SS2 Report
-
Upload
chandra-kanth -
Category
Documents
-
view
92 -
download
1
description
Transcript of SS2 Report
R.V.COLLEGE OF ENGINEERING
R.V. COLLEGE OF ENGINEERING, BANGALORE-560059(Autonomous Institution Affiliated to VTU, Belgaum)
Self-Study Report on,
IOT ON INTELLIGENT TRAFIC SYSTEM(Phase -2)
Submitted by,
MALLIKARJUN MATTI1RV14CS80
Under the guidance of,
Dr. Sharvani G. S, Associate Professor, CSEMr. Anjan K, Assistant Professor, CSE
Mr. Girish Rao Salanke N S, Assistant Professor, CSEMs. Kowcika A, Assistant Professor, CSE
Submitted to,
COMPUTER SCIENCE AND ENGINEERING DEPARTMENT,R.V. COLLEGE OF ENGINEERING.
R.V. COLLEGE OF ENGINEERING, BANGALORE – 560059
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page i
R.V.COLLEGE OF ENGINEERING
(Autonomous Institution Affiliated to VTU, Belgaum)
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
CERTIFICATE
Certified that the Self Study work titled ‘IOT on intelligent traffic system’ is carried out by N
MALLIKARJUN MATTI (1RV14CS080) who is bonafide student of R.V College of Engineering,
Bangalore, in partial fulfillment for the award of degree of Bachelor of Engineering in Computer
Science and Engineering of the Visvesvaraya Technological University, Belgaum during the year
2015-2016. It is certified that all corrections/suggestions indicated for the internal Assessment have
been incorporated in the report deposited in the departmental library. The Self Study report has been
approved as it satisfies the academic requirements in respect of Self Study work prescribed by the
institution for the said degree.
Mr. Girish Rao Salanke N S, Dr. Sharvani G. S, Assistant Professor, CSE Associate Professor, CSE
Ms. Kowcika A, Mr. Anjan K, Assistant Professor, CSE Assistant Professor, CSE
Dr. Shobha GHead of Department,Department of CSE,
R.V. College of Engineering, Bangalore-560059
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page ii
R.V.COLLEGE OF ENGINEERING
Table of Contents:
Abstract 5
1. Problem Statement 6
1.1. What is DSRC and RSU?
2. Analysis 7
3. Design.
3.1. : Implementing intelligent traffic system DLD Component 9
3.2. Design of intelligent traffic controller using embeded system COA Component 13
3.3. CCN trafic Optimization for iot DSC Component 14
3.4. Vision based intelligent trafic management system DSC Coponent 18
4. Implementation 19
5. Future work 20
6. References 21
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page iii
R.V.COLLEGE OF ENGINEERING
List of Pictures:
1. The Graph of Sales High level system architecture with communication
description. 6
2. Measurements visualized on a mobile interface
7
3. Pin diagram for different components.
8
4. A Jumbo Ethernet frame
10
5. Block Schematic of Intelligent Traffic Light Controller with GSM Interface.
6. Optimization problem and Basic Forwarding scheme
7. Sampling optimization 14
8. Dynamically Updated Backgrounds for various values of background constant ‘v’.
9. Traffic Management flowchart.
10. . Background subtraction flowchart
11. Graphic User Interface
12. The concept of MITCS application
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page iv
R.V.COLLEGE OF ENGINEERING
13. Autonomous area
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page v
R.V.COLLEGE OF ENGINEERING
Abstract:
This paper suggests a new schema for applying the IoT (Internet of Things) to intelligent traffic
systems. The intelligent traffic system is implemented using road side units (RSU) with friction
monitoring, vehicles with environmental sensors and a database for data transfer through
different platforms. The system is able to collect sensor data from stationary RSU stations or
moving vehicles and store it to the database. The test results indicate that the developed IcOR
friction monitoring unit is able to distinguish the different road weather categories (ice, snow,
wet and dry asphalt) with sufficient accuracy. Communication is implemented using a V2I/I2V
IEEE 802.11p communication between RSUs and vehicles or 3G/4G mobile connections. In this
article, we describe an implemented IoT ITS concept with current real-life implementation and
future plans.
In recent years, Internet of Things (IoT) has become the hottest issues of Future Internet. It is the
most important concept of Future Internet for providing a common global IT Platform to
combine seamless networks and networked things. However, there is a lack of common fabric
for integrating IoT with current Internet. That results the service providers and operators have no
definite specification to follow. In this study, we construct a simulated bootstrap platform to
provide the discussion of open challenges and solutions for deploying IoT in Future Internet. The
service providers and operators can estimate their migration to IoT by referring to our experience
and experiment results.
.
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page vi
R.V.COLLEGE OF ENGINEERING
Problem Statement:
The Internet of Things (IoT) is a recent communication paradigm that envisions a near future, in
which the objects of everyday life will be equipped with microcontrollers, transceivers for digital
communication, and suitable protocol stacks that will make them able to communicate with one
another and with the users, becoming an integral part of the Internet.
1.1. What is DSRC and RSU?
Dedicated short-range communications are one-way or two-way short-range to medium-
range wireless communication channels specifically designed for automotive use and a
corresponding set of protocols and standards
Vehicles, road side sensor systems and supporting road infrastructure systems gather everyday
information about the traffic environment. Installed road side systems are usually designed to
work independently and provide measurements to only a restricted number of end-users; usually
supporting road maintenance.
In addition to road side sensor systems, new vehicles are equipped with several driver assistance
sensor systems that measure the environment outside the vehicle. Sensor information to the
driver from the traffic environment is limited to car sensors although new mobile phones and
navigators are capable of receiving information almost in real time. In addition, new
communication technologies like the IEEE 802.11p standard for vehicle to vehicle to
communication is available
Measurements from vehicles and RSU are visualized on a user-friendly map. Marking the
measurement location and measurements on a map does the visualization. The novelty of our
approach is on the schema of using existing sensor system and communication platforms to
implement an intelligent traffic system. Vehicle to vehicle communication with road friction
monitoring is also introduced in the WiSafeCar project [1]. Road weather monitoring from the
moving vehicle is introduced in same project [2]. V2I communication is used in the
INTERSAFE-2 project for I2V (Infrastructure to vehicle) communication [3]. The RSU is based
on ASSET concepts together with the IcOR road friction monitoring system
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page vii
R.V.COLLEGE OF ENGINEERING
2. Analysis:
Vehicles and RSU collect measurements and calculate values to be stored in the database, or
send them using 802.11p communication. In addition, RSU allows direct access to measurements
and video images through an HMTL5 interface. The database enables the provision of a map
interface for mobile devices.
Figure 1. High level system architecture with communication description.
The road side unit (RSU) takes images of the road section with a stereo camera and calculates
the road weather type with the use of the IcOR software developed by VTT. From the lookup
table, the system estimates the road friction based on the measurements. The installation of the
RSU unit is shown in Fig.2, i.e. installed on a motorway ramp to monitor the road section at end
of the ramp. The RSU is able to send measurements to vehicles through V2X communication
using CAM/DEMN messages and to the database using 3G mobile connections. One RSU
message contains the measured road condition, the friction value and the GPS location of the
device.
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page viii
R.V.COLLEGE OF ENGINEERING
Vehicles can collect data with various sensors. The vehicle shown in Fig.3 can measure the air
and road surface temperature, ABS/ESP status, 3D-accelerations and the status of the dashboard
controls. In addition, the vehicle contains the same IcOR road friction detection system as the
RSU on the motorway ramp. Vehicles can communicate through V2X communication using
CAM/DEMN with RSU and other vehicles nearby. In addition, vehicles communicate with the
database using mobile 3G connections.
The database is used to store all the measurements from the vehicles and RSU. The intelligent
traffic system database contains weather information from environmental sensors combined with
data from vehicle sensors. It is possible to extend the database so it would contain more accurate
information about, for instance, traffic flows, for emission calculations.
The IoT ITS Pilot uses and provides the user interface for nomadic devices (i.e. smartphones) for
which the penetration is expected to increase in the near future. This will expand cooperative
systems to cover users more specifically, instead of just the vehicles and the infrastructure.
The user interface (Fig 2) shows the locations of vehicles and RSUs (as red, blue, and gray dots).
The user can select a specific unit by clicking on the dot to see more information, i.e. the road
condition or friction value
Figure 2. Measurements visualized on a mobile interface.
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page ix
R.V.COLLEGE OF ENGINEERING
3. Design:
3.1. Implementing Intelligent Traffic Control System for Congestion Control,
Ambulance Clearance
The current problem section, it can be seen that, existing technologies are insufficient to handle
the problems of congestion control, emergency vehicle clearance, stolen vehicle detection, etc.
To solve these problems, we propose to implement our Intelligent Traffic Control System. It
mainly consists of three parts. First part contains automatic signal control system. Here, each
vehicle is equipped with an RFID tag. When it comes in the range of RFID reader, it will send
the signal to the RFID reader. The RFID reader will track how many vehicles have passed
through for a specific period and determines the congestion volume. Accordingly, it sets the
green light duration for that path. Second part is for the emergency vehicle clearance. Here, each
emergency vehicle contains ZigBee transmitter module and the ZigBee receiver will be
implemented at the traffic junction. The buzzer will be switched ON when the vehicle is used for
emergency purpose. This will send the signal through the ZigBee transmitter to the ZigBee
receiver. It will make the traffic light to change to green. Once the ambulance passes through, the
receiver no longer receives the ZigBee signal and the traffic light is turned to red. The third part
is responsible for stolen vehicle detection. Here, when the RFID reader reads the RFID tag, it
compares it to the list of stolen RFIDs. If a match is found, it sends SMS to the police control
room and changes the traffic light to red, so that the vehicle is made to stop in the traffic junction
and local police can take appropriate action. List of components used in the experiment are
CC2500RF module, Microchip PIC16F877A, RFID Reader–125KHz–TTL and SIM300 GSM
module. Figure 2 shows the pin diagrams (or pictures) of components used.
A. ZigBee Module CC2500
The CC2500 is a RF module and has transreceiver, which provides an easy way to use RF
communication at 2.4 GHz. Every CC2500 is equipped with the microcon- troller (PIC
16F877A), which contains Unique Identification Number (UIN). This UIN is based on the
registration num- ber of the vehicle. One of the most important features is serial communication
without any extra hardware and no extra coding. Hence, it is a transreceiver as it provides com-
munication in both directions, but only one direction. The microcontroller and CC2500 always
communicate with the A. ZigBee Module CC2500
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page x
R.V.COLLEGE OF ENGINEERING
The CC2500 is a RF module and has transreceiver, which provides an easy way to use RF
communication at 2.4 GHz. Every CC2500 is equipped with the microcon- troller (PIC
16F877A), which contains Unique Identification Number (UIN). This UIN is based on the
registration num- ber of the vehicle. One of the most important features is serial communication
without any extra hardware and no extra coding. Hence, it is a transreceiver as it provides com-
munication in both directions, but only one direction. The microcontroller and CC2500 always
communicate with the microcontroller via serial communication. Rx pin of CC2500 is connected
to Tx (RC6) of microcontroller and Tx pin of CXC2500 is connected to Rx pin of
microcontroller (RC7). Other two pins are used to energize transreceiver. It is used to transmit
and receive the data at 9600 baud rate. Figure 4.1.a shows the image of transreceiver. Here, we
uses CC2500 ZigBee module and it has transmission range of 20 meters.
B. Microcontroller (PIC16F877A)
Peripheral Interface Control (PIC) 16F series has a lot of advantages as compared to other series.
It executes each instruction in less than 200 nanoseconds. It has 40 pins and has 8K program
memory and 368 byte data memory. It is easy to store and send UINs. At the junction, it is easy
to store large number of emergency vehicles. Before switching to green, it should satisfy all the
conditions. Simple interrupt option gives the advantage like jump from one loop to another loop.
It is easy to switch any time. It consumes less power and operates by vehicle battery itself
without any extra hardware. Figure 2.b shows the PIN Diagram of PIC16F877A.
C. GSM Module SIM 300
Here, a GSM modem is connected with the microcontroller. This allows the computer to use the
GSM modem to com- municate over the mobile network. These GSM modems are most
frequently used to provide mobile Internet connectivity, many of them can also be used for
sending and receiving SMS and MMS messages. GSM modem must support an “extended AT
command set” for sending/receiving SMS messages. GSM modems are a cost effective solution
for receiving SMS mes- sages, because the sender is paying for the message delivery. SIM 300 is
designed for global market and it is a tri-band
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xi
R.V.COLLEGE OF ENGINEERING
Fig. 3(a) ZigBee module CC2500. (b) Pin diagram of PIC16F877A. (c) GSM Module SIM300.
(d) RFID reader–125 kHz–TTL.
GSM engine. It works on frequencies EGSM 900 MHz, DCS 1800 MHz and PCS 1900 MHz.
SIM300 features GPRS multi-slot class 10/ class 8 (optional) and supports the GPRS coding
schemes. This GSM modem is a highly flexible plug and play quad band GSM modem, interface
to RS232, it supports features like voice, data, SMS, GPRS and inte- grated TCP/IP stack. It is
controlled via AT commands (GSM 07.07,07.05 and enhanced AT commands). It uses AC – DC
power adaptor with following ratings DC Voltage: 12V/1A.
D. RFID Reader–125 kHz–TTL
Radio Frequency Identification (RFID) is an IT system that transmits signals without the
presence of physical gadgets in wireless communication. It is categorized under automatic
identification technology, which is well established protocol. The working of an RFID system is
very simple. The system utilizes tags that are attached to various components to be tracked. The
tags store data and information concerning the details of the product of things to be traced. The
reader reads the radio frequency and identifies the tags. The antenna provides the means for the
integrated circuit to transmit its information to the reader. There are two types of RFID
categories, active and passive tags. The tags that do not utilize power are referred to as passive
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xii
R.V.COLLEGE OF ENGINEERING
and they are driven by an antenna that enables the tag to receive electromagnetic waves from a
reader. On the contrary, active tags rely on power and they have inbuilt power sources that
enable it to send and receive signals from RFID reader. RFID range depends on transmit power,
receive sensitivity and efficiency, antenna, frequency, tag orientations, surroundings. Typically,
the RFID range is from a few centimeters to over hundred meters. RFID reader uses frequency
125 KHz with a range of 10 cm.
A. Automatic Signal Control System
In this module, for experiment purpose, we have used passive RFID tags and RFID reader with
frequency 125 KHz. RFID tag, when vehicle comes in the range of the receiver will transmit the
unique RFID to the reader. The microcontroller connected to the RFID reader will count the
RFID tags read in 2 minute duration. For testing purpose, if the count is more than 10, the green
light duration is set to 30 seconds, if count is between 5 and 9, the green light duration is set to
20 seconds. If the count is less than 5, the green light duration is set to 10 seconds. The red light
duration will be for 10 seconds and orange light duration will be for 2 seconds. Figure 3
implementation for automatic signal control and stolen vehicle detection system.
B. Stolen Vehicle Detection System
In this module, for testing purpose, we compare the unique RFID tag read by the RFID reader to
the stolen RFIDs stored in the system. If a match is found, then the traffic signal is immediately
turned to red for a duration of 30 seconds. Also an SMS is sent specifying the RFID number by
using SIM300 GSM module. The LCD display will indicate that stolen vehicle is present as
shown in Figure
fig 4 PIN Diagram for automatic signal control and stolen vehicle detection system.
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xiii
R.V.COLLEGE OF ENGINEERING
3.2. DESIGN OF INTELLIGENT TRAFFIC LIGHT CONTROLLER USING
EMBEDDED SYSTEM
The proposed operations of Intelligent Traffic Light Controller are shown in Figure 1. In this
figure the junctions are shown by letters A to F. The Infrared Sensors to detect vehicles is
mounted on road. The presence or absence of a vehicle is sensed by a sensor assembly mounted
on each road. This acts as an input to the ITLC unit. This input signal indicates the length of
vehicles on each road. The ITLC unit generates output signals for Red, Green and Orange Signal
and monitor their timings taking into considerisation the length of vehicles on each road. The
same information is transmitted to the mobile user which will request for congestion status. If a
vehicle driver at junction send sms on GSM mobile phone to ITLC unit, the driver will get
message indicting congestion status of road. In this case it will inform that junction A is
congested and the best possible route at this instant is Route 1 via junction E. In addition to
above, in the emergency mode, for a vehicle like ambulance, fire fighter or police car, the signals
are altered for the fast and easy movement of these vehicle. Consider Figure 1, if an emergency
vehicle is passing by the route A-B-C-F, the signals on the roads which are crossing this route
will be immediately made red to stop vehicles on these routes. This is a very important feature
which is very useful in case of emergency.
The basic operation of ITLC can be realized by using embedded system which has advantages of
simplicity, user friendly, easily programmable and a facility for GSM mobile interface. In our
proposed model the basic operations are implemented using Microcontroller89c51AT. The main
reason for selecting this microcontroller is ease of programming, sufficient number of input
output lines, manageable size of RAM and ROM and simple architecture. The block diagram of
the proposed model is shown in Figure 2. The heart of the system is microcontroller AT89c51.
For communicating with the external signals additional ports and multiplexers are used.
Additional RAM and ROM are used for storing system program and application program. The
block diagram consists of the microcontroller, input switching matrix, serial communication
interface, GSM interface, Real Time Clock 1307, Clock circuit, Relay Driver ULN 2003, LED
interfacing circuit.
The signals from sensor assembly will be applied to input switching circuit. These input signals
from sensors will be in the form of digital signals which corresponds to presence or absence of a
vehicle. These digital signals from each lane will be given to the input port of microcontroller,
where the microcontroller will determine the length of vehicle at each lane. This information is
the input to microcontroller to determine various timing signals. The on and off time of the four
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xiv
R.V.COLLEGE OF ENGINEERING
junctions will be calculated by microcontroller, in order to keep waiting time minimum. These
signals will be applied to two relay drivers which consist of ULN 2003. These relay drivers are
level shifters and current amplifiers. The output of relay driver is applied to Red, Green and
Orange LED at each junction. IC 24C61 is used for I2C interface. One LCD Display will be
provided with each signal. LCD Display is shown only for prototype mode LCD Display will
indicate the contract LED displays are to be used, which will be visible time left for the signal to
become green i.e. it indicates the time from a longer distance.a vehicle has to wait at a particular
junction.
fig 5 Block Schematic of Intelligent Traffic Light Controller with GSM Interface.
Microcontroller is programmed using Assembly Language. Separate routines are written for
Input section, Relay drivers, LCD Display, GSM interface. All routines are integrated with the
main logic of the system which determines the timing interval at each junction
CCN Traffic Optimization for IoT
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xv
R.V.COLLEGE OF ENGINEERING
The Content-Centric Networking (CCN) recently re- ceived a lot of attention thanks to its
elegant way to optimize content diffusion at the scale of Internet. However, communica- tions
occurring at the edge of Internet, in particular the Internet of Things (IoT), are also a vivid
research topic. Even if CCN was not initially designed to optimize the specific traffic pattern of
the IoT, it can be improved to better support these new CCN is well suited for distributed
environments where each node plays a role in the routing decisions. A node interested in a given
content (consumer) sends an interest request which is forwarded until another node can satisfy it
(producer) by sending a data message using the reverse path. Each CCN node maintains three
vital types of information:
. Problem definition
A common IoT deployment is composed of communicating things in a dedicated network.
Thanks to gateways, such a network is interconnected to the Internet. However, things tends also
to be more and more directly connected, especially using cellular networks, like smartphones or
connected cars.
We promote the use of CCN for low power devices. CCN considers every node as a router and
so enhances direct neigh- bor to neighbor communications as in Figure 1. In addition, CCN is
designed to be incrementally deployed beside IP which is necessary for traversing non-CCN
paths.
However, this paper focuses on the low power devices at left hand in Figure 1. As highlighted
before, CCN is well designed for distributed environments. Data-centric approaches are helpful
for saving resources while the issue of a flat naming scheme is addressed with CCN. In addition,
the mechanisms of CCN can route interests in parallel on multiple paths and faces (and so
communication medium) which thus helps to improve the QoS (opportunistic routing). Besides,
the meaningful nam- ing scheme enables new context-aware routing features[10].
These low power networks are composed of information producers (like sensors) requested by
low power devices (like actuators) and other Internet devices (servers, smartphones, etc).
Therefore, a single information producer may have several information consumers desiring
different granularities. As an example, a temperature sensor can send information to a fire
detection system at a high frequency, every 10 seconds, but such an information provided only
every 5 minutes is enough for automatically adjusting a heating system.
Therefore, the information should not be multi-casted to every consumer at the same regularity
to limit the commu- nication overhead. Sending information temperature every 10 seconds
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xvi
R.V.COLLEGE OF ENGINEERING
during five minutes represents an overhead of 29 useless messages which can be forwarded
through multiple paths and so consume even more resources.
The general scenario is depicted in Figure 2 where there is an intermediate CCN node Ri which
has to forward the
information from M producers transmitted by a previous router Ri−1 towards K destinations
which are accessible through other downstream nodes like Ri+1. The upstream and downstream
routers can be also other producers or consumers as highlighted before. To reduce the number of
messages forwarded by Ri, we consider a sub problem qualified as local regarding one router
and one producer. Resolving the more general problem with multiple producers consists in
resolving multiple times the previous problem.
Formally, the problem is defined as follows. Assuming (1) an intermediate node, a router Ri, (2)
one producer P able to send an information inf every x seconds, (3) N consumers C = c1, . . . ,
cN and (4) that each consumer ci is interested to have inf every x×ii seconds where ii ∈ N
(sampling period), the goal is to minimize the number of messages sent by Ri. Since each sensor
node has a limited capacity, this optimization can only use a limited number of additional
resources Capi. This assumes that a proportion of resources is already reserved for the common
CCN operations.
SAMPLING OPTIMIZATION A. Basic Forwarding strategies
In the following discussions, the targeted router for sam- pling optimization purposes is Ri and
the considered producer is P. For sake of clarity, only one content is supposed for this producer,
named Pcontent , but the approach can be easily applied in parallel to different contents.
By definition, CCN nodes are stateful and only forward data on the back path if an interest was
emitted beforehand. Therefore, a basic usage of CCN requires that information consumers
directly request producers as illustrated in figure 3(a). This strategy is called pull and allows
CCN to send data only when needed (on demand delivery) but the initial interest packet
represents an overhead. However, an interest is not forwarded if a previous similar one has been
already sent, and a given data message is sent only once per face which is then multi-casted by
design. The producer also needs to initialize the FIB of the CCN nodes by announcing the hosted
content. Such a step is not considered in the overhead computation because it is required only
once.
Pushing data over CCN was envisioned in [2] which describes potential solutions and finally
ends up designing a publish-subscribe mechanism where a node interesting in a certain content
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xvii
R.V.COLLEGE OF ENGINEERING
can subscribe for it, but in this case the diffusion is still based on IP. In this paper, we consider a
strategy, push where data is directly transmitted using the FIB. In fact, the subscription is done in
the same way that the registration of contents populating the FIB and the sensors’ data
dissemination is similar to the propagation of interests. In particular, sensors’data are not cached
and corresponds to a one-way message. To do so, a simple option is to disseminate such very
small and ephemeral data directly inside interests like for example
/roomA/temperature/ts=10/value=20 assuming that the temperature sensor has registered the
content /roomA/temperature/. Such an interest is forwarded using the FIB but without creating
any entry in the PIT, since no data will be sent back. This can be easily done using a flag in the
content name or in the interest header. However, a main advantage of creating an entry in the
PIT, in the common usage of CCN, is to avoid routing loops. Unlike [2] that uses standard IP
unicast for avoiding routing loop, we propose to use a timestamp, ts, in the content name in
conjunction with a new field in the FIB called last seen in order to route only the latest content,
which is the most valuable. Every new transmission for a given content will be checked against
the last seen value. More recent data will be forwarded while older obsolete ones will be
discarded, thus avoiding loops and saving bandwidth.
Hence, we assume our aforementioned mechanism. This corresponds to data messages
forwarded in Figure 3(b). However, since the information is transmitted regularly and
independently of the demands, each value out of the sampling period (ii) of a consumer ci is
dropped and represents an over- head of useless messages. Such a mechanism is comparable to
use IP multi-cast where data is forwarded to all subscribers.
fig 6 (a) Optimization problem (b) On demand notification (pull) (c) Subscription based
mechanism (push)
B. Optimal Forwarding strategy
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xviii
R.V.COLLEGE OF ENGINEERING
To reduce the message overhead in an optimal way, the goal is to combine the advantage of both
pulling and pushing modes. The router should only forward data to Ci in pushing mode at the
period t only if Ci has expressed its interest regarding this period: t is a multiple of the sampling
period ii. Hence, when the consumers subscribe, they have to specify their sampling period such
that Ri can keep track of it thanks to a dedicated counter. This is shown in Figure 4(a) where C1
Subscribes to P(name) whit i1 = 2. As highlighted, the counter is directly inserted in the FIB and
updated each time a message of the subscribed data arrives.
However, this optimal strategy needs one counter per subscription. The additional resources Capi
are considered in terms of number of available counters, independently of the implementation.
To limit the number of messages, identical messages are combined into one. Assuming i1 = 2
and i2 = 3, the sixth produced message is not forwarded twice as D1 and D2 but only once as
D1+2 meaning that this data serves both C1 and C2. The different notations of the same message
just help in identifying the purpose of the message. In fact, all these messages are exactly the
same and will be forwarded to both consumers thanks to the propagation mechanism of CCN,
which makes feasible the use of a single message D1+2. This simple mechanism is also applied
to the other strategies described in the next section.
The optimal strategy (one counter per content per con- sumer) cannot be applied with a fixed
Capi value because the number of subscriptions can be very large and the sensors’ resources are
very limited. While not being practical, this strategy is considered as the baseline for evaluating
the performances of the other ones.
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xix
R.V.COLLEGE OF ENGINEERING
(a) Simple filtering (b) GCD-based optimisation (c) Using multiple counters in
IST
Fig. 7: Sampling optimization
Vision Based Intelligent Traffic Management System (DSC)
Vision based intelligent traffic management system is a robust framework that manages the on
road traffic flow in real time by estimating traffic density near traffic signals. We have proposed
a simple yet efficient algorithm to calculate the number of vehicles at various signals on a road to
efficiently manage the traffic by controlling traffic signals to avoid congestion and traffic jam.
The proposed system works by detection of vehicles in video frames acquired by cameras
installed on roads and then perform accurate counting of vehicles at the same time. Dynamic
background subtraction technique and morphological operations for vehicle detection have been
used to achieve better detection efficiency. In order to attain accurate vehicle count in least
possible time, we have used Region of Interest based method for vehicle calculation. The
proposed framework is designed and implemented in several simulation test cases. It is expected
that this work will provide an insight into the design and development of traffic signaling based
system and also serves as a basis for practical implementation of a computer vision technology in
real- time environment. Furthermore, this work also contributes to new design schemes to
increase traffic signaling system’s intelligence.
We have used the dynamic background subtraction method for vehicle detection in a video
sequence. For density estimation, we have defined a region of interest in which the system
calculates no. of vehicles in that particular frame.
A. Vehicle detection
As described above that we have implemented Dynamic background subtraction for vehicle
detection. In this method we use to extract the first frame of the video and consider it as our
background then this background is dynamically updated according to the formula
bgn = (gray * v) * (bgn-1 * (1-v)) (1)
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xx
R.V.COLLEGE OF ENGINEERING
fig 8 (a)For v = 0.2 (b) For v = 0.5 (c) For v = 0.9
. Dynamically Updated Backgrounds for various values of background constant ‘v’
In Equation ‘bgn-1’ is previous background image, ‘bgn is updated background, ‘gray’ is
original video frame and ‘v’ is dynamic background constant.
The value of ‘dynamic background constant’ is adjustable and ranges from 0 to 1. The value of
background constant is tuned according to ifferent working scenarios. If the value of background
constant is adjusted near 0, the updated background has increased impact of previous frames
with a certain weight and vice versa. The elaboration of this background constant is shown in
figure
This background subtraction mechanism can also be done by using static background i.e. by
using a constant background image, but this technique is overturn by intensity changes as light is
the major factor for effecting the image brightness and quality. So this factor has been efficiently
overcome by dynamic enhancement of background image. As the background updates with
every incoming frame, so, it easily compensates for light fluctuations and weather conditions.
This updated background is subtracted from current frame to identify the moving object in a
certain frame.
Figure 9. Background subtraction flowchart.
Figure 10. Traffic Management flowchart.
a. Pre Processing
Incoming video frames are pre-processed and enhanced in order to remove noise i.e. salt and
pepper noise. This noise is voided by using morphological operations. This step is really
important as it decreases the chances of false detection.
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xxi
R.V.COLLEGE OF ENGINEERING
b. Background Subtraction
Background subtraction is an important part of our algorithm because at this vehicle is detected
from an image sequence. Background subtraction is carried out by taking the difference of pre-
processed video frame from dynamically updated background. The dynamically updated
background is used for each video frame to achieve high detection efficiency.
B. Density Estimation
Vehicle counting has been done for those vehicles that pass through a defined area of interest.
This requires less processing as the system does not need to process whole frame for counting
vehicles. When an object is detected in that area of interest, an indicative rectangle is plotted
around it. When a detected object enters in the defined region of interest, another indicative
rectangle is plotted at its boundary vehicles counter is incremented. An important reason for
carrying out this step is to avoid multiple counting of a single vehicle in a video frame.
C. Traffic management
The vehicle count value is updated at the processing hardware after a regular interval of time.
Signal controlling algorithm works on the basis of updated vehicle count. It controls the
electronic traffic
signals according to number of vehicles on all roads, connecting at a specific central point like a
square.
Figure 11: Graphic User Interface
Fault detection capability is also incorporated in this system. There may be an occasion when
camera might fail or due to very poor lighting or weather condition the captured video is not
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xxii
R.V.COLLEGE OF ENGINEERING
good enough to get processed for traffic density estimation. In these cases system detects
irregular traffic patterns or it looks for no vehicle detection in the frames for a longer period of
time. If this situation occurs, the system shifts all electronic signals to fixed times and traffic
signals work in conventional way.
4. Implementation: The New Intelligent Traffic Control System for Taiwan
Ethernet over FiIn Taiwan, the first traffic control system was built in Taipei City some twenty
years ago. Thissystem has not been upgraded in line with the progress in information and
communicationtechnologies. Although new systems have been introduced in other areas of
Taiwan, their framework,function, and control devices are similar to thoseused in Taipei. The
gap between traffic control system andinformation and communication technologies prevents
traffic control systems from benefiting from the advantages of information and communication
technologies, such as devices thatare compact, thin, mobile wireless and so on
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xxiii
R.V.COLLEGE OF ENGINEERING
T
Figure 12. The concept of MITCS application
Non-intrusive vehicle detector, such as optical vehicle detector used in this system, collects
immediate traffic parameters, such as cycle length,split, and offset of traffic light indication,
from the intersections in the same group for the controllers to determine optimal control
Virtual Traffic Police (VTP)
VTP is a fault tolerance which can interconnectadjacent traffic signal controllers to construct a
small-scale control group and implement traffic control when the connections between
controllersand control center are interrupted.
Status Monitor Agent (SMA)
SMA consists of the status monitoring and exception handling procedures which monitor
thesoftware and hardware of MITCS to prevent unstable execution. These procedures keep the
system at work for optimal traffic control. SMA reports malfunctions to control center via VTP,
asshown in Figure 6. Alerts may be issued for thefollowing:
. The execution of software system
. The operation of hardware devices
C. Traffic Control Integration Module (TCIM).
TCIM integrates traffic signal, changeablemessage sign, closed-circuit television, vehicle
detector, and intelligent algorithms to manage thetraffic approaching an intersection. TCIM
allows MITCS to easily install any control device, similar to adding a printer to a desktop
personal computer system. TCIM can also self evaluate traffic dynamics to organize adjacent
intersections into thesame control group for group/area control.
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xxiv
R.V.COLLEGE OF ENGINEERING
Figure 13. Autonomous area
5. Future Work:
FutureInternet
In the future, people and objects will be connected anytime, anyplace, with anything, anyone,
and appropriately utilizing any network and any service. The basic consist of Future Internet in
which is composed of IoT (Internet of Things), IoM (Internet of Media), IoS (Internet of
Services) and IoE (Internet of Enterprises)
Applications of Wireless Sensor Network in Intelligent Traffic System
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xxv
R.V.COLLEGE OF ENGINEERING
Recently, there is a strong interest in developing wireless sensor network (WSN) techniques and
important applications for moving vehicles, to enable WSN communication between roadside
and vehicles or between vehicles. WSN collect data to base station by deploying sensor nodes
that arrange themselves in certain region it is difficult to add energy while the nodes move and
deploy, so saving energy is very important in wireless VSNs
.
6. References:
. [1]. T. Sukuvaara, P. Nurmi, M. Hippi, R. Autio, D. Stepanova, P. Eloranta, L.
Riihentupa and K. Kauvo, “Wireless traffic safety network for incident and weather
information,” 1st ACM International Symposium on Design and Analysis of Intelligent
Vehicular Networks and Applications [DIVANet'11, Miami, FL, 4 November 2011]
. [2]. M. Kutila, P . Pyykönen, K. Kauvo, and P . Eloranta, “In-vehicle sensor data fusion
for road friction monitoring,” IEEE 7th International Conference on Intelligent Computer
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xxvi
R.V.COLLEGE OF ENGINEERING
Communication and Processing (2011 IEEE ICCP) [Cluj-Napoca, Romania, 25-27 Aug. 2011.
IEEE (2011)]
[3]. G. Varaprasad and R. S. D. Wahidabanu, “Flexible routing algorithm for vehicular area
networks,” in Proc. IEEE Conf. Intell. Transp. Syst. Telecommun., Osaka, Japan, 2010, pp. 30–
38.
[5] B. P. Gokulan and D. Srinivasan, “Distributed geometric fuzzy mul- tiagent urban traffic
signal control,” IEEE Trans. Intell. Transp. Syst., vol. 11, no. 3, pp. 714–727, Sep. 2010.
[6]. [1] Liu, “Routing finding by using knowledge about the road network”, IEEE Transactions on
System, man, and Cybernetics- Part A: Systems and Humans. Vol. 27 No. 4, 1997, pp 425-430.
[7]. "Task 1 - Traffic Management Studies for Reconstruction High-Volume Roadways,"
Innovative Pavement Research Foundation, The Texas Transportation Institute, Texas A&M
University System, College Station, Texas, 2002.
[8]. G. Montenegro, N. Kushalnagar, J. Hui, and D. Culler, “Transmission of ipv6 packets over
ieee 802.15.4 networks,” 2007. [Online]. Available: http://tools.ietf.org/html/rfc4944
[9]. A. Carzaniga, M. Papalini, and A. L. Wolf, “Content-based pub- lish/subscribe networking
and information-centric networking,” in ACM SIGCOMM workshop on Information-centric
networking – ICN, 2011
[10]. A.Prati, I. Mikic, R.Cucchiara and M.M Trivedi, Analysis and detection of shadows in
video streams, IEEE Computer Vision and Pattern Recognition Confrence,Volume:2, 8-14
December, 2001.
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xxvii
R.V.COLLEGE OF ENGINEERING
[11]. C.Papageorgiou and T.Poggio. A trainable system for object detection, Int Journal of
Computer Vision, 38:15-33, 2000.
[12]. Department of Transportation, Taiwan, NTCIP-compatible wireless ATMS 1 & 2, 2005-
2006.
[13]. Fongray Frank Young, Jhao-You Hong, Liang-Tay Lin and Jim-Min Lin, "Investigation
Studies for Hierarchical ZigBee WiMAX Combined Networks.", Fourth Workshop on Wireless
Ad Hoc and Sensor Networks, pp. 117-123, 2008
DEPARTMENT OF COMPUTER SCEINCE AND ENGINEERING Page xxviii