TCP QIC: Quadric Increase Congestion Control

Girish Paliwal
[email protected]
Amity University Rajasthan Jaipur
algorithm Implementation and simulation

paper is organized as in the following sections the basic of
Abstract— Now these days most of the devices having internet the TCP in section II, section III related work of the TCP
connections that increased the traffic over the server and TCP congestion control algorithm, section IV simulation setup,
provide the connection-oriented connection. TCP may have a and section V have the simulation result matrix the last
problem with utilizing the full bandwidth of the communication
channel. Numbers of congestion control proposals have been
section is the conclusion of the present quadric congestion
suggested to reduce this problem. This paper presents the control algorithm performance.
implementation of a quadric increase congestion control
algorithm and it’s a simulation through the ns3. This algorithm
is based on binary increased congestion control algorithm (TCP II.TCP BASICS
BIC). In ns3 TCP QIC is separately implemented and tested
with the different congestion control algorithms. These This provides a connection-oriented connection between
congestion control algorithms are TCP Westwood, BIC, sender and receiver. The Connection-oriented communication
NewReno, scalable and Illinois. The performance of the TCP means TCP send packets between computers and provides
QIC having the significance over the other congestion control acknowledgment number through the sequence number.
algorithms in respect of throughput, goodput, delay variance Unacknowledged TCP packets are retransmitted based on the
and round-trip time.
information about successfully received sequence number and
Keywords— Congestion control algorithm, TCP, NS3, TCP in some cases, round trip time out of the packet is used for
QIC: Quadric Increase Congestion control algorithm. retransmission of the packet. TCP congestion control
algorithm proposed over the last 30 years roughly to maximize
the throughput and avoiding the congestion at the internet
I. INTRODUCTION routers and provide the services effectively and efficiently.
Over the past decades, so much work has been already done to TCP congestion control algorithms have been having some
enhance the TCP performance and the result many different common features these features are on the basis of changing in
congestion control algorithms. The number of congestion congestion window acknowledgment received round trip time
control algorithms are already developed and implemented of a packet. On the basis of these parameters each and every
after finding some enhancement result over the existing TCP congestion control algorithm wants to provide good
congestion control algorithms [1]. The implementation and throughput and Plus latency time and use the maximum
comparing them with other congestion control algorithms is capacity of the channel [5].
troublesome without network simulators. In the real world TCP congestion window control the number of packets a
scenario, it is very difficult to compare algorithm performance TCP flow they have in the network at any time the congestion
on the basis of actual devices it is possible with the simulators window is set using additive increase multiplicative decrease
it is less time to consume and less expensive and given the AMD mechanism that probes the available bandwidth
moral as actual results. Computer simulator provides valuable dynamically adapting to changing network conditions the
insight into potential implementation before extending the AMD mechanism work well when the sender continuously has
considerable effort that may be required to build a technology data to send as is typically the case for TCP used for bulk data
simulator allow us to measure and compare the existing transfer current TCP implementation have a range of the
algorithm as well as we can introduce a new algorithm and behaviours for starting up after an ideal period some current
compare in light of existing algorithms. It provides a much TCP implementation slow start after an ideal period longer
more flexible way to implement a new algorithm and finding than the RTO estimates as suggested in [5] RFC 2581 while
out the related result on the basis of some existing networking other implementations don’t reduce their congestion window
topology [2]. This paper presents an implementation of the after an ideal period. When the congestion window is reduced
QIC congestion control algorithm that is based on the TCP the slow start threshold remains as a memory of the recent
BIC. In order to the ns3 TCP BIC that is most widely used by condition window. sending of back to back packets are facing
the Linux packets out over the period of a round trip time and old
operating system [3]. I enhance the congestion control congestion window that has not been fully using for some time
algorithm TCP back that use the methodology of the cannot be trusted as an indication of the bandwidth currently
congestion window increase in a binary manner whereas I available for the floor we would contain that the mechanism to
will change a little bit in this algorithm to check the piss out packet allowed by the congestion window are largely
performance of quadric congestion control window increase used to determine the appropriate size of the congestion
the effect is immense to attract our attention. The rest of this window. [6]The Mechanic to determine the number of RTO in
the most recent ideal period could also be implemented by

using the timer that expires every RTO after the last week it
was Centre Street of a check per packet. The detailed description of TCP BIC algorithm could be found
in [9]. A diagram of this algorithm during the Congestion
III.RELATED WORK Avoidance phase is shown in figure 1

This is very challenging task to design a new protocol that

can satisfy all the criteria like in congestion control these
criteria are RTT fairness TCP friendliness and scalability [1,
7]. a protocol should adopt its window control depending on
the size of the window. TCP Bic the congestion control
problem is viewed as a search problem [8].The TCP BIC
binary increase congestion control that consists two parts
binary search increase and additive increase.
A. Binary search increase
Starting points for this search our current maximum
window size and maximum window size usually Wmin and W
max are the windows I just before the last recovery. The
algorithm rapidly computes the midpoint between Wmax and Figure 1 TCP BIC congestion avoidance representation
Wmin set the current window size to the midpoint and check
for feedback in the form of packet loss.
If there is a packet loss and then present threshold IV. SIMULATION SET UP
minimum increments. This technique is called binary search
increase allow bandwidth propping to be more aggressive
In this section, I have evaluated and analyzed the different
initially when the difference from the current window size to
type of congestion control algorithms as TCP Westwood, BIC,
the target window size is large and become less aggressive as
NewReno, scalable, Illinois, and QIC. I am using a simple
the current window size gets closer to the target window size
network scenario and applied through the network simulator
[10].
NS3. To test the behavior of each Protocol in the bottleneck
B. Additive Increase bandwidth network, I used simple simulation Technology
To ensure convergence at RTT fairness the binary search in consisting of a single sender and single receiver
trees having an additive increase strategy also. When the interconnected to each other through a router at the other end
distance from the midpoint to a current minimum is too large this is shown in Figure 2. Here I create a point to point
increase window size directly to the midpoint. When the connection between the sources and sink and simulate
distance from the current window size to the target is less than according to different bandwidth channel between server to
the Smax, at that time window increase directly to the target the router and varying the time of simulation. Each TCP
[10]. congestion control algorithm is analysis on the basis of the
congestion window, Round trip time and finally get the
C.Slow Start throughput. Table 1 shows the network simulator parameters
that used in simulation through NS3. [4].
The TCP congestion control algorithm runs a “slow start”
strategy to probe for a new maximum, once the current 10 Mb/s, 0.45 ms 2Mb/s,0.01ms
window size is greater than Wmax but smaller than
Wmax+Smax. The congestion window increases in each
Figure 2 TCP congestion Simulation network Scenario
RTT round in steps.
D.Fast convergence
TABLE 1
SIMULATION VARIABLES AND ITS VALUES
Under the fast convergence, the binary search increase
combines in multiplicative decrease converges to fair share Parameters Values
search. since the large window reduces more and Bottleneck Bandwidth 2 Mb/s
multiplicative decrease. the fast convergence strategy the Time of Simulation 50 seconds
binary search increase after a window reduction new Queue Type PfifoFastQueueDisc
maximum and minimum are set. than Readjust the new Nodes 2
Delay 0.01 ms
condition window maximum size to be the same as the new TCP Congestion Control TCP Westwood, BIC, NewReno, scalable,
target window which is the midpoint and then we adjust the Algorithms Illinois, and QIC
target and apply binary search this has an effect to reduce the Access Bandwidth 10 Mb/s
increasing rate of the larger window allows the smaller Access Delay 0.45 ms
window to catch up this strategy called first convergence [10].
 V. RESULT EVALUATION MATRIX
Where the performance of the congestion control algorithms
evaluated three major performance metrics is as average
throughput average goodput comma average Queen Delhi and
the average package reception these metrics are calculated on
the bottleneck link in both directions. To extract these values
trade shows are installed on the network device of the router
each time a packet is and cute and the cute or received by the
device the method which calculates performance metrics
assignment of the timestamp to packet as they enqueue and
dequeue is implemented by the call back function.
A. Throughput
It is measured as the ratio of the total amount of received data
and the total simulation run time [11].
Average Throughput =Total data received / Simulation time
B. Goodput Figure 3Throughput comparison between congestion control algorithms in
increasing order
In computer networks, goodput is the application- level
throughput. The amount of data considered excludes protocol
overhead bits as well as retransmitted data packets
C.Delay (End to End)
dend-end= N[ dtrans+dprop+dproc+dqueue]
where
dend-end= end-to-end delay
dtrans= transmission delay
dprop= propagation delay
dproc= processing delay
dqueue= Queuing delay
N= number of links (Number of routers - 1)
Note: we have neglected queuing delays.
The average end-to-end delay measured by the total time
taken by sent packets, divided by the total time taken by
received packets.

D.Packet Delivery Ratio

Figure 4 Goodput graph for bottleneck bandwidth 2Mb/s in increasing order
It is a ratio of a packet generated by the source and the number
of packets delivered to the destination [11]. In figure 4 represent the goodput graph. The throughput and
Packet delivery time = Transmission time + Propagation delay goodput result of the simulation 50 seconds for 2 Mb/s
bottleneck bandwidth. I have simulated with different cases of
E. Packet Loss
bandwidth (2,5,10) and simulation time(25,50,80) but the
Number of dropped packed when the buffered becomes full or observation is that the when bandwidth increasing the
stored packed time to live exceeds
congestion windows is increasing significantly and congestion
does not occur for this scenario that’s why here only one
simulation results describe.
VI. TCP QIC ANALYSIS WITH OTHERS
In the figure 5shows the end to end delay in transmission. In
On the basis of the simulation result, it is analyzed that the this graph, the average delay calculated that show the lowest
throughput of the TCP QIC having some significance value average delay TCP Westwood, NewReno, and Scalable and
with comparing to another congestion control algorithms. This highest average delay TCP Illinois and BIC. Whereas TCP
is represented by the figure 3. In this graph the throughput QIC having an average delay between lowest and highest. In
representation in increasing order. this regard, the TCP QIC performance is better than the TCP
BIC and Illinois.
Figure 5 Average end to end delay representation Figure 7 packet send speed
The following figure 6 shows the number of packets sends If data packet send speed higher but receiving speed not
by each TCP congestion control algorithm in increasing order similar to the sending speed in this situation congestion occur
these are represented by the graph. It is very simple logic and data packets are dropped and retransmission of same data
when the number of packets sent by any algorithm then data packets. This is affecting the throughput and average end to
more transmitted and it shows the maximum utilization of the end delay.
bandwidth and produced maximum throughput. It is proved by
the above and below graphs that the TCP QIC is having an
advantage in terms of throughput and data transmission. VII. CONCLUSION
This paper proposes a TCP congestion control algorithm based
on the TCP BIC that is a quadric congestion control (QIC).
TCP QIC is a simulation of the different parameters values
using the ns3 simulator. The simulation metrics are generated
after 50-second simulation each of the congestion control
algorithms simulated same time. The different kind of graph
plotted on the basis of simulation Matrix result that represents
the TCP QIC having better throughput and Goodput. TCP Q
IC having efficient data packet transmission speed and it also
receives a maximum number of packets within a 50 second of
simulation time. TCP QIC average delay time is higher from
lowest delay time but it is lower than the highest delay time.
Finally, I say that the TCP QIC having better performance
than the other TCP congestion control algorithms as
Westwood, Illinois, NewReno, BIC and scalable.

REFERENCES
Figure 6 Packet received in 50 seconds simulation
Figure 7 represents the data packet transmission speed that [1] K. Nagori, M. Balachandran, A. Deepak, M. P.
is represented by the lambda. Lambda is the data packet send Tahiliani, and B. R. Chandavarkar, “Common TCP
Evaluation Suite for ns-3,” pp. 9–16.
per second. If data packet sends per second is higher that
[2] P. Yang et al., “TCP Congestion Avoidance Algorithm
means the data packed acknowledgment received speed is
Identification,” IEEE/ACM Trans. Networking, vol. 22,
higher, that show the congestion prevention and congestion no. 4, pp. 1311–1324, 2014.
control is efficiently handled by the TCP QIC rather than TCP [3] Brett Levasseur, Mark Claypool,Robert Kinicki, “A TCP
BIC, Westwood, NewReno, Illinois and Scalable. CUBIC Implementation in ns-3 ∗,” vol. In Proceedings
of the 2014 Workshop on ns-3 (p. 3). ACM., pp. 1–12,
2014.
[4] L. A. Grieco and S. Mascolo, “Performance evaluation
and comparison of Westwood+, New Reno, and Vegas
TCP congestion control,” SIGCOMM Comput. Commun.
Rev., vol. 34, no. 2, p. 25, 2004.
[5] “RFC 2581,” 2001.
[6] “RFC 3465 - TCP Congestion Control with Appropriate
Byte Counting …ABC—,”
[7] W. Hua and G. Jian, “Analysis of TCP BIC Congestion
Control Implementation,” pp. 781–784.
[8] https://www.nsnam.org/docs/models/html/tcp.html.
[9] SeyedShams Feyzabadi, “Identifying TCP Congestion
Control Mechanisms Using Active Probing,”
[10] L. Xu and K. Harfoush and I. Rhee, “IEEE INFOCOM
2004: The conference on computer communications :
Twenty-third Annual Joint Conference of the IEEE
Computer and Communications Societies : proceedings :
7-11 March, 2004, Hong Kong Convention and
Exhibition Centre, Hong Kong, PR China,” (eng), 2004.
[11] Girish Paliwal and Swapnesh Taterh, “Impact of Dense
Network in MANET Routing Protocols AODV and
DSDV Comparative Analysis Through NS3,” pp. 327–
335, 2017.