IJCNC 07

DYNAMIC INTERFERENCE SUPPRESSION FOR TV
WHITE SPACE: THE CASE OF THAILAND

Prathana Thaopanya 1

and Teerapat Sanguankotchakorn2

1Thai Public Broadcasting Service (ThaiPBS), Bangkok, Thailand

2 School of Engineering and Technology, Asian Institute of Technology, Thailand

ABSTRACT

In this work, we study the problem of co-existence between the LTE (Long Term Evolution) and the Digital
Terrestrial Television Broadcasting (DTTB) channel. There are three scenarios: co-channel, upper and
lower adjacent channels. We use the broadcasted signal from channel 3 called ThaiPBS, the actual case of
Thailand for our study where the standard of digital terrestrial television broadcasting is DVB-T2 adopting
8 MHz bandwidth, while the 5 MHz bandwidth of LTE is considered as the interference. We propose a
dynamic interference suppression method for increasing spectrum usage by optimizing TV white space
utilization and minimizing interference. This method adopts the protection ratio concept to suppress the
LTE interference on TV receiver. We implement our proposed algorithm as an adaptive interference
controller using a Radio Frequency (RF) attenuator and a Raspberry Pi board for our testbed hardware.
We illustrate the effectiveness of our proposed algorithm by doing experiment using our testbed and
assessing the quality of the received TV signal by adopting the Quality of Experience (QoE) assessment. In
our testbed hardware, a Log Periodic antenna is used for receiving the DTTB signal, while an RF digital
transmitter is used for generating a 5MHz bandwidth of LTE signal, an ultra-high frequency (UHF) mixer
is used to combine both signals, then a field strength meter is used to monitor video picture quality and to
analyze the spectrum. According to the experiment, our proposed method can reduce the perceived video
distortion by at least 62.5% for co-channel and 87.5% for adjacent channel, while the spectrum usage is
increased by 100%.

KEYWORDS

Adjacent Channel, Co-channel, Protection Ratio, Quality of Experience, Radio Frequency Attenuator,
Raspberry Pi, TV White Space, Visibility Threshold
1.INTRODUCTION
In digital terrestrial television broadcasting system, the VHF (Very High Frequency) and UHF
(Ultra High Frequency) frequency bands are assigned by regulators for television broadcasting
service of licensed users. However, there are many in-band frequencies that are unused in each
area. That is, there are frequencies available for secondary or unlicensed users. In addition, the
need of frequency spectrum for wireless and mobile communication services, i.e., wireless
broadband Internet access, IoT network, etc,. has been significantly increasing in recent years
[24]. One good example is the reassignment of 700MHz frequency band for mobile
communications [19].
Since not all the designed channels have been deployed in terrestrial television broadcasting in
any given region, therefore, the channel called TV White Spaces (TVWS) in which the channels
that are not used for broadcasting may be available for the other purposes. In general, the TV
White Space spectrum ranges from 470 MHz to 790 MHz, but it may be different on any region
[18]. For instance, in Thailand and some countries, the frequency spectrum of 510 MHz to 790
MHz has been assigned for Digital Terrestrial Television Broadcasting [15].

Actually, there were several TVWS trials conducted by various countries [9][11][13]~[15] [18].
In 2011, the largest TVWS trials occurred in Cambridge, and there was an implementation of the
TVWS to connect the city’s infrastructure in Wilmington, North Carolina. In 2013, the TVWS
was used to power a “super Wi-Fi network” in West Virginia University. In 2014, NICT Japan
and partner confirmed a successful implementation of long-range wireless communication by
using IEEE 802.22 and IEEE 802.11af based systems in Tono city, Japan [18]. In Thailand, NBTC
also conducted a first TVWS trial to provide the wireless broadband access in a rural area [15].
Regarding the spectrum usage in Thailand nowadays, although the demand is rapidly increasing,
it is not enough. During the last year, there was the reallocation of the 700 MHz band of frequency
channel used for DTTB to the mobile broadband system [19]. Presently, only the bandwidth of
184 MHz is left for the DTTB. Since the digital terrestrial television broadcasting system in
Thailand has been operating in both Single Frequency Network (SFN) and Multi-Frequency
Network (MFN), this can save the frequency spectrum on UHF (Ultra-High Frequency) band and
can avoid the interference as well. However, only five frequency channels are adopted in some
areas. This results in many unused frequency spectrums and inefficient usage. In case of
coexistence between TV broadcasting signal and wireless broadband signal such as LTE, the LTE
causes interference in DTTB system. In [17], they found that this situation happens when the
traffic load of LTE base station exceeds 80% of its capacity in rural and suburban scenarios. In
ITU recommendation [7], the protection ratios for DVB- T2 being interfered by LTE base station
and user equipment were analyzed. This shows that different traffic load can cause different
protection ratio. A dynamic interference management has been proposed before in the literature
[17]. However, it investigated only rural and suburban scenarios in Havana, Cuba, and Ghent,
Belgium. Certainly, these scenarios are different from the scenario in Thailand due to different
terrain, frequency band allocation and network topology. Additionally, the specifications of the
broadcasting transmitter, the propagation environment and the modulation and coding schemes
are different as well.
Based on the rationale described above, we are interested in the flexible approach that can analyze
the interference case-by-case and the appropriate analysis will lead to the high efficiency of the
system. Furthermore, for future development of wireless communications operated on TVWSs, it
also needs the optimization of spectrum usage efficiency without affecting the users’ Quality of
Service (QoS) and Quality of Experience (QoE).
In this work, the advantages of TVWS are thoroughly studied. The solution to the problem of
operating TVWS in Thailand effectively is investigated. We propose a dynamic interference
management for TVWS, especially in Thailand. In our system, we consider the scenario of DTTB
signal being interfered by LTE signal. Then, we implement our proposed system using Raspberry
Pi as a testbed. One of the inputs to our hardware testbed is the DTTB signal which is actually
received from the Thai Public Broadcasting Service (ThaiPBS) network, which is the major public
broadcasting network in Thailand, while the LTE signal is generated from the signal generator.
We consider the interference of frequency spectrum of 3 scenarios: co-channel, upper and lower
adjacent channels. The effectiveness of our proposed method is evaluated by QoE assessment
method. The contribution of this work is the hardware devices which could be used to suppress
the interference of LTE on DTTB signals.
The rest of this paper is structured as follows: In section 2, we describe the relevant literature
review. Section 3 presents the proposed method including the detailed parameters, algorithms, all
of details of testbed and assessment method, while Section 4 describes the experimental results
including program and testbed validations and assessment results. Finally, we conclude our work
and future work in section 5

2. RELATED WORK
3. indoor and outdoor Customer Premise Equipment (CPEs) in rural environment. The initial
4. transmission power level of the IEEE 802.22 signal was set with the protection ratio to achieve
5. the bit error rate (BER) of 2×10-4
6. . In [11], protection of incumbent service in TVWS was
7. considered. In [5], the significant impact of terrain and frequency on the protection distance was
8. shown. The separation distance values depending on type of Cognitive Radio (CR) device,
9. antenna height and co-channel or adjacent channel adoption. In the interference protection
10. requirements of the FCC, portable CR devices should comply with the minimum co-channel
11. separation distance of 4 km. For instance, in [21], they investigated the coexistence of the DVBT/T2 and LTE downlink services in co-channel, considered interfering LTE signals with different
12. bandwidths. Moreover, the modulation error ratio (MER) is used to evaluate performances of
13. DVB-T/T2 systems. From the study, in co-channel coexistence scenarios, unwanted narrowband
14. interfering LTE signals have less impact on DVB-T/T2 performance than the broadband. In [10],
15. the LTE radio planning pertaining to the maximum acceptable LTE radio interface load, up to
16. which a targeted user data rate can be maintained, was studied. The model of LTE radio scheduler
17. was given. It provides the optimum traffic balancing for two cells. The work of [11] was studied
18. under pure LTE radio spectrum and was analyzed using the actual data from commercial LTE
19. networks. Some works in the literature studied the impact of interference in the co-channel
20. scenario [10],[25]. In [25], the transmitter power on co-channel interference was studied.
21. However, the signal under consideration is WiFi.
22. In [1], the study of the SFN which compares field measurement results with simulation results of
23. propagation models using the PROGIRA software was carried out. It was found that the
24. Okumura-Hata model provides the smallest average error for a suburban area and for a path with
25. obstruction. In [17], the experiment was performed to find how frequent the protection ratio limits
26. are exceeded. Both TOV (Threshold of Visibility; defined in Report ITU-R BT.2035-2/2008) and
27. SFP (Subjective Failure Point; defined in Recommendation ITU-R BT.1368-13/2017) have been
28. defined as a criterion to find a limit for a just error-free picture at a TV screen for protection ratio
29. measurements.
30. PROPOSED MODEL
    3.1. System Model and Parameters
    The block diagram of the proposed dynamic interference analysis system is shown in figure 1. In
    our work, we consider the scenario where DVB-T2, the digital terrestrial television broadcasting
    (DTTB) standard adopted in Thailand, is a primary service coexisting with mobile wireless
    communication (LTE) which is a secondary service. Both signals are the inputs of the dynamic
    interference suppression algorithm, which is implemented on Raspberry Pi, our programmable
    testbed. Then, we assess the effectiveness of our implemented algorithm in reducing the LTE
    interference on DTTB signal by the QoE assessment method. The detail of all these procedures
    is described in the following subsections.