Measuring the local impact of TVWS broadband Richard Thanki ECS Partners, University of Southampton richard.thanki@soton.ac.uk
Evaluating the impact of TVWS broadband The internet has had a tremendous economic and social impact across all areas of human activity TVWS, as an unlicensed radio technology, holds the promise of providing connectivity for people who have so far been excluded by current high cost connection technologies Understanding the potential impacts of TVWS broadband in these contexts is important for regulators, policy makers and potential operators To begin answering these questions we have begun a detailed M&E exercise of three of the Microsoft supported trials: Kenya Mawingu Ghana djungle Wi-Fi Namibia Citizen Connect This presentation focuses on the low-cost, solar powered, public access Wi-Fi being deployed by Mawingu in Kenya
The successes of unlicensed spectrum A substantial amount has been written about the tremendous success of unlicensed spectrum In short: Hundreds of protocols (Wi-Fi, Bluetooth, proprietary) Thousands of use cases Tens of thousands of manufacturers Majority of the world s wireless chipsets Vast majority of the world s internet data delivered in only a sliver of shared spectrum Unlicensed access to the TVWS has the potential to enable low-cost longer range wireless applications, such as rural broadband access
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 Millions An increasingly connected world 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 Fixed Telephone lines Mobile subscribers Internet users World population Year Thanki 2015 (from various sources)
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Internet users per 100 but with large variations between countries 100 90 80 70 60 50 40 30 20 10 0 Countries at differing levels of HDI Very High High Medium Low Year Thanki 2015 (using ITU and UN data)
Internet users per 100 Internet penetration in Southeast Asia 100 90 80 70 60 50 40 30 20 10 0 Singapore Malaysia Brunei Vietnam Philippines Thailand Indonesia Lao Cambodia Burma East Timor Thanki 2015 (using ITU and UN data)
Internet users per 100 Internet penetration in the three M&E countries 100 90 80 70 60 50 40 30 20 10 0 Kenya Namibia Ghana Thanki 2015 (using ITU and UN data)
Unaffordability is the main barrier to connectivity To date 3 billion people around the world use the internet. For the remainder, access to broadband remains unaffordable for the majority (2 2.6 billion) The incomes available to these people is not enough to cover the costs of even basic mobile broadband Our existing methods of delivering broadband (fixed, mobile and satellite) are too costly to serve these populations
The first billion (by income) Micronesia Melanesia Eastern Africa Northern Africa Polynesia Middle Africa South America Southern Africa Western Africa Central Asia Australia and New Zealand Eastern Asia South-Eastern Asia Southern Asia Northern America Western Asia Eastern Europe Central America Caribbean Western Europe Northern Europe Southern Europe Western Asia Average income = $49,266 pa Affordable comms spend = $205 pm Thanki 2015
The second billion Micronesia Melanesia Australia and New Zealand Polynesia Eastern Africa Northern America Middle Africa South America Northern Africa Southern Africa Western Africa Central Asia Central America Western Europe Caribbean Eastern Asia Western Asia Southern Europe Northern Europe Eastern Europe Western Asia Southern Asia South-Eastern Asia Average income = $12,722 pa Affordable comms spend = $53 pm Thanki 2015
The third billion Australia and New Zealand Central America Caribbean Western Europe Western Asia Micronesia Eastern Africa Polynesia Melanesia Northern America South America Middle Africa Northern Africa Southern Africa Western Africa Central Asia Southern Europe Northern Europe Eastern Europe Western Asia Southern Asia South-Eastern Asia Eastern Asia Average income = $5,540 pa Affordable comms spend = $23 pm Thanki 2015
The fourth billion Central America Melanesia South America Middle Africa Northern America Eastern Africa Western Polynesia Europe Caribbean Micronesia Northern Africa Southern Africa Southern Europe Western Asia Western Africa Northern Europe Central Asia Eastern Europe Western Asia Southern Asia Eastern Asia South- Eastern Asia Average income = $2,987 pa Affordable comms spend = $12 pm Thanki 2015
The fifth billion Eastern Africa Middle Africa Australia and New Northern Africa Western Southern Northern Eastern Northern Central Western Caribbean Micronesia South Melanesia Polynesia Zealand Asia Europe Asia America Southern Africa Western Africa Central Asia Southern Asia Eastern Asia South- Eastern Asia Average income = $1,771 pa Affordable comms spend = $7 pm Thanki 2015
The sixth billion South America Southern Northern Eastern Western Central Caribbean Micronesia Melanesia Polynesia America Asia Europe Eastern Africa Middle Africa Northern Africa Southern Africa Western Africa Central Asia Southern Asia Eastern Asia South-Eastern Asia Average income = $1,065 pa Affordable comms spend = $4.4 pm Thanki 2015
The final billion (by income) Eastern Europe Southern Europe Central America Melanesia Caribbean Micronesia Polynesia South America Western Asia Eastern Africa Southern Asia Middle Africa Western Africa Northern Africa Southern Africa South-Eastern Asia Eastern Asia Central Asia Average income = $540 pa Affordable comms spend = $2.25 pm Thanki 2015
Mawingu Low-cost high performance solar panels Low-cost high performance radios Mawingu Wi-Fi and recharging Increasing smartphone penetration Mawingu is an ambitious project to use licenceexempt wireless and solar panels to provide the lowest possible cost, high performance rural nomadic Wi-Fi
The location of the Mawingu Network
Kenya baseline household survey 601 households were randomly sampled in 4 broad locations: Endana 100 households Marura 101 Matanya 299 Ngobit 101 These households contained in total 2,340 individuals In each household, we conducted detailed communications use interviews with 3 randomly selected members. If fewer than 3 members were present we interviewed all the members available. This resulted in 826 detailed communications usage interviews.
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Household access to facilities and communications For phones and PCs access is determined as being a single HH member owning the device Sample size: 601 households Thanki 2015
USD per week Weekly household income 500 450 400 350 300 250 200 150 100 50 0 0% 20% 40% 60% 80%100% Sample size: 601 households The figures below represent weekly reported incomes in USD Mean 26.80 Median 16.50 Top 25% > 27.50 Bottom 25% < 7.70 The average income per year is $358 placing this sample in the world s poorest billion Thanki 2015
Proportion of total consumption spent on communications less than 1% 1% - 2.5% 2.5% - 5% 5% - 10% 10% - 25% 25% - 50% more than 50% 64% of households spend more than 5% of their total expenditure on communications 41% are spending more than 10% Sample size: 601 households Thanki 2015
Communications device usage 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Don't know Never Occasionally Monthly Weekly Daily Sample size: 826 individuals in 601 households Thanki 2015
Communication device use by gender 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Don't know Never use Use elsewhere Have at home Own Sample size: 826 individuals in 601 households Thanki 2015
Phone usage by age category 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0-14 15-24 25-34 35-44 45-54 55-64 65+ Sample size: 826 individuals in 601 households Don't know Never use Use elsewhere Have at home Own Thanki 2015
Smartphone usage by age category 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0-14 15-24 25-34 35-44 45-54 55-64 65+ Sample size: 826 individuals in 601 households Don't know Never use Use elsewhere Have at home Own Thanki 2015
PC usage by age category 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0-14 15-24 25-34 35-44 45-54 55-64 65+ Sample size: 826 individuals in 601 households Don't know Never use Use elsewhere Have at home Own Thanki 2015
Internet usage frequency by age category 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0-14 15-24 25-34 35-44 45-54 55-64 65+ Sample size: 826 individuals in 601 households Don't know Never Occasionally Monthly Weekly Daily Thanki 2015
Barriers to using the internet more 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% There is no interesting content for me Lack of local language content Sample size: 103 individual internet users Too expensive to use The internet isfew people to very slow communicate with on the internet Other Thanki 2015
Could the internet lead to improvements? 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Community improvement Sample size: 103 individual internet users Personal improvement yes no Thanki 2015
Suggested improvements from wider availability of the internet Sample size: 103 individual internet users Thanki 2015
Suggested challenges from wider availability of the internet Sample size: 103 individual internet users Thanki 2015
Next steps This is just the beginning of the work we plan to do There are baseline surveys ongoing in Ghana and Namibia Qualitative research will lead to rich audiovisual and textual materials We plan also to conduct detailed network data analysis that will look at per user usage