1

Using Fuzzy Logic Control to Provide Intelligent Traffic

Management Service for High Speed Networks
Abstract: In view of the fast-growing Internet

exchange of feedback between routers at the borders

traffic, this paper propose a distributed traffic

of a network in order to detect and restrict

management framework, in which routers are

unresponsive traffic flows before they enter the

deployed with intelligent data rate controllers to

network, thereby preventing congestion within the

tackle the traffic mass. Unlike other explicit

network.

traffic control protocols that have to estimate

The fundamental philosophy behind the Internet is

network parameters (e.g., link latency, bottleneck

expressed by the scalability argument: no protocol,

bandwidth, packet loss rate, or the number of

mechanism, or service should be introduced into the

flows) in order to compute the allowed source

Internet if it does not scale well. A key corollary to the

sending rate, our fuzzy-logic-based controller can

scalability argument is the end-to-end argument: to

measure the router queue size directly; hence it

maintain scalability, algorithmic complexity should be

avoids various potential performance problems

pushed to the edges of the network whenever possible.

arising

while

Perhaps the best example of the Internet philosophy is

reducing much consumption of computation and

TCP congestion control, which is implemented

memory resources in routers. As a network

primarily through algorithms

parameter, the queue size can be accurately

systems. Unfortunately, TCP congestion control also

monitored and used to proactively decide if action

illustrates some of the shortcomings of the end-to-end

should be taken to regulate the source sending

argument.

from

parameter

estimations

rate, thus increasing the resilience of the network

to traffic congestion. The communication QoS

operating at end

II SYSTEM DESIGN

(Quality of Service) is assured by the good

The architectural design is the process of

performances of our scheme such as max-min

establishing a framework of the system. The software

fairness, low queuing delay and good robustness

requirements are transformed into definitions of

to network dynamics.

software components and their interfaces, to establish

the framework of the software. Abstracting the

I INTRODUCTION

essential details and ignoring non-essentials obtains a

The Internets excellent scalability and robustness

simplified description of the architecture. The model

result in part from the end-to-end nature of Internet

defines the software components, presents a hierarchy

congestion control. End-to-end congestion control

of control, uses implementation terminology and

algorithms alone, however, are unable to prevent the

shows control and data flow.

congestion collapse and unfairness created by

applications that are unresponsive to congestion.
To address these maladies, we propose and
investigate a novel congestion-avoidance mechanism
called Congestion Free Router (CFR). CFR entails the

III. IMPLEMENTATION
A. LEAKY BUCKET ALGORITHM
The leaky bucket algorithm is an algorithm used in
packet

switched

computer

networks

and

tele-

communication networks. It can be used to check that

data transmissions, in the form of packets, conform to
defined limits on bandwidth and burstiness (a measure
of the unevenness or variations in the traffic flow). It
can also be used as a scheduling algorithm to
determine the timing of transmissions that will comply
with the limits set for the bandwidth and burstiness:

Fig. 1 Architectural Design

see network scheduler.

The source takes the messages, converts them into

packets and forwards it to the In-Router. The InRouter receives the packets and routes it to the Out-

The variation in leaky bucket for the system is that

we make use of dynamic leaky bucket.

Router to check the rate of the Destination. The OutRouter is incorporated with a time-sliding window that
retrieves the rate of the Destination node and
computes the delay for In-Router. Hence, the OutRouter sends a Backward Feedback about the delay to
the In-Router. The In-Router incorporates the LeakyBucket Algorithm and a Rate-Control Algorithm. The
Rate-Control Algorithm receives the BackwardFeedback that contains the delay and then computes
the time of leaving for the packets. If the rate is too
large, then the leaky bucket holds the packets until its
turn to leave. The packet is then forwarded to the
Core-router (Router) and thereby to the Out-Router as
well. The Out-Router then identifies the destination
with the given IP address and routes the packet to the
correct Destination. The message is finally received

Fig. 2 Flowchart of Leaky Bucket Algorithm

by the Destination and displayed on a specific

interface, if needed.
The

variation

here,

with

respect

to

the

B. RATE-CONTROL ALGORITHM

conventional leaky bucket is that, this proposed

This algorithm is deployed in the In-Router of the

system uses a dynamically sized leaky bucket that

proposed system. The Rate-Control Algorithm waits

adjusts its size according to the rate of the Destination.

for the Backward Feedback from the Out-Router. This

Hence, if the Destination is too slow, the bucket size

feedback includes the rate of the destination in the

increases to hold more number of packets.

form of delay. The Rate-Control Algorithm then

computes the rate of leaving of the packets from the

leaky bucket by making use of the delay in milliseconds. This helps the packets to be lined in the leaky
bucket, so that they need not wait for a very long time,
at the same time they should not be flooded on to the
congested links. This algorithm computes the time of
leaving for packets on the basis of the round trip and
the previous time for receiving the feedback.
C. TIME-SLIDING WINDOW
This algorithm checks for the rate of the
destination, computes the delay and sends a backward
feedback

to the In-Router. This algorithm

is

implemented in the Out-router.

IV SNAPSHOTS
Source:

V ADVANTAGES
The first and foremost advantage of this system is
that it designs a novel explicit rate-based trafficmanagement scheme, called the IntelRate Controller
for high speed IP networks. This serves as a
congestion avoidance scheme that completely avoids
congestion and there by eliminates packet loss.
The next advantage is that it uses a Fuzzy
Smoother mechanism that can generate relatively
smooth flow throughput. This increases the throughput
to an extent due to the elimination of packet loss and
thereby improves the performance.
VI CONCLUSION
In this paper, we have presented a novel
congestion-avoidance mechanism for the Internet
called CFR and an ECSFQ (Enhanced Core-Stateless
Fair Queuing) mechanism. Unlike, existing Internet

Destination:

congestion control approaches, which rely solely on

end-to-end control, CFR is able to prevent congestion
collapse

from

undelivered

packets.

ECSFQ

compliments CFR by providing fair bandwidth

allocations in a core-stateless fashion.

CFR ensures at the border of the network that each

flows packets do not enter the network faster than

jitter in transferring such packets would completely

bring down the quality of the audio or video.

they are able to leave it, while ECSFQ ensures, at the

VIII REFERENCES

core of the network that flows transmitting at a rate

lower than their fair share experience no congestion,

[1] Jungang Liu, Oliver W.W.Yang, Using Fuzzy

i.e., low network queuing delay. This allows the

Logic

transmission rate of all flows to converge to the

Management Services for High Speed Networks,

network fair share. CFR requires no modifications to

IEEE, 2013

Control

to

Provide

Intelligent

Traffic

core routers nor to end systems.

[2] K.M.Pasino and S.Yurkovich, Fuzzy Control,
Only edge routers are enhanced so that they can

Addison Wesley Longman, 1998

perform the requisite per-flow monitoring, per-flow

rate-control and feedback exchange operations, while

[3]

ECSFQ requires a simple core-stateless modification

Managing Internet Traffic. John Wiley & sons Ltd.,

to core routers. Simulation results show that CFR

successfully

prevents

congestion

collapse

from

undelivered packets. They also show that, while CFR

is unable to eliminate unfairness on its own, it is able
to achieve approximate global max-min fairness in a
completely core-stateless fashion.
VII FUTURE ENHANCEMENTS
In real time this technique of use Fuzzy Logic can
be used for the congestion avoidance to provide better
throughput and to eliminate the packet loss. The
IntelRate Controller is a novel router design for
Congestion Free Routers. The special design helps to
adapt to the present high speed networks and there by
works well with a wired or wireless LAN. This can be
improved to adapt to Mobile Adhoc Networks where
the mobility is the main concern and that requires a
highly efficient congestion avoidance mechanism.
Adaptation to the wireless networks is one of the
improvements.
Currently the implemented project only takes the
text area as its input from any of the nodes. This is
basically to transfer text messages. This can further be
improved to take real time traffic as the input. The
audio and the video files are the main cause for
congestion these days in high speed networks. Any

M.Welzel,

Network

Congestion

Control: