4th Scientific International Conference – Najaf – IRAQ (4th SICN-2019)

Real Time Video Streaming From Multi-Source

Using Client-Server for Video Distribution
Ibrahim M. I. Zebari Subhi R. M. Zeebaree Hajar Maseeh Yasin
IT Dept. Duhok Polytechnic University IT Dept.
Technical College of Informatics - Akre Duhok, Iraq Technical College of Informatics - Akre
Duhok Polytechnic University [email protected] Duhok Polytechnic University
Duhok, Iraq Duhok, Iraq
ibrahim. [email protected] [email protected]

Abstract— The Real Time Video Distribution (RTVD) is now a occurrence of instance handover may be through video
day a subject of interest in various applications that support sessions which in turn have to be faced of various
different fields, like video conferences, security systems, video communication media restrictions. They addressed to use of
broadcasting, etc. This paper presents a system for a multi-IP their system with the supporting of handover using video
cameras based on the use of client/server to distribute video sessions and Long-Term Evolution telecommunication
signals to many clients. There are three main functions in the
transmissions. While, a cluster mesh-based architecture for
proposed system: the first is that the server will receive the sent
streams by all connected cameras. The second means that the VoD over hybrid content delivery networks and Peer to Peer
server starts viewing video signals and distributes them to all networks that function as videocassette recorder been
connected clients in the network. Finally, the clients will receive addressed by [5]. They illustrated that the users have the
video signals from the server. Hence, the proposed RTVD System capability of review the recorded streams as demand.
(RTVDS) makes the Admin to focus on the received frames sent Video streaming applications can be considered as different
by IP cameras to/from the server within signals broadcasting. In broadcasting approaches. These applications always
addition, RTVDS has the capability of tracking and storing the categorized belong the Online Communication Style (OCS)
frames stream and encode them during the broadcasting process. field through the based media. Hence, the processed signals
Different structures and algorithms have been depended in the
and data follow will be seen and used by hosts and mobile
proposed RTVDS using tools implemented by VC# programming
language. phones [6,7]. While the other categorized belong the demand
strategy to reviewing the sent frames called video on demand
Keywords—IP camera, client/server, stream, broadcasting, real (VoD). So, VoD approach enables the user to watch the video
time frames at any time rather than at live time. In other words more
than one users can benefit from this facility with more than one
I. INTRODUCTION stream type at the same time [8,9]. Video streaming is used in
many applications such as security, surveillance,
Nowadays, the Internet Information and Communication entertainments, medicine, and educations. The aim of this
Technology made deep effects in the human life, the paper is to design and implement a real time video streaming
interchange information and getting more details about from multi-source using client-server for video distribution.
different fields, it is prefer to distribute videos among the users. This paper constructed to seven sections. Section I
Because of the big size and reserving hug spaces, the introduction, section II video distribution using IP, section III
operations of intense frames streams transmitting or storing general organization of the proposed system, section IV
will be much cost [1]. Frames stream can be described mechanism of the proposed system operation, section V
depending on number of criteria which provide the roadmap of obtained results, Section VI discussion and comparison with
the manipulation structure with these criteria [2]. Hence, previews works, VII Conclusion and finally references used in
stream frames might be described as frames stream style, color this paper.
of a minute area of illumination on a display screen deepness,
coding recipe, reaching of frames grade. Because of the huge
size of the obtained stream, MPEG-x compression technique II. VIDEO DISTRIBUTION USING IP
can be used to minimize the stream size [3]. The view point of Distributing video across a network requires
keep tracking the connections of multi accessing using various communication between the sending and receiving device.
tools and technics was addressed by [4]. However, the With the growing capacity in networks to handle large

978-1-7281-4425-2/19/$31.00 ©2019 IEEE 115

Authorized licensed use limited to: UNIVERSITY OF BIRMINGHAM. Downloaded on May 11,2020 at 03:04:16 UTC from IEEE Xplore. Restrictions apply.
 4th Scientific International Conference – Najaf – IRAQ (4th SICN-2019)

amounts of bandwidth, IP-based video transmissions have many responsibilities; its software usually requires special
become more and more popular [10]. IP is a low-level system privileges to receive video signals from all IP-cameras.
networking protocol used by computers and other hardware to Then, displaying them on their viewers (a viewer is a window
communicate across networks. TCP/IP is the most complete screen that used by the server to display video signal, in this
networking protocol available. Because of this it has also system the server uses three viewers). Adding to that, the
become almost universally accepted and can be utilized in server has many special control-options: one of them manages
virtually any networking environment [11]. TCP/IP packets video signals for all viewers in the system such as; turning
are actually “encapsulated” or surrounded by the network on/off video signals, setting mute/unmute sound, recording and
protocol in use, such as Ethernet, Frame Relay, ATM, etc.
Because TCP/IP is so universal, it can be used to provide
functionality and communications across disparate networks
[12].There are two categories of video streaming can be
depended, based on the broadcasting technology used
(centralized and non-centralized), centralized: the server is
responsible for accessing, storing and transferring data. While
the client, is responsible for decoding and playing the received
video-signal [13]. Non-centralized: the clients actively
participate in sending videos to other clients. In this category
solutions based on peer-to-peer networks are surveyed which
enable peers to collaborate without the need of a centralized
component [14]. There are three types of client/server
networks as shown in Fig. 1 [15], [16]:

Fig. 2. General Organization of the System

stopping video signals. Another one is concerned with moving

IP-cameras to all directions. Also, the server has the ability to
calculate received data (i.e. dataflow) in KiloBytes (KB).
Another part of the proposed system is the IP-cameras
part. The IP-cameras used in this system have two image
resolution-modes; one is the Video Graphics Array VGA with
(640*480 pixel) and the other is Quarter VGA (QVGA) with
(320*240 pixel). The default resolution is (640*480 pixels).
This system uses tow hub-switches to connect IP-cameras with
the server and server with clients through UTP cable. Fig. 2
represents the general organization of the proposed system.

During server preparation, it will need a location for

displaying each video signal. After the video signal has been
assigned to the specified location at the sever side, it has the
Fig. 1. Possible Client/Server topological design and strategies
ability to perform some control options on the video signals
and broadcasting them to the clients. Fig. 3 represents a block
III. GENERAL ORGANIZATION AND MECHANISM OF THE
diagram of the proposed system.
PROPOSED SYSTEM
The system is organized with four hardware-sides (IP-
cameras, server, client and hub-switch). The first one
represents IP-cameras which consists of X-IP-Cameras (for the
test X is considered as equal to 3). The second hardware is
server which consists of one host. The third hardware is client
which consists of N- host (for the test N is considered as equal
to 6), while the last hardware is two hub-switches device one
of them has 4 port and others has 8 ports. Each IP-camera, host Fig. 3. Block Diagram of the Proposed System
(server) and host (clients) has a unique IP address and all
addresses must belong to the same network. The system has

116

Authorized licensed use limited to: UNIVERSITY OF BIRMINGHAM. Downloaded on May 11,2020 at 03:04:16 UTC from IEEE Xplore. Restrictions apply.
 4th Scientific International Conference – Najaf – IRAQ (4th SICN-2019)

Adding to that, the server will be able to record the video

signal and save it as a file in a local machine. For such
conditions, it is better to use compression techniques to reduce
the size of the recorded video signals in the hard disk or
broadcasting video signals to exploit the network bandwidth.
The server has the ability to know the size of the received/sent
data.
For video broadcasting from the server to clients it is
preferred to use the socket programming approach. A socket is
Fig. 5. GUI of the Client
a tool that is used to transfer data from one device to another
within the network. After this tool been (defined) and can be
known as Defined Tool (DT), however the initialization of the IV. INPLEMENTED RESULTS, DISCUSSION AND COMPARISON
communication for the received signals will be performed. The WITH PREVIEWS WORKS
DT either will be an active or passive to represent initialization
of the (client and server) respectively. The DT (i.e. socket) will A. Server Receiving Video Signals from IP-Cameras
not be provided by details of internet protocol and ports related After server initialization, the first step that must be done is
to the other side host until using the application, which will get searching for IP-cameras and attempting to detect them. The
one of them or the both. server will start receiving video signals from all detected IP-
The proposed system has many sockets at the server for Cameras and display them on its viewers. Here, there are three
receiving/transferring data/messages from/to IP- cases of receiving video signals for all connected IP cameras.
cameras/clients. The server is listening to a single port (one First case: server received video signals at a daytime when
socket) for all clients. When many clients are connected to the installing an IP camera at outdoor. Second case: server
server; all clients will share the same socket (i.e. data received video signals at a daytime when installing an IP
transferred between the server and all clients through the same camera at indoor. Third case: server received video signals at a
port). To perform efficient connection between the server and night time when installing an IP camera at outdoor. Tables I to
all the clients, the server uses a multithreading approach. III represent dataflow of the server when it has received data
Hence, adding to the main thread that manages the connections from all three IP-cameras in three different cases. Data of these
between the server on one side and all the clients on another three tables are plotted to get related graphs as shown in Figs.
side. Many secondary threads will be generated to perform the 6, 7 and fig.8 respectively.
rest operations.
Multithreading provides a flexible system and not a TABLE I. DATAFLOW OF THE SERVER USING 3-IP CAMERAS IN OUTDOOR AT
DAYTIME
blocked one. So, each thread will operate separately from
others that cause complexity and problems, this situation added
more programming-complexity. To overcome the problem of
grouping all threads with the main one to be illustrated as one
system, the proposed system used a delegate-programming-
technique for each thread in the system. The delegate
conjunction all the secondary threads with the main thread and
all results obtained by the secondary threads appear at main
thread. Fig. 4 represents friendly use Graphical User Interfaces
(GUIs) of the server and fig. 5 represents the GUI of the client.

Fig. 4. GUI of the Server

117

Authorized licensed use limited to: UNIVERSITY OF BIRMINGHAM. Downloaded on May 11,2020 at 03:04:16 UTC from IEEE Xplore. Restrictions apply.
 4th Scientific International Conference – Najaf – IRAQ (4th SICN-2019)

Table III. DATAFLOW OF THE SERVER USING 3-IP CAMERAS IN

OUTDOOR AT NIGHT TIME
Dataflow in Kilo Byte

Time in Sec.

Time (Sec.) 1 IP Camera 2 IP Camera 3 IP Camera

Fig. 6. Dataflow of the server using 3-IP cameras at outdoor at daytime

TABLE II. DATAFLOW OF THE SERVER USING 3-IP CAMERAS AT INDOOR AT

DAYTIME

Dataflow in Kilo Byte

Time in Sec.

Time (Sec.) 1 IP Camera 2 IP Camera 3 IP Camera

Fig. 8. Dataflow of the server using 3-IP cameras in outdoor at night

time

B. Server Broadcasting Video Signals to all Clients in the

Network
After server received video signals from all connected IP
cameras, it will be able to distribute the signals to all connected
Dataflow in Kilo Byte

clients in the network. Table IV represents dataflow of the

server after broadcasting the video signals of three IP cameras
to six clients. The server has the ability to distribute video
signals in three cases. Case one: server broadcasts video
signals to only one client (unicast). Case tow: server broadcasts
video signals to a group of clients in the network (multicast).
Case three: server broadcasts video signals to all connected
clients in the network. The obtained results of dataflow are
illustrated in Table IV and been plotted to get the related
Time in Sec. graphs as shown in Fig. 9. It is clear that the rate of the sending
dataflow increases with increasing the number of clients. The
Time (Sec.) 1 IP Camera 2 IP Camera 3 IP Camera
distance between source and destination is approximately 100
meters.
Fig. 7. Dataflow of the server using 3-IP cameras in indoor at daytime

118

Authorized licensed use limited to: UNIVERSITY OF BIRMINGHAM. Downloaded on May 11,2020 at 03:04:16 UTC from IEEE Xplore. Restrictions apply.
 4th Scientific International Conference – Najaf – IRAQ (4th SICN-2019)

Table IV. DATAFLOW OF THE SERVER USING 3-IP CAMERAS AFTER with load increasing. Table V illustrates a comparison of some
BROADCASTING TO SIX CLIENTS basic features used in some closest previous works with the
proposed system.

TABLE V. COMPARISON BETWEEN PROPOSED SYSTEM AND RELATED

PREVIOUS WORKS

Note: The symbol means the feature is used in the related

reference

V. CONCLUSION
In this paper, we have proposed and implemented a system
for distributing real time video signals to clients in the local
network. Obtained results can be classified in two parts: first
part when server received video signals in a specific period of
time from the same scene in two different times (day times and
night time), the results demonstrate that the size of received
data of the video’s frames at day time is more than the size of
received data at night time, in this part we included that the
light has an important factor to increase the size of data. The
second part of the results is when the server broadcast video
signals to clients in the network, the results demonstrate that
increase number of the clients in the network will rises the
network load and this will lead to losing of data.

REFERENCES
[1] W. Daar, Distribution Agnostic Video Server. 2010.
[2] P. Uppu and S. Kadimpati, “QoE of video streaming over LTE network,”
2013.
Fig. 9. Dataflow of the server using 3-IP cameras after broadcasting to six [3] R. Gill, T. Farah, and L. Trajkovic, “Comparison of WiMAX and ADSL
clients performance when streaming audio and video content,” 2011.
[4] L. Bellido, C. M. Lentisco, M. Aguayo, and E. Pastor, “Supporting
handover between LTE video broadcasting and unicast streaming,”
C. Discussion and Comparison with Previews Works presented at the Next Generation Mobile Applications, Services and
Technologies, 2015 9th International Conference on, 2015, pp. 329–334.
[5] A. G. Sheshjavani, B. Akbari, and H. R. Ghaeini, “CMPVoD: A cluster
A camera needs a light to produce high quality of image or mesh-based architecture for VoD streaming over hybrid CDN-P2P
video. The proposed system tested receiving video signals in networks,” presented at the Telecommunications (IST), 2016 8th
three different ways. First way during natural light (sunlight) International Symposium on, 2016, pp. 783–788.
[6] T. Kawakami, Y. Ishi, S. Matsumoto, T. Yoshihisa, and Y. Teranishi, “An
when installs an IP camera in outdoor. Second way during implementation of a video effect process allocation scheme for Internet
natural light (sunlight) when installs an IP camera in indoor. live broadcasting,” presented at the Consumer Electronics, 2016 IEEE 5th
And the last way during Electric light (nighttime) when installs Global Conference on, 2016, pp. 1–2.
an IP camera in outdoor. As shown in the tables I to III the [7] J. Scheer, C. F. de Tejada Quemada, and O. Grau, “A client-server
architecture for real-time view-dependent streaming of free-viewpoint
server received dataflow of size 12131, 11111 and 11021 video,” presented at the 3DTV Conference: The True Vision-Capture,
kilobytes during 10 seconds from three IP cameras that mean Transmission and Display of 3D Video (3DTV-CON),2017,2017,pp. 1–4.
the received data increased with increasing light intensity. At [8] S. C. Perez, H. A. Facchini, L. Marrone, and F. A. Hidalgo,
the server side it is clear that the rate of the sending dataflow “Experimental Study of Multicast and Unicast Video Traffic in WAN
Links,” IEEE Latin America Transactions, vol. 15, no. 10, pp. 1847–
increases with increasing number of clients, in other words 1855, 2017.

119

Authorized licensed use limited to: UNIVERSITY OF BIRMINGHAM. Downloaded on May 11,2020 at 03:04:16 UTC from IEEE Xplore. Restrictions apply.
 4th Scientific International Conference – Najaf – IRAQ (4th SICN-2019)

[9] K. I. Z. Apu, N. Mahmud, F. Hasan, and S. H. Sagar, “P2P video

conferencing system based on WebRTC,” presented at the Electrical,
Computer and Communication Engineering (ECCE), International
Conference on, 2017, pp. 557–561.
[10] O. El Marai, T. Taleb, M. Menacer, and M. Koudil, “On Improving
Video Streaming Efficiency, Fairness, Stability, and Convergence Time
Through Client–Server Cooperation,” IEEE Transactions on
Broadcasting, vol. 64, no. 1, pp. 11–25, 2018.
[11] R. E. Brydia, B. E. Brackin, R. F. De Roche, J. D. Johnson, and G. B.
Thomas, “Video Over IP Design Guidebook”, Texas Transportation
Institute, The Texas A&M University System College Station, Texas
77843-3135, Published December 2009.
[12] M. S. M. Badhusha, K. Duraiswamy, and G. Subha, “Effective Video
Streaming with Low Latency over MANET.”, International Journal of
Advanced Research in Computer Engineering & Technology, Volume 3
Issue 4, April 2014.
[13] S. R. Zeebaree and I. M. Zebari, “Multilevel Client/Server Peer-to-Peer
Video Broadcasting System.”, International Journal of Scientific &
Engineering Research, Volume 5, Issue 8, August-2014.
[14] C. S. Guynes and J. Windsor, “Revisiting client/server computing,”
Journal of Business & Economics Research, vol. 9, no. 1, pp. 17–22,
2011.
[15] S. C. Yadav and S. K. Singh, “An introduction to client server
Computing”, Published by New Age International (P) Ltd. 2009.
[16] Abdulazeez, A. M., & Zeebaree, S. R. (2018). Electronic Learning
System for Duhok Polytechnic University. Academic Journal of Nawroz
University, 7(3), 249-258.

120

Authorized licensed use limited to: UNIVERSITY OF BIRMINGHAM. Downloaded on May 11,2020 at 03:04:16 UTC from IEEE Xplore. Restrictions apply.