Improving video streaming through increasing protocol

Sisa Juma
[COMPANY NAME] [Company address]
IMPROVING VIDEO STREAMING THROUGH INCREASING
PROTOCOL EFFICIENCIES
[Type here]

Improving video streaming through increasing protocol efficiencies
Abstract
The research is on the improvement of video streaming by increasing the efficiency of the
protocols. For a successful HTTP adaptive solution of streaming video, it must be able to ensure
that the objectives are met as follows: very high utilization of the resources of the network,
farness, stability, as well as convergence over a short period of time to the fair share at the same
time avoiding the issues of buffering. For this reason, this particular paper proposes HTTP
solution which achieves the objectives set by very close relation between the clients and the
servers. With dependence on the data which is present from both the client and the server, the
control of the efficiency, stability and fairness is delegated to the client and server side. From the
client side, the bandwidth available, the history of the bitrate played and the occupancy of buffer
are well known hence making it best for the control of the utilization of the bandwidth,
management of buffer as well as stability. The fairness in contrary, is controlled at the server
because the connected client numbers together with their bitrates played as well as the capacity of
the bottleneck link are known from the server side.

Improving video streaming through increasing protocol efficiencies
Table of Contents
Abstract........................................................................................................................................................1
CHAPTER ONE...............................................................................................................................................5
1.0 Introduction............................................................................................................................................5
1.1 Background.........................................................................................................................................5
1.2 Aim.....................................................................................................................................................7
1.3 Objectives...........................................................................................................................................7
1.4 Research Brief....................................................................................................................................8
1.4.2 Research Overview......................................................................................................................8
1.4.2 Scope...........................................................................................................................................8
1.4.3 Research Problem........................................................................................................................9
1.5 Contribution.......................................................................................................................................9
1.6 Research Organization......................................................................................................................10
CHAPTER TWO............................................................................................................................................12
2.0 LITERATURE REVIEW.............................................................................................................................12
2.1 Introduction......................................................................................................................................12
2.1 Related Studies on Video Streaming................................................................................................12
2.3 Real Time Streaming Protocol..........................................................................................................14
2.4 WebReal Time Communication........................................................................................................15
2.5.1 Distorted Standards....................................................................................................................17
2.5.2 Security threats.........................................................................................................................18
2.5.3 Bandwidth Challenge................................................................................................................18
2.5.4 Congestion of Network..............................................................................................................18
2.5.5 Improving These Inefficiencies.................................................................................................19
CHAPTER THREE.........................................................................................................................................21
3.0 METHODOLOGY....................................................................................................................................21
3.1 Introduction......................................................................................................................................21
3.2 Research Problem and Philosophical Questions...............................................................................21
3.2 Selection of Research Methodology.................................................................................................22
3.3.1 Real Time Streaming Protocol...................................................................................................22
3.3.2 WebRTC....................................................................................................................................23
CHAPTER FOUR...........................................................................................................................................27
4.0 EXPERIMENTS.......................................................................................................................................27

Improving video streaming through increasing protocol efficiencies
4.1 Introduction......................................................................................................................................27
4.2 Experimental Design.........................................................................................................................27
4.3 Client Side Controller........................................................................................................................28
4.3.1 Bandwidth Estimator.................................................................................................................28
4.3.2 Buffer Controller.......................................................................................................................29
4.3.3 Efficiency Controller.................................................................................................................29
4.3.4 Stability.....................................................................................................................................29
4.4 Server Side Controller.......................................................................................................................30
4.5 Evaluation Metrics............................................................................................................................31
4.6 Data Collection.................................................................................................................................31
4.7 Data Analysis....................................................................................................................................32
CHAPTER FIVE.............................................................................................................................................33
5.0 RESULTS................................................................................................................................................33
5.1 Introduction......................................................................................................................................33
5.2 System Configuration.......................................................................................................................33
5.3 Presentation Of Results....................................................................................................................34
5.4 Discussions.......................................................................................................................................45
5.4.1 RTSP Streaming Platform.........................................................................................................45
5.4.2 Direct WebRTC Streaming Platform.........................................................................................46
5.4.3 Analysis of Smartphone to Web streaming Applications...........................................................47
5.5.4 Analysis of smartphones to smartphone streaming applications..................................................48
CHAPTER SIX...............................................................................................................................................50
6.0 CONCLUSION........................................................................................................................................50
6.1 Conclusion........................................................................................................................................50
6.2 Recommendation.............................................................................................................................52
6.3 Future Work.....................................................................................................................................53
REFERENCES...............................................................................................................................................54

Improving video streaming through increasing protocol efficiencies
Table of Figures
Figure 1: Web Real-Time Communication request order...........................................................................25
Figure 2: Network topology.......................................................................................................................28
Figure 3: Star Topology.............................................................................................................................34
Figure 4: The played representation levels of the different clients.............................................................36
Figure 5: The efficiency and fairness evaluation of ESTC, FESTIVE and PANDA..................................37
Figure 6: The comparison of the allocated bitrates to Client 1 between ESTC, FESTIVE and PANDA....38
Figure 7: The comparison of the allocated bitrates to Client 2 between ESTC, FESTIVE and PANDA....39
Figure 8: The comparison of the allocated bitrates to Client 3 between ESTC, FESTIVE and PANDA....39
Figure 9: The comparison of the allocated bitrates to client 4 between ESTC, FESTIVE and PANDA.....39
Figure 10: The buffer occupancy evaluation of the different clients..........................................................40
Figure 11: The evaluation of the clients’ stability in case of ESTC, FESTIVE..........................................41
Figure 12: Comparison of the overall stability, fairness and efficiency metrics.........................................42
Figure 13: The efficiency comparison between ESTC, FESTIVE and PANDA when running larger
number of clients........................................................................................................................................43
Figure 14: The fairness comparison between ESTC, FESTIVE and PANDA when running larger number
of clients.....................................................................................................................................................43
Figure 15: Comparison of the estimated bandwidth for the CBR video.....................................................44
Figure 16: Comparison of the estimated bandwidth for the VBR video.....................................................44
Figure 17: General flow of streaming platform..........................................................................................47
Figure 18: Average delay time...................................................................................................................48
Figure 19: Average stream delay time between smartphone to smartphones..............................................50
List of Tables
Table 1: THE LINK CAPACITIES AND ARRIVAL TIMES...................................................................34
List of Equations
Equation 1: Client Side Controller.............................................................................................................30
Equation 2: Server Side Controller.............................................................................................................31

Improving video streaming through increasing protocol efficiencies
CHAPTER ONE
1.0 INTRODUCTION
1.1 Background
Video calling gives flexibility even if one is outside; he can start a video calling on a laptop,
iPod, and on other mobile devices (Abelson et al., 2015). Besides meetings and business
purposes, video calling is used for remote working such as interview, e-learning, and
telecommuting. However, video calls are no more a novelty which is affordable to few; it has
become a practical option in everyone life. As far as organizations are concerned, many big
industries rely on the video calling for everything from management of the project to the
interfacing their other apex executives. By showing oneself and seeing others, one can work the
fascination of body language, which is very important in business as well as other activities
engaging human interaction. Seeing someone while communicating them changes taxonomy of a
conversation, whether it is in business or personal relationship. Video calling between companies
as well as loved ones is becoming a more frequent place due to the progression in technology.
The internet has become an essential part of almost every industry as well as for everyone life. A
strong internet connection is required for effective video calling. Voice over internet protocol is
one of the most popular standards for video calling. Usually, video call depends on the
advancement of the technology used for it. There are various factors which impact the quality of
video callings, such as internet speed, devices, and bandwidth. At least 1.2 Mbps speed is
required for High clarity video calling. Video calling is highly dependent on the network
connections (Neustaedter et al., 2015). Bandwidth issues affect the transmission of video, files as

Improving video streaming through increasing protocol efficiencies
well as sound, which leads to an interruption in communication or cooperation. For getting the
best experiences of video calling, it is necessary to comprehend video calling bandwidth, which
represents the capacity of the connection. The common problem while video calling is poor
internet speed or absence of bandwidth. A slow internet connection always interrupts in the video
calling. Bandwidth, as well as internet speed, is used by the ISP’s as well as by customers. The
relationship between the two determines the quality of video calling.
Video calls are usually done via a computer’s webcam or electronic devices such as Smartphone
or computer or tablet (Rowles, 2017). It may include point to point communication or multipoint
communication. Video calls are conducted using software apps. Though there are a number of
research that has been conducted with regard to the video calling and its importance in the life of
people, business and corporations, however, there is limited research on the inefficiencies of the
systems that are used in video calling or for the smooth functioning of the video calling system.
In this report, discussion about smooth video calling using the slow internet will be done. In
addition to this, the illustration will be given with regard to problems occurring due to the slow
internet connection in video calling and how to tackle these problems instead of having a slow
internet connection. It is found by a study that video calling has been adopted universally for
social use among the friends as well as families as it has become more accessible as well as easy
to use. One out of three persons does video calling once in a week. It has become an equal
alternative of voice calling.
There are several reasons due to which high-speed internet connection are inefficient to perform
up to the benchmark level. The internet provider could be at fault. The old router, as well as the
outdated modem, can be one of the reasons behind the slow internet speed. The internet service

Improving video streaming through increasing protocol efficiencies
provider may have an issue regarding routing signals to the location. The slow internet hampers
the process of smooth video calling.
1.2 Aim
The aim of this research paper is to evaluate the importance of video calling and the factors
needed for a smooth video calling. As the internet speed is considered as one of the major factors
for the smooth video calling, but this is not right all the times. There are various factors on which
the internet speed depends. The report aims to find the reason behind the slow internet connection
and the problems occurring in video calling due to the slow internet speed. The research paper
proposes a recommendation for the appropriate algorithm required to improve the efficiency of
internet speed and smooth video calling.
1.3 Objectives
In order to fulfill the aims of the research paper, the objectives are:
• To assess the significance of the appropriate algorithm in the smooth functioning
of internet connectivity.
• To identify the reasons for the failure of the systems used for internet and video
calling.
• To evaluate the factors for improving the speed of internet connectivity.
• To discuss the relationship between the bandwidth and the speed of internet
connectivity.
• To examine the reasons behind the slow working of internet connection in spite of
being a high-speed internet connection.
• To discuss the various approaches which have been used until now are to be used
in the future.

Improving video streaming through increasing protocol efficiencies
• To determine the methods to fasten the internet connectivity using the new
technologies.
1.4 Research Brief
1.4.2 Research Overview
There was a time when video calling was used only by the large organization, but it has become
one of the mainstream ways of communication as well as the collaboration of companies and
people of this era. Video calling has full-fledged in use exponentially throughout the period of the
last decade (Gardner et al., 2015). Almost every industrial sector such as financing, education,
medicine, military use the video calling for their professional purpose (Carla et al, 2016) Video
calling is an efficient way of real-time conversation that makes the people able to communicate
the persons across the distance in a very productive as well as in a convenient way. It has become
more popular for both professional as well as personal use (Fernández et al., 2014). To eliminate
the time as well as space barriers, it is conducted to contact with the partners, colleagues, or
customers anywhere and anytime.
1.4.2 Scope
The scope of this research will extend to evaluate the video calling process using a slow internet
connection and how to fasten the speed of internet connection. This research paper will act as a
call for all the research community as well as industries to look at the processes to fix the
challenges as well as complexities of video calling and how to find most of the possible
opportunities and algorithm that could be implemented.

Improving video streaming through increasing protocol efficiencies
1.4.3 Research Problem
The problem that has been identified studying the literature and research papers is that slow video
calling is mainly due to slow connectivity and internet issues. At the present era, communication
has been one of the important paradigms that has facilitated innovation and smooth functioning
of all organizations, corporates and other sections. The interruption in video calling has slowed
down the communication system thereby causing communication gap at distant place. Slower
connectivity not only distorts the process of video calling but also slower down communication
over voice calls as well.
1.5 Contribution
In this research various protocols as well as the former research with regard to the reasons for
slow internet connectivity and video calling will be researched altogether under one shed. This
would provide an insight into the research gap pertaining to the smooth internet connectivity and
video calling as well as other forms of communication using internet and network connections.
This research will serve as the base academic literature. This paper will also involve a full
assessment of commonly used video streaming protocol, emphasizing on the WebRTC and the
RTSP protocol. The research will serve as an example for the new academic researchers that
would enhance suggested video streaming platform. As the findings are scientificbased and
hence, the results as well as the information will also be reliable.

Improving video streaming through increasing protocol efficiencies
1.6 Research Organization
The research report will have 6 chapters as described below;
Chapter One -Introduction
This is the first section of the report which cntain the background infromation of the research
topic, the aim and objectives of carrying out the research, the scope and the contribution of the
research are also discussedin this section.
Chapter Two- Literature Review
This is the section which contains all the previous researches which were done by previous
researches and the gap of the research is also deduced from this section.
Chapter Three- Methodology
In this section, the technique which is utilized in collection of raw data related to the topic under
investigation is discussed here,
Chapter Four- Experiments
In this section, the data collected are tested and experimented in order to get the analysis of the
research topic for the purpose of drawing conclusion, the experimental design, data collection and
analysis are discussed here.
Chapter Five- Results
In this section, the data found out in the experimentation section is analyzed and presented for
conclusion drawing and understanding the experimental purpose.
Chapter Six-Conclusion

Improving video streaming through increasing protocol efficiencies
This is the last chapter of the report. The section summarized the whole report, main na dmojor
points are presented in the conclusion section. The section also presents the recommendations
and the future plans of the topic under investigation.

Improving video streaming through increasing protocol efficiencies
CHAPTER TWO
2.0 LITERATURE REVIEW
2.1 Introduction
With the introduction of the internet and the mobile network, the world has indeed become a
smaller place to live in. There is a number of changes that took place throughout the past few
decades in network usage and network types. From 1sdt generation networking to 2nd generation
and now to the latest generation that is unable to accept the traditional system of calling and
messaging. Services such as texting have also been introduced. At present, 5G has taken over all
over generation networks by providing a speed up to 100MB per second for the new generation
customers lacking patience in their search and networking tasks. Many of the nations still lack
even 3G speeds or any change in their bandwidth speed that can help in facilitating smooth video
calls and networking through media platforms. Besides this, it has been evident that having own
router makes it difficult for voice over internet protocol service provider to study or examine the
router as well as troubleshoot the issues from any remote locations (MegaMarketing, 2017).
Different locations with dissimilar internet speeds can have a significant impact on network
fluctuation can cause failure to place a video call.
2.1 Related Studies on Video Streaming
According to the point of view of the scholars, Steel et al. (2011) digital connection, or rather the
connection across mobile networks has become an integral part of our daily lives. Starting from
the 1st generation of mobile networks to the fourth generation of mobile networks, the world has
come a long way in this sphere starting from voice calling which later evolved into texting in the

Improving video streaming through increasing protocol efficiencies
following generation and then the provision of video calling which can be found in the 4G
devices using LTE technology. Goldman et al. (2011), on the other hand, contends that video call
is of extensive use in the present scenario and is abundant in personal as well as corporate or
business use. According to the point of view of the scholars_ Pheasant et al. (2010) while the
personal use systems require a lucid point to point connection, the corporate usage calls involving
more than two people require a multi-point connection. The main problem that video calls go
through is the bandwidth size, which severs the connection every now and then. Another notable
issue that can be raised is that a lot of people are not yet exposed to the latest generations of
networking and that obstructs a hassle-free mobile networking experience. However, these are
the aspects that are seen on the surface level. Richardson (2011), on the other hand, contends that
there exists an integrated set of functions and protocols that go behind the streaming of these
videos that the users get to see during the video calling experience. The system of communication
is evolving daily, and streaming of videos can be undoubtedly pinned down as the one, most
famous and evolving. This carved out a way for itself because communication through videos
diminishes the physical distance that exists between people.
It also ensures point to point as well as multipoint communication. One of the most significant
aspects that the users look for while resorting to this kind of communication is a high quality of
experience and service. Even though communication through videos brings the world closer,
there are still some drawbacks or inefficiencies they face. According to the point of view of the
scholars Santos-González et al. (2017), the size of the bandwidth, level of reliability, congestion
of network, et cetera turn out to be obstructions in the way of flawless communication. In this
report, what shall be taken into account are the two protocols that were most widely used in video
streaming-Real Time Streaming Protocol (RTSP), and Web Real-Time Communication
(WebRTC). Their inefficiencies shall be reflected upon and also the algorithms that can be used

Improving video streaming through increasing protocol efficiencies
to solve them. A number of research, though, have been conducted in this paradigm. However,
there is no adequate reasons or evidence that have been presented by academics. This report
covers all the instances and inefficiencies of the networks and systems that lead to an interruption
in the video calling or networking speed. Examples of relevant applications have been presented
so as to provide a complete scenario with regard to the usage of the networking systems by such
applications in the real world and the problems faced by the users.
2.3 Real Time Streaming Protocol
The Real Time Streaming Protocol (RTSP) is a network protocol or a communication system of
network working at the application strata (Austerberry, 2013). From the name itself, an idea can
be derived that how this protocol is related to streaming data. According to the point of view of
the scholars Kokkonis et al. (2017), it transfers data to the device that lie at the end of the
network, and it does so by directly setting up a communication with the server that is streaming
the data. RTSP takes a hold on the streaming of media between the server and the device of the
client or the end user. Yasumoto et al. (2016) on the other hand contends that It works as a
control or a mediator for multimedia like audio or video, but it has to be synchronized according
to time and the media in context should be continuous. But it is worth mentioning that the Real
Time Streaming Protocol does not undertake the action of streaming the multimedia. Instead, it
sets up a communication with the server which does the same. This can be better explained with
an example. If a client is watching a movie online and pauses it, RTSP immediately notifies it to
the server which is concerned with the video streaming.
According to the point of view of the scholars, Diallo et al. (2012), the working of the Real Time
Streaming Protocol can be explained through a fictitious example involving a client and a server.
If the client wants to watch a video on standup comedy and attempts to stream the video from a

Improving video streaming through increasing protocol efficiencies
source, the device of the client by default sends an RTSP request to the respective server which is
in charge of streaming the video. The server, in turn, uses the same protocol to communicate to
the client the sorts of requests it can potentially comply with. Navaz et al. (2015) in their work
contends that Upon knowing the process of placing a request, it sends a media request to the
particular server. The server again comes up with the media description. Another request is made
by the client in terms of setup to which the server communicates the transport mechanism. Now
with all the sending and receiving of requests, the process of setup has reached an end, and the
client can effectively start to stream the video.
Real Time Streaming Protocol mainly started as a protocol enabling the users to stream the
multimedia of their choice directly from the vast expanse of the Internet without the hassle of
actually downloading it to their respective personal devices. However, RTSP is only used for the
purpose of streaming multimedia but also for other uses like video calling, online education et
cetera. According to the point of view of the scholars Yasumoto et al. (2016) Real-Time Transfer
protocol puts to use the same integrations as Hypertext Transfer Protocol. This is the reason why
RTSP is easily functional with the existing networks based on HTTP.
2.4 WebReal Time Communication
WebRTC is an abbreviation for Web Real-Time Communication. One of the biggest advantages
of this protocol was that it did not require any plug-ins for its functionalities. According to the
point of view of the scholars Zeng et al. (2011), if a timeline is taken into account that dates
before Web Real-Time Communication, protocols can be found that offered identical sorts of
functionalities, but they came with a number of external attachment of plugs. WebRTC put to use
a number of APIs, but they were all plug-free. The protocol in context was definitely powerful,
but it was also disruptive. It established a standard upon which the functionalities rested. Chen

Improving video streaming through increasing protocol efficiencies
and Lee (2011) on the other hand contend that another sphere where the protocol in context had
the edge over the others was that it was not only compatible in the browsers accessed through the
desktop but was also compatible in the browsers accessed through mobile phones.
WebRTC showed dynamism in its actions. It could be implemented over a wide range of
activities, but one of the biggest benefits that WebRTC provided was a point to point connection
over a common network (Carullo et al., 2016). However, some of the primary requisites for the
communication process to start was that the browser of both parties should agree to the
communication taking place. Secondly, the browsers of both parties should know how to identify
the location of each other. Thirdly all the security barriers like those of firewall should be
penetrated. Finally, the multimedia intended should be subjected to transmission.
However, if a bit reflective insight is made, it can be seen how there are evident challenges that
exist in terms of the communication over browsers using Web Real-Time Communication. The
first and foremost challenge that arose was the identification of the browsers and detecting their
location. The second challenge was the establishment of a socket connection with the browser of
the other user. Transfer of data from one device to another or unidirectional transfer is less
complicated and easy to obtain. Hence the other challenge that might arise is the transfer of data
from both ends.
For example, if a user wishes to open a certain website, all that the user has to do is to provide the
name of the web page in terms of an HTTP(Hypertext Transfer Protocol) request over a DNS
server and the server replies accordingly by displaying the desired web page. Now taking into
account a different side to the issue where the end user is not a web server. If a person wishes to
connect a video call with a person, there are certain challenges that will arise. The end user is not
a web server, and so the implementation of an HTTP request would not yield the desired results.

Improving video streaming through increasing protocol efficiencies
So accordingly, a protocol must be used that would ensure the bi-directional transfer of
multimedia data of audio and video from both ends without the involvement of any external
server for keeping the essence of security intact. This is the time when WebRTC would be put
into use to curb these challenges and establish the network.
2.5 Inefficiencies of The Protocols
Even though these protocols had been widely used in all sorts of transmission of media, including
the functions of communication and they served in solving a lot of challenges that existed before
their origin, a number of loopholes still kept these protocols from being the best in the genre and
sweep the entirety of the market all by themselves (Rosas & Martínez, 2016). These were the
inefficiencies that paved the way for the emergence of new and improved systems that solved
these existing flaws to a great extent. Bandwidth strength, Security issues, congestion of network,
and the likes are some of the inefficiencies to name a few. Below discussed are some loopholes
of these systems in detail:
2.5.1 Distorted Standards
For a user accessing an application of communication, video calling in this context, one of the
primary requirements would be for the connection to be established in the first place despite all
sorts of Internet speed. For example, Facebook Messenger is one of the most widely used Video
calling applications, and there are viable reasons behind it. Facebook messenger ensures a
communication experience where even though if the connection is slow, the image quality may
temporarily distort for a bit but again gets back to normal with a little stability, but the voice
quality remains intact (Carullo et al., 2016). The protocol working behind these systems ensure
that, but again on the flip side, these protocols work in a way that a phone call cannot even reach
the other in the first place. A real-time example will explain this better. Skype is accessible to

Improving video streaming through increasing protocol efficiencies
people over a large worldwide platform. However, a more constricted network like FaceTime
does not allow more than a remote connection.
2.5.2 Security threats
In the descriptions of these protocols, the process of their functionality is laid down. One of the
most significant elements of these protocols is that the firewall or the security barriers should be
penetrated successfully so that effective communication can be established. But there is a
negative side to it. These protocols ensure communication in most cases but not of the security of
data. Hence, the data that is subjected to transmission if not properly encrypted is subjected to
breach and misappropriation by third parties involved in cyber-crime.
2.5.3 Bandwidth Challenge
Access of information and video calling and other communication activities is a lot easier when
done over a Wi-Fi network which is home-based. However, the success of such effective
communication over these protocols can be compromised to a great extent (Carullo et al., 2016).
This is because the bandwidth requirements over a cellular network are relatively demanding in
comparison to the home-based networks. The different generations of the network used by people
also vary, and that comes in the way of the protocols to function effectively.
2.5.4 Congestion of Network
Prerequisite to any kind of successful communication is an unhindered network. If technical
noise exists inside a potential network, communication cannot take place. This is where the
loopholes of these protocols come into the light. They cannot function effectively if the network
where they are functioning is flanked with any kind of congestion (Carullo et al., 2016). These
congestions might be caused due to over flooding of data over a single network, mismanagement

Improving video streaming through increasing protocol efficiencies
of data traffic, et cetera. Hence, congestion of network diminishes the potentiality of these
protocols.
2.5.5 Improving These Inefficiencies
Since the description, importance, and inefficiencies of the two most widely used protocols have
been extensively talked about, some ways can be laid down which can solve this problem to a
great extent if not completely. According to the point of view of the scholars Santos-González
(2017), it is important because of the growing demand for face to face and video communication
in the prevalent times. This, as a result, called for the drawbacks to being taken into account and
solved so that the users are guaranteed a hassle-free communication experience. The following
ways were diagnosed in the earlier literary work that has been though implemented by the system
engineers, but limited results were attained.
Firstly, end to end delay between the start and the end user should be diminished. This means that
when the packets of information from the sender are transmitted over the server through these
protocols, the delay in the information reaching the end user should be as less as possible.
According to the point of view of the scholars Begen et al. (2011), this can be ensured by
bringing in necessary modifications in the protocols that are being implemented.
Secondly, proper synchronization should exist between the video as well as the audio streaming
through RTSP. It uses the TCP protocol, and thus the packets that are sent should be
synchronized so that when they are put to decoding by the end user, they receive a set of
synchronized data in terms of all the multimedia that it contains.
Thirdly, since the above-described protocols work on the client side, algorithms must be put to
use for minimum loss of packet data that had been transmitted. Because when a protocol cannot
ensure the authenticity of packet data, that means that the client would receive information that is

Improving video streaming through increasing protocol efficiencies
different from the original. According to the point of view of the scholars McGrew et al. (2010),
the authenticity of the information would be compromised, and that is the sphere where the
protocols would need to modify their structural interface to curb the issue.
Fourthly, provisions should be made for allowing the client or the end user to provide feedback.
Feedbacks are important elements for any sort of improvement that is aimed at. Algorithms
should be put to use in such a way that the service quality of these protocols can be put under
effective scrutiny by the clients who can provide real-time feedback on the service provided.
According to the point of view of the scholars Bridge et al. (2009), this would pave the way for
extreme transparency where the shortcomings of these protocols would be in clear view and also
a real-time approach could be undertaken to improve the same.
Fifthly, security is still a sphere which is much talked about while we take transmission of data
into account. As mentioned earlier, protocols like RTSP and WebRTC allow the multi-
dimensional flow of data, but the concept of stern encryption of the same is still questionable. It
is subject to breach, and that violates the essence of privacy that the clients should be assured
with. Hence, the protocols should be introduced with proper encryption techniques and proper
firewalls within themselves, which would keep data from misappropriation. Therefore,
modifications within the protocol in terms of its barriers should be taken up at the earliest so that
the security threats are mitigated to a great extent.

Improving video streaming through increasing protocol efficiencies
CHAPTER THREE
3.0 METHODOLOGY
3.1 Introduction
Research design is basic to every scientific venture. For this research paper, interpretivism
research design has been selected. This design is based on the naturalistic method of data
collection, including interviews as well as observation. The interview is the major part used for
this type of research design. According to the point of view of the scholar Chilisa (2011),
Interpretivism research design has become more important in information system than other
research design types. It is based on the hypothesis that social reality is neither objective nor
single. This design is invested in methodological as well as philosophical ways of assuming
social reality. The research has used the interpretivism design so as to highlight the significance
of the study field with respect to the changes necessary to be implemented in the protocols. In
addition to this, the approach has been made towards the adoption of the required algorithm for
improving the networking and the systems for smooth video calling by researching on a range of
operations as presented in the findings section.
3.2 Research Problem and Philosophical Questions
The problem that has been identified studying the literature and research papers is that slow video
calling is mainly due to slow connectivity and internet issues. At the present era, communication
has been one of the important paradigms that has facilitated innovation and smooth functioning
of all organizations, corporates and other sections. The interruption in video calling has slowed
down the communication system thereby causing communication gap at distant place. Slower

Improving video streaming through increasing protocol efficiencies
connectivity not only distorts the process of video calling but also slower down communication
over voice calls as well.
3.2 Selection of Research Methodology
A research method refers to a systemic approach in which research is conducted. For this
research paper, the qualitative research methodology has been selected. The qualitative research
methodology is preferred to other methodologies as a qualitative research methodology is
deductive as well as does not involve hypothesis to start the research. According to the point of
view of the scholars Knowles and Cole (2008), the data collection method in qualitative research
is field based as well as personal or interactive and circular. Qualitative research refers to market
research that emphasizes on the collecting data from the conversational communication or open-
ended conversation. This method focuses not only “what” a community thinks but also “why”
they do think so.
No numerical evaluation techniques have been undertaken in this research due to time constraints
and the complexity of the data evaluation using the quantitative method. It is an evident fact that
mixed method research has always been proven to one of the most desirable approaches;
however, due to the authenticity of the academic literature as well as the approach of the
researcher towards validating the data and the information, the only qualitative method has been
used.
3.3.1 Real Time Streaming Protocol
The RTSP is an application layer and non-connection-oriented protocol which makes use of the
sessions that are associated with the help of an identifier. The RTSP typically uses the UDP
protocol in order to share the video as well as audio data and also, on the other hand, uses the

Improving video streaming through increasing protocol efficiencies
TCP control for controlling operations. It has a similar syntax like that of an HTTP protocol and
provides the below-mentioned operations.
• Retrieving media from the server: The customer can request description by HTTP or any other
technique of presentation. It includes the multicast addresses and ports used in the ongoing
media.
• Invitation of a media server to a conference: The Media server can be called to either perform
the media or record a part or all of the media on a presentation during a current meeting.
• Media addition to a present presentation: It is of use when the server is to be made able to
inform the client about the additional media availability.
The URL structure of RTSP is the same as that of HTTP with only one small difference in the
syntax RTSP:// in RTSP instead of using HTTP. The RTSP also has new request functions such
as Setup, play, describe, teardown, and pause. The Describe feature is used to provide a short
introduction on the use of the Description Protocol appointed by the URL RTSP.The server
provides a description of the request as a response. The Setup function is mainly used to develop
as to how the stream will be transported, a transport specification, the URL of the multimedia
stream which includes the port to the video and audio data of the RTP. The Play request enables
the data stream processing by the server by making use of the ports that had been set up during
the configure function. The Pause function halts the one or all streams temporarily. The teardown
function stops the transfer of data and releases all the resources.

Improving video streaming through increasing protocol efficiencies
3.3.2 WebRTC
This was developed by the World Wide Web Consortium (W3C) and includes an Application
Programming Interface (API). The program enables video calls and chats for the apps and also
utilizes P2P files without a single plugin. It was introduced first by Google, then published as an
open source. The main features of WebRTC are:
• getUserMedia: It is helpful in obtaining the audio and video streams right from the
camera and microphones hardware and also allows to take screenshot and screen sharing
feature with other users.
• RTCPeerConnection: It helps is making up the video and audio stream and
performs different tasks such as processing of the signals, code executions, administrating
the bandwidth, and streaming security.
• RTCDataChannel: Helps to share video or audio information between linked
users. The interaction between colleagues and exchanges and information type is two-
way. It uses WebSockets to link the two-way interaction between the customer and server
by using TCP as the quickest interaction or UDP.
• geoStats: It is a call to the API that helps in getting different information about a
single session of the WebRTC.
The steps involved in making a call by using the WebRTC are as follows:
• It first gets an offer for stream from the RemotePeer.
• It then instantiates the PeerConnection from the application.
• After the Connection is done, a stream of the media and the track of audio and
video are created by the PeerConnectionFactory and adds this stream on to the
Connection.

Improving video streaming through increasing protocol efficiencies
• The RemotePeer is communicated by the application and kicks off the
communication of media.
The WebRTC protocol makes use of the Datagram Transport Layer Security (DTLS) protocol as
a security protocol measure. It is based on the TLS protocol, which helps in establishing
parameters of the media communications. In order to encrypt all the media communications, The
WebRTC protocol makes use of the Secure Real-Time Transport Protocol (SRTP) which uses the
AES-256 algorithm in order to encrypt the media that gets communicated for securing it from the
breach and any unauthorized use.
Figure 1: Web Real-Time Communication request order.
The systems researched in the findings were executed through applications which were Android-
based in which the logic of the application and application of the web server performs different
functions while supporting the system. Both systems architecture had been developed in the same

Improving video streaming through increasing protocol efficiencies
manner with the only change in the protocols and the area that is compulsory to communicate
with them. The main difference is that the video is sent only to the server in case of using RTSP
which then sends it to the user, whereas every user is required to distribute the video to all the
different users in case of using WebRTC protocol because it uses peer-to-peer communication
method (Begen et al., 2011).
Two new streaming platform had been implemented through two new Android applications. The
Android apps created for that purpose are based on the Google development principle and design
principles and use the Java language by using the Android Software Development Kit
(SDK).Two servers are unique to the systems that execute the two protocols that had been used.
While these were two independent servers that communicate, they could also be placed on the
same physical server. However, they are physically separated for the flexibility of the system.
The servers so used are MEAN and Live555 or KURENTO. The MEAN server is a point server
that stores who’s online and not at all times. It also has information on system users regarding
public IPs, URLs as well as other private data. The KURENTO and Live555 servers are, on the
other hand, media servers, acting if needed as an intermediate in the streaming process.

Improving video streaming through increasing protocol efficiencies
CHAPTER FOUR
4.0 EXPERIMENTS
4.1 Introduction
In this section, the data collected are tested and experimented in order to get the analysis of the
research topic for the purpose of drawing conclusion, the experimental design, data collection and
analysis are discussed here.
4.2 Experimental Design
The figure below shows the network topology which is used in the simulation. The proposed
scheme of HTTP approach is implemented using ns-3 simulation tool. Our experiment are closely
configured in order to match the scenarios whereby 4 clients compete for 100mpbs bandwidth.
Another experiment is also done with 20 clients, 50 clients and 100 clients competing for the
same 100mbps bandwidth. The main metrics under evaluation are client side controller and
server side controller. In the client side controller metrics, efficiency, stability, buffer controllers
and bandwidth estimation are investigated. In the server side controller metrics, fairness, and
server efficiency controllers are investigated.

Improving video streaming through increasing protocol efficiencies
Figure 2: Network topology
4.3 Client Side Controller
The core role of this controller is to maximize the quality of the video which is perceived and
minimizing the process of oscillations and making sure that the video playback smoothness is
achieved upon streaming. It also maintains the buffer up the threshold. The output which is given
by this controller is final.
4.3.1 Bandwidth Estimator
The experiment is carried out with comparison of the algorithm proposed with the scheme below;
ESTC
PANDA
FESTIVE

Improving video streaming through increasing protocol efficiencies
4.3.2 Buffer Controller
A simulation experiment is done in order to determine buffer when different number of clients
are connected to the server. In the simulation, the max and min buffer is defined as βmax and
βmin respectively.
4.3.3 Efficiency Controller
In this experiment, the largest and highest video quality possible is selected taking into
consideration the state of buffer and the bandwidth vulnerability. This is calculated by using the
equation below
Where Q(.) is the function of quantization which returns the highest index level together with the
Expected Fetch Time.
4.3.4 Stability
Another simulation experiment is done using 4 clients, to determine the stability of the client side
controller.
The algorithm which is used in the client side controller is as below;

Improving video streaming through increasing protocol efficiencies
Equation 1: Client Side Controller
4.4 Server Side Controller
From the server side, the current bitrate of downloading, shared bandwidth, smoothed throughput
and the total number of clients connected are all known. This is because the clients send this
information to the server using the HTTP GET request. Fairness controller and efficiency
controller are also tested and experimented using the ns-3 simulation using the below algorithm;

Improving video streaming through increasing protocol efficiencies
Equation 2: Server Side Controller
4.5 Evaluation Metrics
Different metrics are used in the performance of the simulation in this particular section. In terms
of comparison, FESTIVE and PANDA algorithms are used and implemented in the experiment
utilizing the configurations described above. This evaluation is based on the following main
metrics; the fairness as various clients complete for the same amount of bandwidth and
bottleneck link, the stability of each and every client connected to the server, the occupancy of
the buffer, and finally the total utilization of the bandwidth under investigation.
4.6 Data Collection
Data collection is the method of collecting as well as measuring finding on inconsistent of
interest in a very comprehensive manner, which makes one able to understand the research

Improving video streaming through increasing protocol efficiencies
question as well as test hypothesis and calculate outcomes. For this research paper, the secondary
data collection has been chosen. Secondary data collection refers to a source of findings other
than the researcher. This type of data collection is selected for this research because it is easily
approachable, and less time is needed for collecting all the appropriate information. It is also less
costly compared to primary data collection method. In the case of qualitative research method,
secondary data has been collected from the interviews, newspapers, diaries, transcript, etc. The
information with regard to the secondary data has been taken from various literary works and
networking articles that provide relevant facts and figures in line with the research objectives.
4.7 Data Analysis
Data analysis is the process of evaluating data by analytical as well as logical reasoning to inspect
each of the elements of data collected. This is an essential part of the research. For this research
paper, the thematic approach of data analysis has been used (Coleman, 2013). Thematic data
analysis is extensively used for a qualitative research method and emphasizes on inspecting
themes within the collected data. This method of data analysis has been selected for this research
paper because it provides an extremely elastic approach which can be customized as detailed
theoretical as well as technological knowledge is not required and provides a more approachable
form of analysis. Architecture based approach has been used in the research. Qualitative analysis
has been carried out to identify the strength as well as to minimize weaknesses of the network
system. Charts have been used to compare all the researched software with the proposed software
architecture

Improving video streaming through increasing protocol efficiencies
CHAPTER FIVE
5.0 RESULTS
5.1 Introduction
In order to do evaluation on the performance of the ESTC scheme, a number of simulations were
conducted in the experiment section using the ns-2 simulation software. Based on the above
metrics, the FESTIXVE and the PANDA is also compared and evaluated.
5.2 System Configuration
In this particular paper, a star topology which is made up of one server as well as four clients
which are connected to the central router is utilized to conduct several simulations. Server and
clients’ communication is then established from the TCP protocol. As shown from the figure
below, the network topology which was used is depicted, in the topology, the client is requesting
a video sequence of around 10 minutes which is pre-encoded into 12 representations as follows;
11000, 9000, 7000, 5000, 3000, 2200, 1800, 1200, 800, 400, 100, 30, kbps.

Improving video streaming through increasing protocol efficiencies
Figure 3: Star Topology
The bandwidth which is shared from the server is set to be 10 mbps, this bandwidth amount is
also subtracted in order to avoid saturation of the link called the bottleneck that is set to 400 kbps.
Also the length of the chunk is also set at τ = 2s. the minimum chuck which is to be requested
from the representation is also set at δ = 5 chunks. . βmin is set at 7 and βmax is set at 15 chunks.
5.3 Presentation Of Results
For the ESTC performance evaluation against the FESTIVE and the PANDA, the following
scenario is put into consideration whereby, the four clients having different capacities of
bandwidth, both of them connect to the server streaming at different sets of time. The table below
shows the different capacity of links as well as the time the links of the simulated clients arrives.
Table 1: THE LINK CAPACITIES AND ARRIVAL TIMES
Link Capacity in mbps Arrival Time in seconds
Client 1 10 0
Client 2 10 300
Client 3 0.9 100
Client 4 0.9 150
For the purpose of creating a discussion and illustration, the four clients results are provided in
order to show the traits of various solutions and the ways each of the links converges to the fair
share that is not at all possible especially when running a lot of clients. Also, other simulations

Improving video streaming through increasing protocol efficiencies
were done using more clients example, 20 to 50 clients and there results which were found are
also discussed in sub-section 6 of this evaluation section. From the observation made from the
performance of the ETSC as compared to the FESTIVE and the PANDA, it remains the same
regardless of the number of clients carrying out the simulation.
Efficiency: The representation levels played by different clients are displayed and plotted as
from the below figures 4a, 4b and 4c for PANDA, FECTIVE as well as ESTC schemes. From the
figure representing the ESTC, the client 1 is making self-connection to the server when t=0 hence
making benefits from the bandwidth in order to reach l10 when time is 80s which is considered as
the highest level that is lower as compared to its capacity of bandwidth. When time is 100s, the
figure shows client 3 joining the server and the client is trying to improve the representation
acquired level till the client achieves the highest l3 level. The fact that client 3 joins the server,
does not affect client 1 since the bandwidth shared is enough to satisfying the requests which are
being done by both clients and the bandwidth is therefore able to keep then stable. When time is
at 150s, client 4 joins the server as the representation level of client 1 decreases. This does not
affect client 3 at all since its level played is lower as compared to the average level of bandwidth.
When the time cis at 300s, client 2 then joins the server and begins its session, the representation
level increases systematically till reaching l7 that is considered the highest level which can be
attempted. From the graph, it can be noted that client 3 and client 2 maintains their levels are
remains very stable since they have their levels which is far much lower as compared to the
average level of the bandwidth. The representation level of the client 1 is seen to decrease by one
level which in turn allows the client 2 to increase in its level. This particular strategy allows at the
very similar time maintaining property smoothness as well as avoiding the level switches which
are not necessary that could end up making the client more stable.

Improving video streaming through increasing protocol efficiencies
Figure 4: The played representation levels of the different clients.
Figure 5: The played representation levels
In the case of FESTIVE as illustrated in the graph, different clients are getting different
considerable delays in getting the fair allocation level l10 at t = 155s, level l3 at t = 129s, level l2 at
t = 168s and level l8 at t = 435s which are for clients 1,3,4, and 2 in that respective order. Also in
the FESTIVE case as shown in the graph, the client is not able to maintain level of fair allocation
for longer period of time as compared to client 1 who is in this case the first client who
automatically gets the higher fair allocation of the level.

Improving video streaming through increasing protocol efficiencies
The main objective of the PANDA algorithm is in order to maintain and improved the stability of
the client. This is as illustrated in the graph for PANDA analysis where all the clients are stable.
Though from the observation, it is noticed that each and every client in the PANDA case is
getting its level lower that the fair share even when the TCP throughput is allowing a bit higher
bit rate.
Figure 6: Feedback for the played representation
This main cause of this is the asymmetrical rate shifting of levels and features that permits
conservative rate of upshift in level and also a downshift which is more responsive because of the
formula which has been employed at the bandwidth share step. This particular formula is aimed
at mitigating the observed throughput impact when it is improved hence making the client more
stable though less efficient as far as utilization of bandwidth is concerned. The different values

Improving video streaming through increasing protocol efficiencies
which were recorded from the efficiency metric for the algorithms were also plotted as shown in
the figure 5a, from the graph, when client 1 makes a connection to the server, it is seen that the
maximum value of efficiency which is attained is 0.9 when time was 81s, this is so because of the
discrete nature of the levels of representation due to the fact that the nest level which is 11mbps is
much far greater as compared to the maximum capacity of the bandwidth for client 1 and also the
move for that level will create instability issues.
Figure 7: The efficiency and fairness evaluation of ESTC, FESTIVE and PANDA
This particular efficiency was further improved to the vicinity value of 1 after all the 3 clients
have made connection to the server when time is 100s. The beginning of client 4 to make
connection to the server when time is 150s, its causes the efficiency to drop to 0.82. this is due to
the fact that when the quality level of the fourth client increases, then the played level of the
client 1 decreases though the amount decreased is much greater as compared to the amount of
bandwidth increased. Similarly, as the client 2 connects to the server, with the same capacity of
bandwidth as client 1, it results to the drop of the value of efficiency to 0.66 measure for the
obvious critical reason that the level of representation of client 1 falls to 5mpbs at level 8 with the
difference of around 2mpbs which is between level 8 and level 9. The value of the efficiency
rises gradually as the level of representation of client 2 is increasing till it reaches the efficiency

Improving video streaming through increasing protocol efficiencies
value of 0.92 when time is 35 seconds. The efficiency value remains stable from this particular
time until client 1 gets out of the connection which also causes a considerable drop in the
efficiency value to 0.42, this is due to the fact that client 1 occupied the biggest shared bandwidth
amount. For this reason, client 2 is given the opportunity to improve in level pf played quality till
it reaches level 9 with efficiency value of 0.82 before the client 4 and client 3 leaves the
connection.
As compared to the FESTIVE, ESTC shows at least better efficiency though both FESTIVE and
ESTC are all far much efficient as compared to PANDA. As shown from fig 5a, it is seen that
FESTIVE value of efficiency metric is above the value of 1 meaning that the bitrate accumulated
from all the clients is far much more than the bandwidth shared. This in turn creates
maximization on the utilization of the bandwidth. This is however dangerous because it might
lead to instability when the video quality is rapidly decreasing or buffering when keeping levels
higher for long periods of time.
In Figs. 6(a), 7(a), 8(a), 9(a), shows the throughput of 4 different clients in the case of ESTC,
while the throughput for the clients in PANDA and FESTIVE are rarely illustrated.
Figure 8: The comparison of the allocated bitrates to Client 1 between ESTC, FESTIVE and
PANDA

Improving video streaming through increasing protocol efficiencies
For PANDA, Figs. 6(c), 7(c), 8(c), 9(c) shows how PANDA takes the highest level of
representation that is lower as compared to the smoothed bandwidth.
Figure 9: The comparison of the allocated bitrates to Client 2 between ESTC, FESTIVE and
PANDA.
Figure 10: The comparison of the allocated bitrates to Client 3 between ESTC, FESTIVE and
PANDA

Improving video streaming through increasing protocol efficiencies
Figure 11: The comparison of the allocated bitrates to client 4 between ESTC, FESTIVE and
PANDA
Fairness: As seen in figure 5b above, the variation in the fairness value metrics is utilized in
calculation of the throughput of the session of the clients. From the figure, it becomes clear that
approach proposed is better that PANDA and FESTIVE algorithms in terms of fairness on the
competing clients. The lowest fairness values occur when the client leaves or connects to the
server. The lowest values from the graphs is noticed to be when time is at 100 seconds, 150
seconds, 300 seconds, with correspondence to the client 3.4. and arrival times respectively. The
proposed system is observed to attain the highest fairness at most times.
Figure 12: The buffer occupancy evaluation of the different clients.
Stability: As seen in the fig 4a, all the clients connected to the server are 100% stable with
exemption of client 1 only which is unstable. The results from the graph in an overall stability of
the system of 0.99875. Its normal for the system to be unstable because it takes place when a new
client makes a connection to the server, this does not affect the experience of the user except at
312 seconds when client 1 switches to level 9 and then drops to level 8 again after 3 seconds. In
fig 4b, the clients of FESTIVE are unstable as compared to the clients of ESTC.

Improving video streaming through increasing protocol efficiencies
The overall stability of FESTIVE case is 0.944167. this is caused by the fact that client 1 that
plays a bit rate higher as compared to the fair allocation. The fig 4c shows the PANDA clients are
completely stable. A comparison between the stability of the clients in PANDA, FECTIVE and
ESTC cases are shown in Fig 11.
Figure 13: The evaluation of the clients’ stability in case of ESTC, FESTIVE
Buffer Occupancy: This is clearly shown in fig 10 a, b and c that the approaches make sure that
there are packets always in the buffer which are displayed therefore the buffer never goes without
the packets. The difference in terms of buffer management is noticed between the three cases
whereby the FESTIVE clients maintains the occupancy of the buffer in a buffer reference which
is defined. PANDA on the other side, tries in keeping the buffer in the minimum buffer while the
ESTC makes sure that the buffer is kept as full as possible.
Convergence Time: this is illustrated and demonstrated in fig 4a and 4b where new clients make
a quick convergence to the fair share in the case of the solution proposed as compared to the
FESTIVE case. For instance, client 1 reaches the level 10 at 80 second which for the FESTIVE
case, this is achieved at 156 second. The convergence in this case is smooth regardless of the
bitrate increase or decrease. The level of quality of the previous clients which had already

Improving video streaming through increasing protocol efficiencies
connected is only affected when necessary. As opposed to the gradual level rate shifting strategy
which is employed by the FESTIVE and the ESTC as seen in Fig 4c where the PANDA makes
use of the aggressive level switching representation hence reaching the destination very quickly
though this might affect the user.
Figure 14: Flows of packets
Fig. 12 makes a summary of the overall values of various metrics for all the 3 solutions. It is
worth noting that ESTC is helping in the improvement of the efficiency by 23% and stability by
43% and the fairness is improved by 21% and this is compared to the FESTIVE. The solution
proposed also makes a quick convergence to the fair allocation as compared to the FESTIVE as
witnessed from the Fig. 4a AND 4b.

Improving video streaming through increasing protocol efficiencies
Figure 15: Comparison of the overall stability, fairness and efficiency metrics
Results of performance Using Large Number of Clients: For further validation of the approach
proposed, simulations were conducted using many clients with different capacities of bandwidth,
and the joining the server time is taken randomly. For the purpose of exactness, simulations were
conducted using 20, 50 and 100 clients and results were also recorded. As seen in figure 13a, 13b
and 13c for the purpose of efficiency and fig 14a, 14b and 14c for the metrics of fairness.
Figure 16: The efficiency comparison between ESTC, FESTIVE and PANDA when running
larger number of clients
Using 20 clients, where they started randomly within 0-200 seconds, with competence of
100mbps of the bandwidth shared, at the same time, 50 clients starting within 0-300 seconds, and

Improving video streaming through increasing protocol efficiencies
100clinets also starting within 0-300 seconds both sharing 300mbps. For the 20 clients, as shown
in fig13a, ESTC works best than both FESTIVE and PANDA all the time.
Figure 17: The fairness comparison between ESTC, FESTIVE and PANDA when running larger
number of clients.
For the 50 clients and 100 clients, as shown in fig 13b, 13c, 14b and 14c, ESTC still works best
than both FESTIVE and PANDA always.
Figure 18: Comparison of the estimated
bandwidth for the CBR video
Figure 19: Comparison of the estimated
bandwidth for the VBR video

Improving video streaming through increasing protocol efficiencies
To evaluate the bandwidth, regular plan utilizes quick throughput, PANDA utilizes the ω, κ
parameter, and FESTIVE uses the symphonious mean. In this assessment, we set the ω = 0.3, κ =
0.14. Fig. 15 and Fig. 16 demonstrate a correlation of the evaluated bandwidths.
Figure 20: Bandwidth comparison
The proposed plan beats different plans, and the enormous expectation mistakes in the tail show
up in light of the fact that the bandwidth watched for each fragment relies upon the quantity of
contending customers. The exactness of the throughput estimation strategy empowers a
progressively steady cradle inhabitance, in this manner lessening the underlying buffering delay.

Improving video streaming through increasing protocol efficiencies
5.4 Discussions
5.4.1 RTSP Streaming Platform
The RTSP streaming platform architecture is made up of the use of RTSP protocol P2P
communication in locations where both the users are within the similar network and make use of
a Media server Live555 if no network or internet was dedicated. The Live555 media server
supports the most prevalent type of media files, a fully open source RTSP server. In addition, this
server can simultaneously stream multiple streams from the same or different files with UDP
protocol, but can also transmit stream through the TCP protocol if necessary (Mcgrew et all,
2015).
Depending on the application scenario, the established Android application functions as a server
and/or customer. This application enables to connect to several users and send their streams in
real time. In the application of this scheme, Android 6.0 was chosen, which is the recent version
available for the Android SDK API 2.3. The RTSP protocol had neither been used in this version
of Android or any other. It was, therefore, necessary to use external libraries that were running
the mentioned protocol. It was necessary to utilize Libstreaming open source library, both on the
client side as well as on the server side, to implement the RTSP protocol. This library enables the
microphone and/or the Android device camera to be streamed.
The general flow of the system was now described. First, a request is sent to the internet server
when a user decides to enable streaming when it is beginning with a video call. The server then
answers back with a positive reply, and the smartphone begins the transmittal of the video stream.
If a customer wishes to see a stream and start the application, a request to find out what the
internet stream and its URLs are, will be automatically sent to the server. Finally, the application

Improving video streaming through increasing protocol efficiencies
connects the server media, and then it downloads and reproduces streaming when the user
chooses a stream of the list of internets streaming.
5.4.2 Direct WebRTC Streaming Platform
The direct WebRTC Platform’s architecture consists in using the WebRTC protocol for P2P
communications in places in which the two users are in the same network, whereas the internet
connection is used for those who have no dedicated network. Depending upon the application
situation, the established application functions as customer and/or server. This enables one or
more users to connect and send their current real-time stream (Mcgrew et al., 2015). It was
decided to use Android 6.0s recent accessible version that corresponds to android SDK API 2.3.
Neither it is implemented in android Core in the version of Android and nor in any past version;
it was this necessary to use the libraries that would implement the WebRTC protocol. The
libjingle library had been chosen at that time, which is an open source library published in c++ by
Google that allows P2P links to be established and P2P application development. In order to use
the library that is present in Android, it was important to assemble the source code in order to
form the JNI API so that implementation of JAVA can make use o the implementation of C++
for the alike API. The library is able to stream the microphone or the camera of an Android
device.
The workflow of the WebRTC could now be described. At first, the user sent a request to the
server when it is willing to broadcast along with the port as well as the Public IP that will be used
by the device in order to use the broadcast for streaming. This data sent by the user is further
derived by the Google STUN server. The server responds to the request by sending an [positive
feedback, and then the smartphone begins transmitting the video stream to the web server. As
when the client application is opened, an automatic request to the server is sent with the motive

Improving video streaming through increasing protocol efficiencies
of finding out the online streamings, public IP and ports that are used. The transmission finally
begins soon after the stream is selected by the user from the list of online streaming (Umesh,
2015).
5.4.3 Analysis of Smartphone to Web streaming Applications
For this analysis, video streaming platforms like Hangouts and Facebook has been used.
Measurements had been obtained of the packet delay of the stream for Facebook as well as
Hangouts which are the video streaming platforms as well as the two dedicated systems. The
analysis consisted of doing a call in the form of a video call between the web application as well
as the smartphone. Communications packets were analyzed during this call by making use of the
Wireshark analyzer to determine the stream packet delay. After this, the packet delay stream of
the two implemented system WebRTCand RTSPhad been realized. The outcomes so obtained
reflects how there had been a much lower delay in Direct WebRTC streaming platform than that
of RTSP streaming platform. On average, the delay time of RTSP after receiving 10,000 packets
is 37,807 milliseconds whereas it is only 8072 milliseconds in the case of WebRTC streaming
platform.
While analyzing the Google Hangouts application, it was clear in the packets that were captured
that the STUN protocol was used. This clearly showed that video streaming had gone directly
between the clients and did not go through any media server. Therefore, it could be likely
concluded that WebRTC is used by Google Hangouts. The delay time in the streaming was also
calculated through the data packets that had been so collected. The average measurement that was
calculated in the delay of the stream time had resulted in 8.832 milliseconds of delay.

Improving video streaming through increasing protocol efficiencies
The other analyzed system was a Facebook application. In this case also, as seen in the case of
Hangouts, it could be seen that video streaming directly travels among the clients, without
travelling through any kind of media server. The average delay time that was obtained through
the measurements of the media packets should take a delay in the stream time by 11.093
milliseconds.
Figure 21: Average delay time
5.5.4 Analysis of smartphones to smartphone streaming applications
In this section of the experiment, the analyzed communication is between two smartphones. Also,
some of the application used in the previous experiment is also analyzed. Various measures on
packet delay of the stream had been taken for Facebook, Skype, Hangouts, etc. and also for the
WebRTC video streaming platform. The packet traces of the systems run were analyzed and
collected just like it had been done in the previous analysis. In this process, a video call was
generated between both smartphones by making use of different android applications.
Communication packets were collected and analyzed in order to know the delay in the stream.

Improving video streaming through increasing protocol efficiencies
After this, Direct WebRTC stream packet delay was measured. It was evident that there has been
a less delay in this test as compared to the previous test of the smartphone to web application test.
An average delay of 5.112 milliseconds had been recorded in this platform.
The analysis done on the google hangouts application showed the use of STUN protocol, as
shown in the previous test. The data packets collected were used to calculate the stream delay.
The results were almost similar to that of Direct WebRTC protocol and reflected a stream delay
of 6.87 milliseconds.
On analyzing the Facebook messenger application, it could possibly be assumed that it uses the
WebRTC protocol. The stream delay was calculated and showed a delay of 16.234 milliseconds.
Similarly, on analyzing the Google Duo video call application, it was evident that it makes use of
STUN server and also the stream directly goes between the two smartphones and does not pass
through any web server. The stream delay showed a result of 5.424 milliseconds of delay in the
stream.
Figure 22: Average stream delay time between smartphone to smartphones

Improving video streaming through increasing protocol efficiencies
CHAPTER SIX
6.0 CONCLUSION
6.1 Conclusion
For a successful HTTP adaptive solution of streaming video, it must be able to ensure that the
objectives are met as follows: very high utilization of the resources of the network, farness,
stability, as well as convergence over a short period of time to the fair share at the same time
avoiding the issues of buffering. For this reason, this particular paper proposes HTTP solution
which achieves the objectives set by very close relation between the clients and the servers. With
dependence on the data which is present from both the client and the server, the control of the
efficiency, stability and fairness is delegated to the client and server side. From the client side, the
bandwidth available, the history of the bitrate played and the occupancy of buffer are well known
hence making it best for the control of the utilization of the bandwidth, management of buffer as
well as stability. The fairness in contrary, is controlled at the server because the connected client
numbers together with their bitrates played as well as the capacity of the bottleneck link are
known from the server side. From the results of the simulation, the proposed approach makes an
improvement of the bandwidth fairness, utilization and stability while making sure that the buffer
is kept as full as possible. Also it showed from the results obtained that convergence rate of the
clients are quicker and smoother for the fair share in the cases of decreasing and increasing
bitrate. It is true that video calls have helped in making communication with distant people easier
at an affordable rate. However, the fluctuations in the internet network and connectivity obstruct
any video creating misunderstanding among people, losses to business or conflicts amid people.
Emerging nations have less internet speed with huge population connecting their internet with

Improving video streaming through increasing protocol efficiencies
2G. In order to solve this problem, the above protocols and architecture have been developed that
would help in ensuring smooth video calls without disruption and high-speed internet
connectivity. This paper involves a full assessment of the most common video streaming
protocols, with particular emphasis on WebRTC as well as RTSP protocols. In addition, two new
streaming platforms for comparing and optimizing their operation have been developed. These
implementations have been developed with the most prevalent systems and requirements for use
with Android applications in mind. It was concluded from the experiments that significant
improvements were achieved for WebRTC over RTSP, for both the time the package has been
sent. In addition, two studies have contrasted the implemented schemes with the most popular
business apps. In comparison with the most common smartphone and web video apps, new
implemented platforms were used with external software as code for such proprietary
applications cannot be accessed. This test showed comparable and better conduct than the other
compared schemes both on Hangouts application as well as the new Direct WebRTC platform.
On the other hand, it was used with the above-mentioned internal software to compare the
implementation of the Direct WebRTC to the most popular smartphone to Android smartphone
call apps because the codes of such proprietary applications are not available either. The
implementation of the Direct WebRTC scheme showed a good answer again and showed the
highest outcomes in comparison with the Google Duo application. It can, therefore, be confirmed
at this stage that the usage of the WebRTC protocol gives greater QoE and QoS than other
protocols and that the Direct WebRTC System implemented offers excellent outcomes based on
the studies that have been conducted. The research may lead in the future to fresh works that
would enhance and further enhance the suggested video streaming platform, thanks to the know-
how gained in the research and execution of the two protocols mentioned. In addition, the

Improving video streaming through increasing protocol efficiencies
research of fresh metrics for comparing the presented streaming platforms with the commercial
ones will allow for other elements of the submitted research.
6.2 Recommendation
From the research work and the conclusion drawn, it is therefore recommended that HTTP
DASH based approach be used in the improvement of the video streaming technology. The
primary reason to develop the above architecture is to demonstrate to the people living in
emerging nations using old generation device or having low internet speed that hampers their
video calls every time. It is an evident fact that people have their friends living in a diverse part
of the nation; moreover, they may have family members staying out of the state or country. As
the video call helps in making the conversation more interactive; therefore, it is imperative to
ensure that the speed of internet connectivity does not disrupt the conversation. One of the major
objectives of this project is providing all the people with smooth video call facility. People
perhaps use diverse internet networks such as 2G, 3G, or 4G. They might face difficulties in
placing a video call to their close ones. The proposed recommendations will facilitate their
requirements and demands.
The above findings obviously demonstrate a greater efficiency than the others with the use of the
WebRTC protocol. The following table indicates the average time of all studies in milliseconds
by using various protocols and apps studied during the job. The experiment could not be
performed in some of the applications because of which no average appeared in those cases.
In order to improve the implementation of the WebRTC, the usage of UDP protocol in all
WebRTC communications with respect to RTSP could be linked, while TCP for control is used
for the RTSP protocol. The protocol of RTSP does not, however, remove video packets and if
necessary, the WebRTC protocol can. Finally, it sends a video to the other pair in case of peer to

Improving video streaming through increasing protocol efficiencies
peer communications, while the video is sent to the server in the event of RTSP and the server to
the other pair. In addition, the findings achieved in this research on improving RTSP execution of
WebRTC could be generalized as integrated systems had applied the norms faithfully and use
enormously-tested libraries of sources.
In addition, the streaming platform shows improved results over the stream reception time and
stream setting times in all cases, so the dedicated streaming platform provides a higher QoS other
than the applications studied in taking those measurements into account.
6.3 Future Work
In the future research work, implementation of the proposed solution of improving the video
streaming using the HTTP approach in real life scene will be done, with consideration being on
the cases of more than one servers as part of the commercial content delivery network CDN.

Improving video streaming through increasing protocol efficiencies
REFERENCES
Abelson, H., Anderson, R., Bellovin, S. M., Benaloh, J., Blaze, M., Diffie, W., ... &
Rivest, R. L. (2015). Keys under doormats: mandating insecurity by requiring government
access to all data and communications. Journal of Cybersecurity, 1(1), 69-79.
Austerberry, D. (2013). The technology of video and audio streaming. Abingdon:
Routledge.
Begen, A.; Akgul, T.; Baugher, M. Watching video over the web. (2011): Part 1:
Streaming protocols. IEEE Internet Comput15, 54–63.
Bridge, P.D.; Jackson, M.; Robinson, L. (2009) The effectiveness of streaming video on
medical student learning: A case study. Med. Educ. Online. 14, 1–5.
Carla, L., Fantacci, R., Gei, F., Marabissi, D., & Micciullo, L. (2016). LTE enhancements
for public safety and security communications to support group multimedia
communications. IEEE Network, 30(1), 80-85.
Carullo, G., Tambasco, M., Di Mauro, M., & Longo, M. (2016). A performance
evaluation of WebRTC over LTE. In 2016 12th Annual Conference on Wireless On-
demand Network Systems and Services (WONS) (pp. 1-6). IEEE.
Chen, C. M., & Lee, T. H. (2011). Emotion recognition and communication for reducing
second‐language speaking anxiety in a web‐based one‐to‐one synchronous learning
environment. British Journal of Educational Technology, 42(3), 417-440.
Chilisa, B. (2011). Indigenous research methodologies. Sage Publications.

Improving video streaming through increasing protocol efficiencies
Coleman, R. (Ed.). (2013). Deleuze and research methodologies. Edinburgh University
Press.
Diallo, O., Rodrigues, J. J., & Sene, M. (2012). Real-time data management on wireless
sensor networks: A survey. Journal of Network and Computer Applications, 35(3), 1013-
1021.’
Fernández, Saldana, Navajas, Sequeira, & Casadesus, C., (2014). Video Conferences
through the Internet: How to Survive in a Hostile Environment. Video Conferences
through the Internet: How to Survive in a Hostile Environment, Retrieved from
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3926290/
Gardner, M. R., Jenkins, S. M., O'Neil, D. A., Wood, D. L., Spurrier, B. R., & Pruthi, S.
(2015). Perceptions of video-based appointments from the patient's home: a patient
survey. Telemedicine and e-Health, 21(4), 281-285.
Goldman, R., Pea, R., Barron, B., & Derry, S. J. (2014). Video representations and the
perspectivity framework: Epistemology, ethnography, evaluation, and ethics. In Video
research in the learning sciences (pp. 17-52). Routledge.
Knowles, J. G., & Cole, A. L. (2008). Handbook of the arts in qualitative research:
Perspectives, methodologies, examples, and issues. Sage.
Kokkonis, G., Psannis, K. E., Roumeliotis, M., & Schonfeld, D. (2017). Real-time
wireless multisensory smart surveillance with 3D-HEVC streams for internet-of-things
(IoT). The Journal of Supercomputing, 73(3), 1044-1062.
Marketing, M. (2017). Two Secrets to Reliable Video Conferencing. Two Secrets to
Reliable Video Conferencing, Retrieved from https://www.megapath.com/blog/blog-

Improving video streaming through increasing protocol efficiencies
archive/two-secrets-to-reliable-video-conferencing/
McGrew, D.; Rescorla, E. Datagram Transport Layer Security (DTLS) (2010). Extension
to Establish Keys for Secure Real-time Transport Protocol (SRTP). RFC 5764 Available
online: https://datatracker.ietf.org/doc/rfc5764
Navaz, A. S., Jayalakshmi, P., & Asha, N. (2015). Optimization of Real-Time Video Over
3G Wireless Networks. September–2015, International Journal of Applied Engineering
Research, Vol No-10, Issue No-18, 39724-39730.
Neustaedter, C., Pang, C., Forghani, A., Oduor, E., Hillman, S., Judge, T. K., ... &
Greenberg, S. (2015). Sharing domestic life through long-term video connections. ACM
Transactions on Computer-Human Interaction (TOCHI), 22(1), 3.
Pheasant, R. J., Fisher, M. N., Watts, G. R., Whitaker, D. J., & Horoshenkov, K. V.
(2010). The importance of auditory-visual interaction in the construction of ‘tranquil
space’. Journal of Environmental Psychology, 30(4), 501-509.
Richardson, I. E. (2011). The H. 264 advanced video compression standard. New Jersey:
John Wiley & Sons.
Rosas, A. S., & Martínez, J. L. A. (2016). Videoconference system based on WebRTC
with access to the PSTN. Electronic Notes in Theoretical Computer Science, 329, 105-
121.
Rowles, D. (2017). Mobile marketing: how mobile technology is revolutionizing
marketing, communications and advertising. London: Kogan Page Publishers.
Santos-González, I., Rivero-García, A., Molina-Gil, J. and Caballero-Gil, P., 2(017).
Implementation and analysis of real-time streaming protocols. Sensors, 17(4), p.846.

Improving video streaming through increasing protocol efficiencies
Santos-González, I., Rivero-García, A., Molina-Gil, J., & Caballero-Gil, P. (2017).
Implementation and analysis of real-time streaming protocols. Sensors, 17(4), 846.
Steel, K., Cox, D., & Garry, H. (2011). Therapeutic videoconferencing interventions for
the treatment of long-term conditions. Journal of Telemedicine and Telecare, 17(3), 109-
117.
Umesh, A. Performance Analysis of Transmission Protocols for H. 265 Encoder. Ph.D.
Thesis, BlekingeInstitute of Technology, Karlskrona, Sweden, 2015.
Yasumoto, K., Yamaguchi, H., & Shigeno, H. (2016). Survey of real-time processing
technologies of iot data streams. Journal of Information Processing, 24(2), 195-202.
Zeng, D., Guo, S., & Cheng, Z. (2011). The web of things: A survey. JCM, 6(6), 424-438.