Development of SDN-based wi-fi AP using openwrt and raspberry pi 3

d either Ethertype is LLDP or the 
packet's destination address is a Bridge Filtered address, 
then drop packet and return Step 1. 
 If destination is multicast, then flood the packet and 
return Step 1. 
 If port for destination address in our address/port table, 
then flood the packet and return Step 1. 
 If output port is the same as input port, then drop packet 
and similar ones for a while and return Step 1. 
 Install flow table entry in the switch so that this flow 
goes out the appropriate port and then, send the packet 
out appropriate port. 
 Go back to Step 1. 
III. EXPERIMENTAL RESULTS AND DISCUSSION 
In this section, key designed parameters of the 
developed SDN-Wi-Fi access point device were tested and 
verified. Our developed device can support typical Wi-Fi 
standards including 802.11 b/g/n and be compatible with 
Openflow 1.5. It also can be controlled by both popular 
interfaces including graphical user interface (GUI) and 
command line interface (CLI). The device networking 
functions are deployed and controlled by an SDN 
controller. Thanks to using OVS, the device is compatible 
with typical SDN controllers such as Opendaylight, Ryu, 
 However, for simplicity, we used POX version 0.5.0 in 
our experiments. We have also evaluated the performance, 
in terms of bandwidth, of the device in different 
experimental scenarios. 
A. System Configuration Verification 
We have practically tested our developed prototype of 
SDN-based Wi-Fi AP in order to verify the designed 
configuration. We set up an experimental testbed including 
our SDN-based Wi-Fi device controlled by a POX 
controller which is installed in a laptop and a mobile phone 
as a Wi-Fi access device. Note that our Open vSwitch can 
work well with other SDN controllers such as Ryu, NOX, 
Opendaylight,  
Figures 3 and 4 describe the OpenFlow configuration 
of our developed device as designed. It was demonstrated 
that the AP prototype was successfully connected to the 
controller at the IP address of 192.168.1.10 through a TCP 
port (TCP port number of 6633). Moreover, the versions of 
DEVELOPMENT OF SDN-BASED WI-FI AP USING OPENWRT AND RASPBERRY PI 3 
the deployed OVS software and the operating OpenFlow 
protocol were 2.8.5 and 1.5 respectively. 
Figure 3. SDN configuration of the Wi-Fi AP prototype. 
Figure 4. Versions of SDN components (OpenFlow and Open 
vSwitch). 
The SDN-based Wi-Fi AP prototype is then activated 
as a wireless switch by applying the corresponding POX 
control program (forwarding.l2_learning) which has been 
introduced in the section II. Figure 5 shows that the device 
worked properly to provide a flow connection. In fact, 
depending on the applied control program, our device can 
be deployed with any appropriate networking functional 
device like switch, router, firewall,  
Figure 5. CLI of POX controller on the laptop. 
Moreover, Figure 6 depict the Web-based configuration 
and O&M (Operation and Maintenance) interface of the 
developed device. Similar to those of common Wi-Fi Aps, 
the device configuration in term of Wi-Fi parameters 
including channel number, Wi-Fi standard version and 
associated station information can be set up and controlled. 
Our device can support three Wi-Fi standards, 802.11b/g/n. 
As being demonstrated in the Figure 6, one laptop was 
connected through Wi-Fi 802.11g with the speed of 54 
Mbit/s on the channel number 11 with the carrier frequency 
of 2.462 GHz. Detailed information including MAC 
address, host name, current signal/noise value and RX/TX 
rates can be also provided on the Web interface. 
Figure 6. Configurable Web-based interface of the SDN-based 
Wi-Fi AP prototype. 
B. Performance Evaluation 
In this part, we have practically tested the performance, 
in terms of connection speed, of our developed SDN-based 
Wi-Fi AP. Using a similar experimental setup to that from 
the previous part, the developed device is controlled by a 
POX controller installed in a laptop to provide Wi-Fi access 
while a mobile or a laptop is used as a Wi-Fi station. The 
Wi-Fi station used Wi-Fi 802.11g which offers the 
maximum speed of 54 Mbps. Each experimental scenario 
was repeated five times and its average result was then 
calculated and summarized. 
Figure 7. Transmission speed. 
Figure 7 shows the obtained speed of the developed 
prototype when the traffic generated by the mobile station 
was increasing from 5 Mbps to 60 Mbps. The results 
confirm that the maximal transmission speed is about 54 
Mbps, the limited rate of 802.11g. The received rate is 
increasing with the generated traffic when it is less than 54 
Mbps however, when the transmitted traffic is greater than 
54 Mbps, the mobile station can receive only up to 54 Mbps 
and drop the remaining traffic. 
Hai-Chau Le and Khac-Tuan Nguyen 
Figure 8. Dependence of bandwidth on distance. 
 In addition, in order to estimate the cover range of the 
SDN-based Wi-Fi AP, we tested and measured the 
transmission speed with different connection distance (the 
distance between the AP and the mobile station). Figure 8 
shows the impact of connection distance on the bandwidth. 
It informs that the attained bandwidth is decreased when 
the distance becomes longer. The cover distance of the 
developed AP is limited due to the use of the built-in Wi-
Fi card. In fact, the cover range will be extended by using 
a suitable external antenna however, this extention is 
ignored in this paper and will be investigated more details 
in next stage of the research to figure out suitable 
applications for the developed device. 
Figure 9. The percentage of received power versus distance. 
Finally, we also measured the received power of the 
mobile station to explain the dependence of the 
performance on the connection distance. Figure 9 
demonstrates the dependence of the received power on the 
connection distance. The power received at the mobile 
station is significantly reduced as the mobile station is 
moving far from the AP. This power decrease explains why 
the bandwidth of the Wi-Fi connection is lessen with the 
increase of the connection distance. Again, in order to cope 
with long range connections, it is required to equip the 
developed SDN-based Wi-Fi AP device with an 
appropriate external antenna. 
IV. CONCLUSION 
In this paper, we have successfully developed a cost-
effective scalable SDN-based Wi-Fi AP prototype that 
enables OpenFlow 1.5 for IoT communication. The 
proposed device, that is based on a Raspberry pi 3 and 
therefore, it is very cost-effective, uses OpenWrt firmware 
and Open vSwitch software. By using SDN technology, our 
developed device can be deployed as a network device with 
various functions such as hub, switch, firewall, ... which 
depend on Python-based controlling program installed in 
the SDN controller. The SDN-based Wi-Fi AP 
performance has been verified by testing experiments. The 
obtained numerical results proved the effective and 
scalable performance of the developed SDN-based Wi-Fi 
AP prototype. With a built-in Wi-Fi card and without 
external antenna, the SDN-Wi-Fi access point prototype 
can cover the range of about 20 meters. This device can be 
a promising approach for creating flexible and effective 
IoT network devices. 
REFERENCES 
[1] Bizanis, Nikos, and Fernando A. Kuipers, “SDN and virtualization 
solutions for the Internet of Things: A survey”, IEEE Access, 
vol. 4, pp. 5591-5606, 2016. 
[2] Samaresh Bera, Sudip Misra, and Athanasios V. Vasilakos, 
“Software-Defined Networking for Internet of Things: A Survey”, 
IEEE Internet of Things Journal, vol. 4, no. 6, pp. 1994-2008, 2017. 
[3] Qin Zhijing, Grit Denker, Carlo Giannelli, Paolo Bellavista, and 
Nalini Venkatasubramanian, “A software defined networking 
architecture for the internet-of-things”, 2014 IEEE network 
operations and management symposium (NOMS), pp. 1-9, 2014. 
[4] Bizanis Nikos, and Fernando A. Kuipers, “SDN and virtualization 
solutions for the Internet of Things: A survey”, IEEE Access, vol. 
4, pp. 5591-5606, 2016. 
[5] Bakhshi, Taimur, “State of the art and recent research advances in 
software defined networking”, Wireless Communications and 
Mobile Computing 2017, 2017. 
[6] Diego Kreutz, Fernando M.V. Ramos, Paulo Esteves Verıssimo, 
Christian Esteve Rothenberg, Siamak Azodolmolky, and Steve 
Uhlig, “Software-defined networking: A comprehensive 
survey,” Proceedings of the IEEE, vol. 103, no. 1, pp. 14-76, 2015. 
[7] Hu Fei, Qi Hao, and Ke Bao, “A survey on software-defined 
network and openflow: From concept to implementation”, IEEE 
Communications Surveys & Tutorials, vol. 16, no. 4, pp. 2181-
2206, 2014. 
[8] Haque Israat Tanzeena, and Nael Abu-Ghazaleh, “Wireless 
software defined networking: A survey and taxonomy”, IEEE 
Communications Surveys & Tutorials, vol. 18, no. 4, pp. 2713-
2737, 2016. 
[9] Jararweh Yaser, Mahmoud Al-Ayyoub, Elhadj Benkhelifa, Mladen 
Vouk, and Andy Rindos, “SDIoT: a software defined based internet 
of things framework”, Journal of Ambient Intelligence and 
Humanized Computing, vol. 6, no. 4, pp. 453-461, 2015. 
[10] Vipin Gupta, Karamjeet Kaur and Sukhveer Kaur, “Developing 
Small Size Low-Cost Software-Defined Networking Switch Using 
Raspberry Pi,” Next-Generation Networks, pp. 147-152. Springer, 
Singapore, 2018. 
[11] Kim, Hyunmin, Jaebeom Kim, and Young-Bae Ko, “Developing a 
cost-effective OpenFlow testbed for small-scale Software Defined 
Networking," IEEE 16th International Conference on Advanced 
Communication Technology (ICACT), pp. 758-761, 2014. 
[12] Behnam Dezfouli, Vahid Esmaeelzadeh, Jaykumar Sheth, and 
Marjan Radi, “A review of software-defined WLANs: Architectures 
and central control mechanisms,” IEEE Communications Surveys 
& Tutorials 21, no. 1, pp. 431-463, 2018. 
[13] Hisham Elzain, and Yang Wu, “Software Defined Wireless Mesh 
Network Flat Distribution Control Plane,” Future Internet, vol. 11, 
no. 8, pp. 166- , 2019. 
[14] Won Jin Lee, Jung Wan Shin, Hwi Young Lee, and Min Young 
Chung, “Testbed implementation for routing WLAN traffic in 
software defined wireless mesh network,” 2016 Eighth International 
Conference on Ubiquitous and Future Networks (ICUFN), pp. 
1052-1055, 2016. 
0
20
40
60
80
100
1 3 4 5 6 7 12 14 20
R
ec
ei
v
ed
 P
o
w
er
(%
)
Distance (m)
DEVELOPMENT OF SDN-BASED WI-FI AP USING OPENWRT AND RASPBERRY PI 3 
[15] Kristián Košťál, Rastislav Bencel, Michal Ries, Peter Trúchly, and 
Ivan Kotuliak, “High Performance SDN WLAN 
Architecture,” Sensors, vol. 19, no. 8, pp. 1880- , 2019. 
[16] Ahmed Abdelaziz, Ang Tan Fong, Abdullah Gani, Suleman Khan, 
Faiz Alotaibi, Muhammad Khurram Khan, “On Software-Defined 
Wireless Network (SDWN) Network Virtualization: Challenges and 
Open Issues”, The Computer Journal, Volume 60, Issue 10, pp. 
1510–1519, October 2017. 
[17] Hsu, Huai‐Wen, Kuei‐Li Huang, Yi‐Chih Kao, Shi‐Chun Tsai, and 
Yi‐Bing Lin, “Deploying WLAN service with OpenFlow 
technology”, International Journal of Network Management 27, no. 
3, pp. e1970, 2017. 
[18] Alshnta, Abdalkrim M., Mohd Faizal Abdollah, and Ahmed Al-
Haiqi, “SDN in the home: A survey of home network solutions using 
Software Defined Networking”, Cogent Engineering 5, no. 1, pp. 
1469949, 2018. 
[19] Miano Sebastiano, and Fulvio Risso, “Transforming a traditional 
home gateway into a hardware-accelerated SDN switch”, 
International Journal of Electrical and Computer Engineering 10, 
no. 3, pp. 2668, 2020. 
[20] OpenWrt offical homepage, https://openwrt.org/ 
[21] John C. Shovic, “Raspberry pi IoT projects,” Apress, 2016. 
[22] “Raspberry Pi Foundation.”, https://www.raspberrypi.org/ 
[23] https://openflow.stanford.edu/display/ONL/POX+Wiki.html 
[24] OpenvSwitch Offical Homepage, https://openvswitch.org 
[25] https://standards.ieee.org/standard/802_11-2012.html 
[26] 802.11-2012 - IEEE Standard for Information technology–
Telecommunications and information exchange between systems 
Local and metropolitan area networks–Specific requirements Part 
11: Wireless LAN Medium Access Control (MAC) and Physical 
Layer (PHY) Specifications. Tech. Rep. IEEE Std 802.11TM-2012, 
IEEE-Inst. 
[27] Wi-Fi Alliance,  
NGHIÊN CỨU PHÁT TRIỂN HỆ THỐNG TRUY 
NHẬP WI-FI ĐỊNH NGHĨA BẰNG PHẦN MỀM SỬ 
DỤNG RASPBERRY PI 3 VÀ OPENWRT 
Lê Hải Châu và Nguyễn Khắc Tuấn 
Học viện Công nghệ Bưu chính Viễn thông 
* Email liên hệ: chaulh@ptit.edu.vn 
Tóm tắt: Trong bài báo này, chúng tôi đã nghiên cứu và 
phát triển thành công một mẫu thiết bị truy nhập Wi-Fi định 
nghĩa bằng phần mềm cho truyền thông IoT. Thiết bị được 
phát triển dựa trên nền tảng Raspberry pi 3, sử dụng 
firmware OpenWrt dựa vào nhân Linux và phần mềm 
chuyển mạch SDN mã nguồn mở Open vSwitch. Hệ thống 
này có khả năng tương thích với giao thức OpenFlow 1.5 
và hỗ trợ giao diện WAN qua một cổng Ethernet dựng sẵn 
100 Mbps. Nhờ việc khai thác các ưu điểm của công nghệ 
SDN, công nghệ mã nguồn mở và nền tảng máy tính cỡ nhỏ 
giá thành rẻ, mẫu thiết bị truy nhập Wi-Fi định nghĩa bằng 
phần mềm có giá thành chấp nhận được trong khi vẫn rất 
linh hoạt, có khả năng mở rộng tốt và cho phép hỗ trợ đầy 
đủ các tính năng mạng tiên tiến. Thiết bị này cũng có thể 
được triển khai thành các thiết bị mạng với các tính năng 
khác nhau như hub, switch, router hay firewall,  bằng 
cách sử dụng phần mềm điều khiển được lập trình và cài 
đặt trên bộ điều SDN. Cấu hình và hiệu năng của mẫu thiết 
bị này được kiểm thử và xác minh thông qua các thí nghiệm 
đo kiểm. Các kết quả đạt được đã thể hiện sự hiệu quả và 
khả năng mở rộng linh hoạt của thiết bị được phát triển. 
Từ khoá: Mạng định nghĩa bằng phần mềm, điểm truy 
nhập Wi-Fi, OVS, OpenFlow. 
Hai-Chau Le received the B.E. 
degree in Electronics and 
Telecommunications Engineering 
from Posts and 
Telecommunications Institute of 
Technology (PTIT) of 
Vietnam in 2003, and the M.Eng. 
and D.Eng. degrees in Electrical 
Engineering and Computer 
Science from Nagoya University 
of Japan in 2009 and 2012, 
respectively. From 2012 to 2015, 
he was a researcher in Nagoya 
University of Japan and in 
University of California, Davis, 
USA. He is currently a lecturer in 
Telecommunications Faculty at 
PTIT. His research interests 
include optical technologies, 
network design and optimization 
and future network technologies. 
He is an IEEE member. 
Khac-Tuan Nguyen is a 5th year 
bachelor student in the field 
of Electronics and 
Telecommunications Engineering 
at Posts and 
Telecommunications Institute 
of Technology of Vietnam, and is 
currently an active member of 
Information network research 
group of Telecommunications 
Faculty I in PTIT. His research 
interests include SDN, NFV, and 
future network technologies.