

Char* vessel-speed;

Char* vessel-direction;

int vessel-call-sign;

int vessel-IMO;

};

Refresh information database;

Return 0 ;}

3.3 Design of inland shipping management

information system

Inland shipping management information system

(ISMIS) employs wireless sensor network, RFID

technique, IOT technology and other modern IT

technique into inland vessel administration. ISMIS

realizes precisely control over inland ships by the

aid of WSN, IOT, and RFID, for the center control

and management system of ISMIS can receive su-

pervisory scope information concerning ships, navi-

gation lock, fairway, traffic density, and other cor-

relative data.

WSN in ISMIS plays the role of monitoring

most targets, such as ships sailing in inland rivers,

buoys, environment state, navigation danger, and so

forth. WSN combines with RFID constitutes ISMIS

bracket and network nerve. With the help of such

modern IT technique, ISMIS can effectively sched-

ule distributed vessel management subsystem and

ship sailing in inland rivers so as to prevent traffic

jam and accidents happening. Structure of ISMIS is

showed in Fig. 2.

311

Figure 2. Architecture of inland shipping management information system

3.4 Achievement of dynamic node connecting to

local cluster network

As vessels are shipping on inland rivers, nodes of

WSN are changing constantly. Therefore locomotive

ship nodes connect to local WSNs via wireless sen-

sor networks gateway and concrete realization as

follows:

Step 1: apply for joining native sub-WSN

A ship which functions as traveling node need to

join native wireless sensor sub-network to formulate

real time ad-hoc network. First thing for the mobile

node is applying for registration in the sub-network.

It broadcasts gateway request messages over the

sub-network. The nearest sink node of the sub-

network will build request-join node’s path distance

variable-Hops as soon as it receives registration

messages delivered by ship node, and its initial value

is 0. Then sink node sends permission messages of

joining sub-network and application affirm infor-

mation to the application node by cluster’s nodes.

That very node will deliver response verification

messages to sink node such that sink node can dy-

namic updates cluster nodes information.

Step 2: registration in the sub-network

Ship node transmits its relevant information, in-

volving path hops, routing information and other in-

formation to sink node to login the cluster as a tem-

porary member. Head-node forwards the receiving

information to upper-computer system through wire-

less gateway. Upper- computer system generates

provisional ID for new ship node and stores its in-

formation into ship node database. After that, ex-

temporaneous ship node ID, registration identifying

code and other essential data packets are sent to sink

node and vessel sensor. Ship node can complete reg-

istering procedure so long as it receives extempora-

neous ID and registration code. So far, sink node

possess newcomer’s information while it obtains

upper-computer system’s consent orders, such as

registration node ID, registration area and so forth.

Till then it is legal for the newcomer communicates

with sink node and executes the task assigned by su-

per-stratum node. The representation of the node

joining local sub wireless sensor network is dis-

played in Fig. 3.

Figure 3. Schematic of new node joining into native sub-wsn

312

Step 3: task process

On the one hand, the new-coming member node

needs to coordinate with sink node, shore-based

nodes, environment supervising nodes and other

nodes to realize information sharing that supervised

and transmitted by upper nodes.

Ship-borne information collecting device, namely

ship node transfers its data and acquired information

to sink node which will transmit them to local vessel

management department after data filtering and

screening by network base station. Simultaneously,

native environment nodes and bridge nodes and oth-

er relative nodes broadcast their monitoring data and

messages to vessel controlling center which com-

prise of traffic density, water quality, embargo an-

nouncement, fair-way condition ahead and so on to

shipping management centre to realize maximum ef-

ficiency of the whole wireless sensor network. In

addition each sink node and nodes in every cluster

communicate with each other through optimal path,

while gateway link sensors are deployed in every

navigation bridge, distributed to implement remote

control and converting protocols between different

networks and sink nodes.

4 DATA PROCESSING BASED ON CLOUD

COMPUTING

Currently there is no uniform international definition

of cloud computing. Domestic IT industry gives the

definition of cloud computing as follows: cloud

computing is the fusion of the traditional IT technol-

ogies and some newly developing techniques; the

former involves grid computing, distributed compu-

ting, parallel computing, utility calculation, network

storage, virtualization, load balance and the latter

mainly comprises grid technology.

Inland shipping management information system

as a large-scale inland crafts management system, it

will handle, transmit, and store mass data as well as

huge quantity of information and it really a chal-

lenge for we to implement sufficient facilities to ac-

complish such function. However cloud computing

can meet our demand in acceptable overhead. For

this reason we make use of could computing to con-

duct, store and broadcast our data as same as infor-

mation.

The specific procedures as follows: firstly, data or

messages delivered by RFID middleware systems

are decrypted, filtered, transformed in the light of

stated format and requirement by distributed calcu-

lating center, which includes virtual computing cen-

ters and physical calculating centers, in cloud com-

puting platform. Then vessel central control

monitoring system which is the command center of

inland shipping management information system

will publish warning information for ships if any ob-

structions exist in front of the waterway. In addition,

cloud computing can also forward, convert format of

correlative upper computers’ orders for rock-bottom

devices and equipments. Besides, WSN middleware

systems dispatch their gathering information to ves-

sel center monitoring system through wireless

transmission protocol and then all messages are con-

ducted by cloud computing which as with processing

RFID devices’ information.

5 MIDDLEWARE SYSTEMS DEVELOPMENT

In the inland shipping management information sys-

tem middleware systems are essential for the data

and information of ship-borne RFID devices and

WSN nodes are incompatible in upper computer sys-

tems. Consequently we need to develop RFID mid-

dleware system and WSN middleware system re-

spectively. Function modules of RFID middleware

system comprise formatting data, ensuring data

communication security, data caching and filtering

module, supplying application program interfaces

etc [12]. That’s to say, all data and messages deliv-

ered by ship-borne facilities will be processed before

they are sent to vessel central control monitoring

system. To the contrary, all messages would be

transformed to compatible formats before they are

transmitted to lower layer or bottom users broad-

casted by ship central control supervisory system.

WSN middleware system is somewhat resem-

bling RFID middleware system. The middleware

system offers several terminal users’ interfaces for

outside applications, including ship node middle-

ware interface, bridge node middleware interface,

water-supervisory node middleware interface, and

other comprehensive interfaces for different kind of

sensor nodes. These interfaces will transmit and re-

ceive data or information as required in WSN pro-

cessing and monitoring center. Internal function

modules of WSN middleware will execute their reg-

ular work automatically as long as relative data or

information access into database of middleware sys-

tem. The architecture of WSN middleware system is

shown is Fig. 4.

313

Figure 4. Architecture of WSN middleware system

6 CONCLUSION

This paper proposed a novel design of inland ship-

ping management information system (ISMIS) for

improving inland shipping management level. Since

the design is based on wireless sensor network

whose cost-effective is steadily higher the recent

years, ISMIS can extend its supervising objects dy-

namically with the help of WSN, IOT, cloud compu-

ting, middleware technology and other IT tech-

niques. Namely, inland shipping will be more

efficient, real-time, and convenient than present in-

land shipping management system and inland ships

will enjoy better information service in envisioned

future with the establishment of ISMIS. In future we

will pay more attention to the improvement and im-

plementation of inland shipping intelligent manage-

ment level using WSN and IOT techniques.

ACKNOWLEDGEMENT

This work was supported by Science & Technology

Program of Shanghai Maritime University

(20110028), Innovation Program of Shanghai Mu-

nicipal Education Commission (09YZ247) and

Shanghai Leading Academic Discipline Project

(S30602).

REFERENCES

[1] I. F. Akyildiz, W. Su, Y. Sankara, subramaniam and

E.Cayirci, “Wireless sensor networks: a survey,” Computer

Networks, vol. 38, pp. 393-422, Mar. 2002.

[2] Ma Di, Er Meng Joo, “A Survey of Machine Learning in

Wireless Sensor Networks From networking and applica-

tion perspectives,” IEEE Information, Communications

&Signal Processing, pp. 1-5, Dec. 2007.

[3] William S. Hortos, “Analytical Models of Cross-Layer Pro-

tocol Optimization in Real-Time Wireless Sensor Ad Hoc

Networks,” Lecture Notes of the Institute for Computer

Sciences, Social Informatics, and Telecommunications En-

gineering, vol.28, pp.762-779, 2009.

[4] Ahmad Sardouk, Rana Rahim-Amoud, Leïla Merghem-

Boulahia and Dominique Gaïti, “Data Aggregation Scheme

for a Multi-Application WSN,” Lecture Notes in Computer

Science, vol.5842, pp.183-188, 2009.

[5] Xavier Carcelle, Bob Heile, Christian Chatellier and Patrick

Pailler,“Next WSN applications using ZigBee,” IFIP Inter-

national Federation for Information Processing, vol.256,

pp.239-254, 2008.

[6] Fuxing Yang, Chuansheng Yan, “Design of WSN Gateway

Based on ZigBee and TD,” Insitute of Electrical and Elec-

tronics Engineers, Inc, 2010 3th International Conference

on Electronics and Information Engineering (ICEIE 2010),

pp. v V2-76 - V2-80, 2010.

[7] Miaomiao Wang, Jiannong Cao, Jing Li, Sajal K. Dasi,

“Middleware for Wireless Sensor Networks: A Sur-

vey,”Journal of Computer Science & Technology, vol. 23,

pp 305-326, May. 2008.

[8] Jianhua Liu; Weiqin Tong, “Dynamic Service Model Based

on Context Resources in the Internet of Things,” Wireless

Communications Networking and Mobile Computing

(WiCOM), 2010 6th International Conference, pp. 1-4,

2010.

[9] Stephan Haller, Stamatis Karnouskos, Christoph Schroth,

“The Internet of Things in an Enterprise Context,” Comput-

er science, vol. 5468/2009, pp 14-28, 2009.

[10] Aitor Gomez-Goiri, Diego Lopez-de-lpina, “A Triple

Space-Based Semantic Distributed Middleware for Internet

of Things,” Computer Science, vol. 6385, pp 447-458, 2010

[11] Welbourne E., Battle L., Cole G., Gould K. etal, “Building

the Internet of Things Using RFID:The RFID Ecosystem

Experience,”

Internet Computing, IEEE, vol. 13, pp 48-55,

May. 2009.

[12] Huafeng Wu, Xinqiang Chen, Qingyun Lian etc.“ Design

of Cloud-computing-based Middleware for Internet-of-

ships in Inland Shipping,”Insitute of Electrical and Elec-

tronics Engineers,Inc, 2010 3th International Conference on

Power Electronics and Intelligent Transportation System,

pp. 289-292, Nov.2010.

[13] LIU Lin-feng; LIU Ye, “Research on Wireless Sensor

Network Architecture Model in Ocean Scene,” Computer

Science, vol. 37, pp. 74-77, Oct. 2010.

[14] MA Xiao Tie, LI Kai, “Application of wireless sensor

network in human physical parameters collection,” Micro-

computer & Its Applications, vol. 29, pp. 39-42, 2010