By Leonard Mabele, CB4ICT Project

In the wake of the Covid-19 pandemic, the Communications Authority of Kenya (CA) – Kenya’s National Regulatory Authority (NRA), has been deeply engaged on studies to validate the draft of the regulatory framework for Television White Spaces (TVWS) in the country. These studies were spearhead by researchers within the country led by Strathmore University and working with international partners as well.

The studies were funded by the former UK Department for International Development (DFID) in December 2019 and ran from January to June 2020, receiving an extension until October 2020. DFID merged with the Foreign Commonwealth Office (FCO) in September 2020, creating the Foreign, Commonwealth and Development Office (FCDO). Notably, the CA has a memorandum of understanding (MoU) with Strathmore University until 2022 exploring research in spectrum innovation alongside cybersecurity. The research on spectrum innovation covers work on TVWS, Long Term Evolution (LTE) as well as 5G within the umbrella of dynamic spectrum access (DSA) together with technical concepts of software-defined radios (SDRs) and cognitive radios (CRs). It also involves joint development of capacity on spectrum management skills.

The rise in the spectrum demand due to the proliferation of more wireless devices such as phones, tablets, personal computers etc. and the rapidly growing Internet of Things (IoT) ecosystem requires a new spectrum management paradigm that can guarantee efficient utilisation of spectrum. Primarily, global studies point towards a situation where spectrum allocated to incumbents is not fully utilised based on time, frequency and location. That is, the channels allocated to the licensees remain unused a good fraction of the time or are not even utilised in some locations altogether. CA, as one of the forward-thinking NRAs on the continent, has already identified this gap and has been leading collaborations with other NRAs in Africa and other continents to leapfrog efforts that can drive opportunistic spectrum usage. Strathmore University, on the other hand, has been conducting studies for over three years with partners in the UK, Ghana, Zambia, Malawi and Kenya on dynamic spectrum access, specifically on TV White Spaces. Hence, this shared vision of Strathmore University and the Communications Authority of Kenya (CA) has been the torque leading the wheels of spectrum innovation to deliver the first DSA-based regulatory framework for the country. Potential applications that are envisioned to be the long-term fruits of this framework include provision of fixed wireless broadband connectivity to the rural homes, businesses, public services such as education and healthcare as well as IoT applications.

Throughout these studies, the two partners have worked on investigation of spectrum sharing for Fixed Broadband services in the UHF Television (TV) band (470-694 MHz) as an enabler of rural wireless Internet access. Field studies were conducted in Kisumu, Kitui and Laikipia counties to determine the spectral opportunity in the same UHF TV band. Engagements with the public as well as relevant stakeholders were additionally carried out in these studies to obtain the necessary views on the first step towards spectrum sharing for Kenya.

In this article, a summary of that draft regulatory framework for Kenya on TVWS is highlighted. The draft framework is to be released by the end of November 2020. The framework immensely borrows from the model rules developed by the dynamic spectrum alliance (DSAL). Other guiding standards on it include the European Telecommunications and Standards Institute (ETSI) harmonised rules, the ITU Geneva 2006 (GE06) agreement for Digital Broadcasting as well as the IETF’s Protocol to Access TV White Space Database (PAWS).

 

The Regulatory Approach

‘White Spaces’ refer to radio spectrum that is either allocated for licensed use but not assigned to a particular licensee due to limited demand, or not being used by licensees at all times and in all geographical locations. The white spaces in the frequency band of TV broadcasting services is known as TV White Spaces (TVWS). TVWS can hence, shortly be defined as the unused TV channels assigned to TV broadcasting. Unlike other countries such as the USA, where utilisation of TVWS is in both UHF and VHF bands at a bandwidth of 6MHz,  the draft of the Kenyan rules spell out that the utilisation of the TVWS will operate in the UHF band only - 470-694 MHz at a bandwidth of 8 MHz. This is also a UHF band in the ITU region 1, which includes Europe, Africa, Middle East west of the Persian Gulf including Iraq, the former Soviet Union and Mongolia. The operation will be on a non-protected and non-interference basis by the radio equipment to be deployed, i.e. the white space devices (WSDs). This approach is sensible enough as it needs to be remembered that the UHF band – 470-694 MHz (which houses 28 channels – CH21-CH48) is primarily allocated to the broadcasting services (formally referred to as digital terrestrial television - DTT) who pay for the operating license and hence have to be protected from any interference. The utilisation of the “white spaces” between these bands is therefore a secondary and opportunistic usage but highly desired due to the gigantic value it provides through the provision of wireless Internet access. TV white spaces are particularly suitable for delivering Internet access to the rural and underserved sub-urban areas due to the great propagation characteristics and penetration features they possess, particularly in the UHF band.  

Protecting the incumbents has enormously been at the centre stage of allowing spectrum sharing for TVWS utilisation. This is the same for all the other shareable spectrum bands, including 5G. The traditional licensing model gives high priority to the providers who pay to use the spectrum on an exclusive basis; they therefore must be guaranteed the desired quality of service without interference to their service pipeline.  In the era of dynamic spectrum access (DSA), three approaches can be used to establish the regulations that protect the incumbents. These are the use of spectrum sensing, the beacon reception model and the use of the geolocation databases. CA has adopted the use of the geolocation databases (mostly referred to as white space databases – WSDB) to protect the DTT service providers from the WSD interference.

Just like the way services are provided by the network facilities providers (NFPs) or the Internet Service Providers (ISPs), the geolocation databases will be delivered by providers who must meet the requirements stipulated by CA. The CA has developed the procedure for qualification of geolocation databases, which accompany the draft framework to act as the requirements for provision of geolocation database services. The procedure outlines a 3-stage approach process that includes application, evaluation and testing and finally qualification. The geolocation databases will be expected to have the ability to receive periodic updates from the CA’s spectrum database, perform coexistence calculations and provide transmission parameters to the WSDs for non-interference operation. A bit similar to the UK approach, the CA will implement a listing server that lists the approved databases to operate in Kenya as well as the operators. All of this, in line with the PAWS protocol which the database providers adhere to in the development of their service.

The white space devices (WSDs) will be expected to meet the minimum technical specifications and be Type Approved prior to installation by NFPs. They shall be authorised to operate at specific locations and times determined by the geolocation database(s) used. The WSD operational requirements and channel usage parameters will be obtained from the geolocation databases through the interfacing (described within PAWS) that the master WSDs will have with the databases. A master WSD will provide its location and operational characteristics to a geolocation database that has information on licensed DTT transmissions. The geolocation database will then use the information to determine the channels available for secondary use by the WSD and the associated power levels. The database will perform calculations for the location and technical characteristics provided by a WSD and will communicate the available channels and powers to the master WSD to initiate transmission. The consumer premises equipment (or the slave WSDs) will obtain the operating parameters from the geolocation database through the master WSD. A master WSD must periodically reload the list of qualified databases at least every 24 hours. In the initial communication to the geolocation database, the master WSDs will not use TVWS channels. An illustration of the TVWS network where the master interfaces with the geolocation database is shown in figure 1.

The licensing model adopted in the draft is the lightly licensed one. In the lightly licensing approach, the service providers will require an annual authorisation and will be expected to pay an annual fee within the range of KES. 10, 000 per master WSD in addition to a nominal fee and the fees they will have to pay to the geolocation database providers. Moreover, in this approach, the WSDs will be fully managed through the geolocation databases. In the event of harmful interference, CA will request cease of transmission on the interfering WSDs. WSDs that are not able to verify their operational parameters with the geolocation databases will also be expected to cease transmission. This will be the same for the client WSDs(CPEs or slave WSDs) that can verify their operating parameters with the master WSD(s).

Figure 1: An illustration of the TVWS network where the master interfaces with the geolocation database

 

Technical Highlights

Under the lightly-licensed regime, the white space devices (WSDs) will need to comply with the minimum technical and operational requirements to mitigate the risk of harmful interference to the incumbents (DTT service providers). The primary device to obtain the operational parameters will be a master WSD. Borrowing from the ETSI standard, type-A and type-B devices (fixed and portable WSDs) will be allowed to operate with integral or external antenna and integral or dedicated antenna respectively. All the WSDs will hence have a unique serial number provided by the manufacturer, the model number and the identifier of the device manufacturer. This will allow ease of identification of the devices by the geolocation databases, which will be able to provide the particular channels and powers being used by the specific WSDs.  Based on the emission class of the WSDs, the database(s) will be able to calculate the operational parameters.

Operational parameters generated by the geolocation database(s), will be the instructions of operation to the WSDs. These instructions include:

1. The lower and upper frequency boundaries within which a WSD may transmit.
2. The maximum permitted Effective Isotropic Radiated Power (EIRP) spectral density.
3. Limits on the maximum total number of DTT channels that may be used.
4. The time interval during which the operational parameters are valid.
5. The time interval within which a master WSD must check with a geolocation database that the operational parameters are valid.

Once a WSD obtains the operational parameters, it will have to report to the geolocation database(s) what the actual operating parameters are. These are referred to as the Channel Usage parameters. These parameters include – the location of the WSD(s), the lower and upper frequency boundaries and the maximum in-block EIRP spectral density.

An illustration of the UHF band with the DTT (broadcasting services) coexisting with services in adjacent bands is shown in figure 2. For TVWS transmission, the channels to be used by the WSDs will be limited in power to protect the DTT transmission. Similar power limits will be used to manage the out-of-band (OOB) emissions from WSDs to the 450-470 MHz mobile services. The 9 MHz guard band protects the services of the mobile networks in the 700 MHz. Cross-border emissions will be protected in alignment with the GE06 Agreement to which Kenya is a party and will be restricted to the trigger field strength levels of 21 dBµV/m for channels 21-34 (470-582 MHz) and channels 35-51 (582-718 MHz). This approach will ensure that Kenya does not cause harmful interference to the DTT services of its neighbours due to its WSD transmission.

Figure 2: The UHF band with DTT band coexisting with adjacent UHF services

 

The details of the white space availability based on frequency, height, space and time as well as the key definitions and requirements of TVWS use are discussed in detail in the draft framework. Additionally, coexistence calculations were also carried out in the DFID-funded project leading to a deviation in the thermal noise power limit for the 8 MHz bandwidth from the one provided in the DSAL rules. The coexistence calculations compute spectrum allocation parameters for WSD taking into account geolocation, device requirements for the WSD as well as the DTT protection information. All these are deeply covered in the draft framework and in the DSAL model rules. Moreover, more paper publications are being worked on and will be shared in the coming days by the researchers from Strathmore University, experts from CA and consultants from Kaiote who worked on the simulations of the calculations.

 

Conclusions

This summary is a part of the broader document on the draft regulatory framework for TV White Spaces in Kenya. In a way, it also touches and includes some links to other relevant documents that have been instrumental in the development of the regulatory framework for Kenya. It is worth noting that the framework also received inputs from various stakeholders seeking clarifications and pointing out items that required corrections. As an evolving approach to managing underutilised spectrum in order to address the digital divide problem, CA has developed a blueprint that can lead to more activities leading to dynamic spectrum access with the overall objective of ensuring efficient utilisation of the lucrative resource - radio frequency spectrum. Strathmore, on the other hand is launching a new masters course on spectrum management in the coming days. Among other significant topics, the course will cover the novel way of spectrum sharing and dynamic spectrum management approach. Moreover, it is in its final stages to complete the setup of the spectrum research division that will look at spectrum innovation, particularly in shareable bands and the scalability of the Internet of Things (IoT).