Traffic And Market Report June 2012
Upcoming SlideShare
Loading in...5
×
 

Traffic And Market Report June 2012

on

  • 508 views

 

Statistics

Views

Total Views
508
Views on SlideShare
498
Embed Views
10

Actions

Likes
0
Downloads
5
Comments
0

1 Embed 10

http://www.linkedin.com 10

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Traffic And Market Report June 2012 Traffic And Market Report June 2012 Document Transcript

  • TRAFFIC ANDMarket reportON THE PULSE OF THE NETWORKED SOCIETYJune 2012
  • CONTENTS Mobile subscriptions update 4 Subscriptions outlook 6 Subscriptions outlook: devices 9 Mobile traffic update 12 Traffic development 13 Coverage 15 Smartphone users: network quality 16 Accelerating mobile content delivery 18 Traffic variations between networks 19 Traffic variations: data plan 23The contents of this document are based on a number of theoretical dependencies The signature of humanity 25and assumptions and Ericsson shall not be bound by or liable for any statement,representation, undertaking or omission made in this document. FurthermoreEricsson may at any time change the contents of this document at its solediscretion and shall not be liable for the consequences of such changes. Key figures 272  TRAFFIC AND MARKET REPORT  JUNE 2012
  • Ericsson Trafficand Market ReportJune 2012Everything is going mobile. This evolution is driven by video,cloud-based services, the internet and machine-to-machine(M2M) connectivity. It changes how people behave and how theyleverage mobility to communicate and to improve their daily lives,through new and existing services. Users now demand connectivityanywhere and anytime. About this reportImportant driving forces include new affordable smartphones, and Ericsson has performed in-depth datathe many new connected devices on the market. The total number traffic measurements since the earlyof mobile subscriptions globally (excluding M2M) will reach around days of mobile broadband from a9 billion in 2017, of which 5 billion will be for mobile broadband. With large base of live networks coveringan increased number of subscriptions, evolved devices and 24/7 all regions of the world.connectivity to use them, we expect global mobile data traffic to grow15 times by the end of 2017. The aim of this report is to share analysis based on theseAccess to the internet is a prerequisite and will drive further build-out measurements, internal forecastsof mobile networks. By 2017, an astonishing 85 percent of the world’s and other relevant studies to providepopulation will be covered by WCDMA/HSPA networks. insights into the current traffic and market trends.In today’s competitive markets, it is crucial to ensure best userexperience and provide differentiation. A recent consumer study shows We will continue to share traffic andthat network coverage and speed are the most important factors for market data, along with our analysis,satisfaction with mobile services. on a regular basis.We invite you to share our enthusiasm in the insights this data offers. Publisher: Douglas Gilstrap,We hope you find it engaging and valuable. Senior Vice President and Head of Strategy, EricssonTRAFFIC 12 SMARTPHONES 16 STUDY 19Mobile data traffic Coverage and data Video representswill grow 15 times speed drive customer the largest databy 2017. satisfaction. traffic volume. JUNE 2012 TRAFFIC AND MARKET REPORT  3
  • Mobilesubscriptions updateFigure 1: Subscriptions by region, Q1 2012 Net additions 6.2 Billion (millio n) 2.3 5.9 total mobile 9.9 subscriptions Mobile subs 18 cripti ons ( mil li o 3 30 5 40 n) ica Western h Amer Nort Euro pe 0 65 er ica Ce nt Am ra 60 l in 0 & t La Ea s te nEr uro 39 pe 30 6 80 Africa China 1,000 ast dle E 2 60 Mid 6.3 C A AP of e st Ind R ia 60 1,1 90 0 25 170 29 million net additions Source: Ericsson (June 2012) Q1 2012 Figure 1Mobile subscription figures are estimates as of Q1 2012. Mobile net The estimate of mobile net additions has been made based onadditions are estimates during Q1 2012. APAC = Asia Pacific. historic information from external sources and regulatory and operator reports, combined with Ericsson analysis. Historical data may be revised when operators report updated figures.4  TRAFFIC AND MARKET REPORT  JUNE 2012
  • >  lobal mobile penetration reached 87 percent G  >  WCDMA networks deployed worldwide have All in Q1 2012 and mobile subscriptions now total been upgraded with HSPA. Around 75 percent around 6.2 billion. However, the actual number of of the HSPA networks worldwide have been subscribers is around 4.2 billion, since many have upgraded to a peak speed of 7 Mbps or above .2 several subscriptions. and around 40 percent have been upgraded to 21 Mbps.>  India and China accounted for approximately 40 percent of the estimated 170 million net >  Around 15 percent of HSPA networks now have additions during Q1 2012, adding around 25 and speeds up to 42 Mbps in whole or parts of the 40 million subscriptions respectively. network following a wave of upgrades. Today, we are already seeing evolutionary steps towards>  Brazil (+10 million), Indonesia (+9 million), and increasing speeds to well over a 100 Mbps. Bangladesh (+5 million) follow in terms of net additions.>  Mobile subscriptions have grown around Figure 2: Percentage of WCDMA networks upgraded to HSPA and to 7.2, 21 and 42 Mbps respectively 12 percent year-on-year and 3 percent quarter-on-quarter. 100 Source: Ericsson and>  Mobile broadband subscriptions1 have grown GSA (Q1 2012) around 60 percent year-on-year and have reached 80 1.1 billion. 60>  There is continued strong momentum for % smartphone uptake in all regions. Approximately 40 35-40 percent of all mobile phones sold in Q1 were smartphones, compared to around 30 20 percent for the full year 2011. Only around 10-15 percent of the worldwide installed base of 0 HSPA HSPA 7.2 HSPA 21 HSPA 42 subscriptions use smartphones, which means that there is considerable room for further uptake.Figure 3: Mobile subscriptions penetration in Q1 2012 Western Europe 131% Central & Eastern Europe 125% Latin America 109% Middle East 101% North America 93% APAC excluding China & India 91% China 75% India 73% Africa 63% Source: Ericsson (June 2012)1 Mobile broadband is defined as CDMA2000 EV-DO, HSPA, LTE, Mobile WiMAX and TD-SCDMA. Subscriptions vs subscribers PCs/tablets and for mobile phones. In addition, it takes time before inactive subscriptions are There is a large difference between the number removed from operator databases. Consequently, of subscriptions and subscribers. This is due subscription penetration can easily reach above to the fact that many subscribers have several 100 percent, which is the case in many countries subscriptions. Reasons for this could include today. It should however be noted that in some users lowering their traffic cost by using optimized developing regions, it is common for several people subscriptions for different types of calls, maximizing to share one subscription, having for example coverage, having different subscriptions for mobile a family or village phone. JUNE 2012 TRAFFIC AND MARKET REPORT  5
  • SUBSCRIPTIONSOUTLOOKSubscriptions development – Figure 4: Fixed and mobile subscriptions 2008-2017fixed and mobile 9,000 Source: Ericsson (June 2012)By the end of 2011, total mobilesubscriptions reached around 8,0006 billion and are expected to reacharound 9 billion by the end of 2017. 7,000M2M will add to this figure. 6,000The number of mobile broadband Subscriptions/lines (million)subscriptions reached close to1 billion, and is predicted to reach 5,0005 billion in 2017. 4,000 ~5 3,000 Billion 2,000 mobile broadband 1,000 subscriptions by the end of 2017 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017PC and tablet mobile subscriptions Mobile subscriptions Mobile broadband Mobile PCs/tabletsare increasing and are expected Fixed VoIP Fixed broadband Fixed narrowband voiceto grow from around 200 million in2011 and almost close the gap with ‘Mobile broadband subscriptions’ is a sub-segment ofthe number of fixed broadband ‘Mobile subscriptions’. ‘Mobile PCs/tablets’ is a sub-segmentsubscriptions by 2017, totaling of ‘Mobile broadband subscriptions’.around 650 million.The number of fixed PSTN voicesubscriptions will continueits downward trend as users Users per fixed subscriptionincreasingly switch to mobile The number of fixed broadband users is at least three timesand VoIP substitutions. the number of fixed broadband connections, due to multiple usage in households, enterprises and public access spots. This is the opposite to the mobile phone situation, where subscription numbers exceed user numbers. In the latter years of the forecasting period, it is likely that the usage trend for mobile PCs will be similar to fixed broadband usage today, with several users per subscription. This is especially the case in developing markets where mobile access will be the mainMobile PC: laptop or desktop PC source of internet connection.devices with built in 3G modem orexternal USB dongle.6  TRAFFIC AND MARKET REPORT  JUNE 2012
  • Mobile technology Figure 5: Mobile subscriptions by technology, 2008-2017Figure 5 illustrates reported 10,000 Source: Ericsson (June 2012) LTEmobile subscriptions by 9,000 WCDMA/HSPAtechnology. In this graph, GSM/EDGE Reported subscriptions (million) 8,000subscriptions are defined by TD-SCDMA 7,000the most advanced technology CDMAthat the mobile phone and the 6,000 Othernetwork are capable of. 5,000 4,000GSM/EDGE will continue to 3,000lead in terms of subscriptionnumbers until the latter years 2,000of the forecast period, despite 1,000rapid HSPA subscriptions 0growth today. This is 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017because new low-end usersentering networks in growingmarkets will likely use thecheapest mobile phonesand subscriptions available.In addition, it takes time to ~9 Billionupgrade the installed base ofphones. However, the rapid mobilemigration to more advanced subscriptions bytechnologies in developed the end of 2017countries means global GSM/EDGE subscription numberswill decline after 2012. Figure 6: Mobile subscriptions by region, 2008-2017LTE is currently being deployed 10,000 LAand built-out in all regions and Source: Ericsson (June 2012) NA 9,000will be used by around one APAC Reported subscriptions (million)billion subscribers. These will 8,000 CEMArepresent the high-end share 7,000 WEof the total subscriber base 6,000by 2017. 5,000Regional growth 4,000 3,000Figure 6 illustrates mobilesubscriptions in each region 2,000up until the end of 2017 and 1,000is characterized by steady 0growth. This is especially 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017evident in the Asia Pacificregion where there are manydeveloping nations with strongpopulation and GDP growth. JUNE 2012 TRAFFIC AND MARKET REPORT  7
  • Regional technology maturity Figure 7: Mobile subscriptions by technology and region 2011By looking at each region and 100 LTEcomparing subscriptions in each WCDMA/HSPAradio technology, it is easy to GSM/EDGE 80see the different maturity levels TD-SCDMAbetween regions. Less mature CDMAregions are dominated by 2G 60 Othertechnologies, like GSM/EDGE, %while more mature regions like Source: EricssonWestern Europe are dominated 40 (June 2012)by HSPA. In North America,LTE subscriptions have beengrowing since 2011 and the 20region now has the majority ofglobal LTE subscriptions. 0North America is characterized NA WE APAC LA CEMAby early growth in LTE, makingLTE, HSPA and CDMA ofapproximately equal size in2017, with GSM/EDGE nolonger present. The fast growthin LTE subscriptions is drivenby the CDMA operators’ early Figure 8: Mobile subscriptions by technology and region 2017commitment to migrating to the 100 LTEnewer technology. Overall North WCDMA/HSPAAmerican subscription growth is GSM/EDGEbased on multiple subscriptions 80 TD-SCDMAper individual – for example, CDMAadding a tablet – rather thanpopulation growth. 60 Other % Source: EricssonLatin America has a large GSM/ (June 2012) 40EDGE subscriber base, whichwill reach its maximum during2012. The strong growth in 20subscriptions in this region willbe driven by GDP development.In 2017, WCDMA/HSPA will 0be the dominant technology; NA WE APAC LA CEMAhowever GSM/EDGE will stillhave a significant presence.As a mature market, Western Europe will show little plateau in 2012. Markets like Japan and Korea willsubscriptions growth in the years to come. What take up LTE subscriptions very early compared togrowth there is will come from an increasing number of late uptake in less developed countries. China willconnected devices. Subscriptions that have GSM/EDGE add substantial numbers in the latter years. LTEas the highest technology are declining as end users are subscriptions will cover both FDD and TDD.upgrading their phone to more advanced HSPA models.In 2017, LTE is expected to have penetrated around 25 Central and Eastern Europe, the Middle East andpercent of the subscriptions base in Western Europe. Africa (CEMA) shows strong subscriptions growth driven by population and GDP growth. HSPA willThe Asia Pacific market will see a massive increase grow significantly, while LTE will be limited and takein subscriptions. Leading up to 2017, 1.5 billion new place in the latter years. The CEMA region is diverse,subscriptions will be added, driven by GDP and with mature countries showing patterns similar topopulation growth. Almost half of subscriptions will Western Europe, whereas most African countries willbe GSM/EDGE at the end of the period, reaching a have completely different patterns.8  TRAFFIC AND MARKET REPORT  JUNE 2012
  • subscriptionsoutlook: devicesSmartphones Figure 9: Smartphones, PCs and tablets subscriptions with cellular connection, 2008-2017Total smartphone subscriptions 4,000 Mobile PCs Source: Ericsson (June 2012)* and tabletsreached around 700 million in 2011 and 3,500are expected to reach around 3 billion Smartphonesin 2017. Subscriptions (millions) 3,000Ericsson has identified several 2,500smartphone operating systems that, onaverage, are associated with high traffic 2,000volumes. These are referred to as High 1,500Traffic smartphones (HT smartphones).HT smartphones are defined as a 1,000subset of open-OS phones(e.g. iPhone, Android & Windows). 500These devices typically generate5-10 times more traffic than low-traffic 0devices. It is estimated that the HT 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017share reached around 50 percent at the *Smartphone forecast from 2011.end of 2011, and will represent the vastmajority in 2017.Factors such as screen size, year ofdevice release and popularity amonghigh-end users have a stronger ~3 Billioncorrelation with the behavior of activeusers generating more than 1 MB smartphoneof traffic per day on average. They subscriptions bytherefore have a stronger effect on the end of 2017median traffic than the OS.Data-heavy devicesEricsson estimates that the totalsubscriptions of data-heavy deviceswill grow from around 850 million bythe end of 2011 to around 3.8 billionin 2017. This includes smartphones,mobile PCs and tablets with cellularconnectivity. JUNE 2012 TRAFFIC AND MARKET REPORT  9
  • Mobile PC and tablet Figure 10: Mobile PCs and their share of subscriptions 2017subscriptions 100 Wi-Fi/Ethernet PCsIn 2017, around one third of the 3G/4G PC with embeddedinstalled base of mobile PCs modem, no subscription 80is estimated to have a 3G/4G 3G/4G PC with embeddedsubscription: the remaining will use modem, with subscriptionWi-Fi or ethernet. Some mobile 60 3G/4G donglesPCs with 3G/4G capability do not + PC with subscription %have an active mobile subscription. Source: Ericsson (June 2012) 40At the same time, only around halfof tablets are expected to havea 3G/4G modem built in. Some 20of those will not have an activesubscription. Note that in the trafficmeasurements presented later 0in this report, only traffic on3G networks is included, notWi-Fi traffic. Figure 11: Tablets and their share of subscriptions 2017Mobile broadband 100 Wi-Fi tablets 3G/4G tablets withoutThe Ericsson term mobile subscriptionsbroadband is defined by the 80 3G/4G tablets withdevice’s radio standard, and the subscriptionsprerequisite that the same standard Source: Ericsson (June 2012)is available in a live network. This is 60a good measure for understanding %the uptake of new technologiesand it is easy to track. 40Figure 12 illustrates mobile 20broadband subscriptions, splitby region at the end of 2011. Ithighlights the large share held by 0Asia Pacific and North America.Figure 13 shows global mobile Figure 12: Mobile broadband subscriptions by region, 2011broadband subscriptionsdevelopment. The majority of APACdevices are, and will continue NAto be, mobile phones. Mobilebroadband will gain a larger share WEof total broadband subscriptions CEMAin many markets, complementing LAxDSL in certain segments and Source: Ericsson (June 2012)replacing it in others. Mobilebroadband also includes somefeature phones, but this share isdecreasing over time. Mobile broadbandMobile broadband is defined as CDMA2000 EV-DO, HSPA, LTE, Mobile WiMAX and TD-SCDMA. It includes mobile PCs,tablets and mobile phones, both smartphones and feature phones. The vast majority is mobile phones.10  TRAFFIC AND MARKET REPORT  JUNE 2012
  • Figure 13: Mobile broadband subscriptions, 2008-2017 6,000 Source: Ericsson (June 2012) 5,000 4,500 4,000Subscriptions (millions) 3,500 3,000 2,500 2,000 1,500 1,000 500 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 technology development expectations, documents Forecast methodology such as industry analyst reports, on a national or Ericsson performs forecasts on a regular basis regional level, together with internal assumptions to support internal decisions and planning as and analysis. well as market communication. The subscription and traffic forecast baseline in this report are Note that a large part of data traffic is generated based on historical data from various sources, by a limited number of users in each device validated with Ericsson internal data, including category. These users may considerably change extensive measurements in customer networks. their usage if operators implement data volume Future development is estimated based on caps or other traffic management schemes. macroeconomic trends, user trends (researched Measures like this could significantly impact the by Ericsson ConsumerLab), market maturity, traffic forecast. JUNE 2012 TRAFFIC AND MARKET REPORT  11
  • Mobile trafficupdateGlobal traffic in mobile networks Figure 14: Global total traffic in mobile networks, 2007-2012Figure 14 shows the total monthly traffic 800 Source: Ericsson (2012)split for voice and data. It depicts a Voice Datastable trend of data traffic growth with 700some seasonal variations. Mobile data Total (uplink + downlink) monthly traffic (PetaByte/month)subscriptions grow strongly, and drivethe growth in data traffic along with a 600continuous increase in the average datavolumes per subscription. 500Mobile voice traffic continues to grow 400at a steady rate mainly driven by newsubscriptions in Asia Pacific and MiddleEast & Africa. 300It should be noted that there are big 200differences in traffic levels betweenmarkets, regions and operators. 100These measurements have beenperformed by Ericsson over several 0years using a large base of livenetworks that together cover all regions Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 07 07 07 07 08 08 08 08 09 09 09 09 10 10 10 10 11 11 11 11 12of the world. They form a representativebase for calculating world total traffic inmobile networks2. 2x mobile Data traffic almost doubled between Q1 2011 and Q1 2012 >  Data traffic almost doubled  between Q1 2011 and Q1 2012 >  The quarterly growth between Q4  2011 and Q1 2012 was 19%2 Traffic from 2G and 3G – does not include DVB-H, Wi-Fi, and Mobile WiMax.12  TRAFFIC AND MARKET REPORT  JUNE 2012
  • 15xTraffic Mobile datadevelopment traffic will grow ~15 times by the end of 2017Traffic outlook Figure 15: Global mobile traffic: voice and data, 2010-2017Overall mobile data is expected 10,000 Data: Source: Ericsson (June 2012) mobile PCs/tabletsto have almost doubled during Data:2011. Mobile PCs dominate traffic mobile phones 8,000in most mobile networks today, Voice Monthly PetaBytes (1015 B)but smartphone traffic is growingfaster, due to high growth in 6,000subscriptions. In the latter years,data traffic will be split fairly equallybetween mobile phones and 4,000mobile PCs and tablets.Accessing the internet from mobile 2,000devices will drive mobile trafficdevelopment. Mobile data trafficis expected to grow with a CAGR 0of around 60 percent (2011-2017), 2010 2011 2012 2013 2014 2015 2016 2017driven mainly by video. This entailsgrowth of around 15 times by the Figure 16: Smartphone and mobile PC traffic per month and subscription 2011 and 2017end of 2017. 10 Smartphones Source: Ericsson (June 2012)Traffic per subscriber partly relates Mobile PCsto the screen size of the user’sindividual device. On average, a 8mobile PC generates approximately GB/month/subscriptionfour times more traffic than a 6HT smartphone. By the end of2011, an average mobile PCgenerated approximately 2 GB per 4month versus 500 MB per monthproduced by HT smartphones. Anaverage smartphone generates 2around half of the volume of an HTsmartphone. By the end of 2017, itis estimated that a mobile PC will 0generate 8 GB per month, and a 2011 2017smartphone just above 1 GB. Traffic refers to aggregated traffic in mobile access networks. DVB-H and Mobile WiMaxNote that there are large or Wi-Fi traffic have not been included. M2M traffic is not included.differences between user patternson different networks, markets anduser types. See results from livenetwork measurements covering Data revenuesdifferent types of devices, OS and Mobile data revenue continues to increase while exhibitingregions on page 19. considerable variations. On average it represents around 35 percent of mobile operator service revenue. This figure includes SMS revenues, however the main bulk of the increase is generated from data traffic. JUNE 2012 TRAFFIC AND MARKET REPORT  13
  • Regional mobile traffic variations Figure 17: Mobile traffic by region and type, 2011By looking at each region in figure 100 Data: mobile PCs/tablets17 and comparing the traffic Data: mobile phonesgenerated from different device Voice 80types, it is easy to see the diversematurity levels between regions. Source: Ericsson (June 2012)In 2011, regions with high voice 60usage per subscription, such asAsia Pacific and North America, %had a high share of the total traffic. 40North America and Western Europehave a significantly larger shareof total traffic volume than their 20subscription numbers alone wouldimply, due to the high penetrationof 3G/4G networks, as well as that 0of PCs, smartphones and tablets. NA LA WE CEMA APACThe Asia Pacific region has thelargest amount of generated trafficin absolute volumes. Figure 18: Mobile traffic by region, 2011Voice traffic growth will remainat a steady level, whereas data APACtraffic will increase strongly. By NA2017, voice traffic volumes will be WEvery small compared to data traffic CEMAvolumes in all regions. As seen in LAfigure 18 and 19, Asia Pacific isexpected to increase its share of Source: Ericsson (June 2012)global volume from around onethird today to almost 50 percentin 2017. Figure 19: Mobile traffic by region, 2017 20x APAC NA WE Mobile data traffic CEMA for smartphones LA will grow ~20 times by the end of 2017 Source: Ericsson (June 2012)14   TRAFFIC AND MARKET REPORT  JUNE 2012
  • 85%Coverage WCDMA/HSPA population coverage in 2017Coverage of the world’s mobile Figure 20: Technology coverage, 2011 and 2017networks is constantly increasing as 100more base stations are deployed. GSM/ >90% Source: Ericsson (2012)EDGE technology has by far the widest >85% 85%reach and today covers more than 80 Rural85 percent of the world’s population. World population distribution % population coverageGeographically, only rural areas remainto be covered by GSM/EDGE. 60 55% 50% >50% >45% Sub-WCDMA/HSPA covered over urban45 percent of the population by the 40end of 2011, but is now accessible by50 percent. There are more densely 20 Urbanpopulated areas lacking WCDMA/HSPAcoverage. Further build-out of WCDMA/ 5% MetroHSPA coverage will be driven by the 0availability of affordable smartphones 2011 2017 2011 2017 2011 2017 2011 2017and the surge in mobile broadbandservices and faster speeds, as well as GSM/EDGE WCDMA/HSPA LTE CDMAregulator requirements to connect theunconnected. By 2017 an estimated 85 , Figure 21: Traffic generation per area, 2017 Source: Ericsson (2012)percent of the world’s population will 100 Ruralhave the opportunity to access Suburbanthe internet using WCDMA/HSPA 80 Urbannetworks3. Metro % populationToday, the combined 2G and 3G 60population coverage for CDMA isestimated to be above 50 percent. 40CDMA coverage is expected to growslightly since most large CDMA 20operators have announced a migrationplan to LTE. 0 Estimated 2017 EstimatedLTE rollout world population distribution 2017 trafficSeveral major operators have * Metro: > 4,000 people/sq km Urban: 1,000-4,000 people/sq km Suburban: 300-1,000 people/sq km Rural: < 300 people/sq kmstarted LTE deployments, but interms of population coverage thereis a long way to go. In February, Effects of urbanizationLTE was estimated to cover 325 Urbanization is a major global trend. Figure 21 shows that by 2017, overmillion people globally. In five 30 percent of the world’s population are expected to live in metro andyears, it is expected that LTE will urban areas. These areas represent less than 1 percent of the Earth’scover around 50 percent of the total land area, yet are set to generate around 60 percent of mobile trafficpopulation. by 2017. In these areas, heterogeneous networks will complement macro network3 The figure refers to population coverage improvements, serving the traffic and providing good coverage and highof each technology. Other factors, such asaccess to devices, are also needed in order quality user experience. In less-densely populated areas the focus will beto utilize the technology. on building cost-effective coverage and capacity. JUNE 2012 TRAFFIC AND MARKET REPORT  15
  • Smartphone users:Network qualityThe growth of smartphones has been tremendous.40 percent of the world’s smartphone users accessinternet and apps even before getting out of bed. Study backgroundOnce out of bed, internet and apps are used almost Ericsson’s ConsumerLab has performed aconstantly, peaking during the daily commute with smartphone user network satisfaction survey70 percent usage. in two mature mobile broadband markets with different characteristics:Smartphone users are developing new internet the Netherlands and Finland.access habits at a tremendous speed in manynew localities and we have studied their evolving The survey was administered online and theperception of network quality. 51 percent of sample size was 1,000 3G smartphone userssmartphone users are very satisfied with their aged 18-69 per country who useoperator’s network and only 3 percent are outright apps or access the internet more thandissatisfied, leaving almost half in a position where once a week.their satisfaction could easily be improved. 47% Figure 22: Experience of smartphone usage problems 40 Source: Ericsson perceive a slow network ConsumerLab (2012) to be the most common 35 cause of smartphone internet usage issues 30 25There is intensive smartphone use during 20 %commuting, yet at the same time, features such assubways and tunnels create obstacles to internet 15access. It may come as no surprise therefore, thataround half of all smartphone users experience 10internet access issues on a weekly basis, accordingto our recent Ericsson ConsumerLab study. 5 0When issues do occur they are perceived to beirritating, and the frequency of the issues is in itself Daily Weekly Less often Nevera cause of frustration for some. Medium to highnetwork satisfaction levels indicate that users acceptsome issues today, but also expect improvements inthe near future.16   TRAFFIC AND MARKET REPORT  JUNE 2012
  • User perceptions Figure 23: Responsibility for internet access problemsEricsson’s new smartphone user network satisfaction 50 Netherlandsstudy in Finland and the Netherlands indicates that Finlandas users are maturing, they have stronger opinions 40on what causes different issues. There is therefore a Source: Ericssongreater focus on improving factors which affect the ConsumerLab (2012)user experience. 30A slow mobile network is perceived to be the most %common cause to smartphone usage issues, stated 20by 47 percent of the sample, or more specific percountry, 56 percent in Finland and 37 percent in theNetherlands. 10Users also perceive the network to be the maincontributor to internet access problems. One third of 0Finnish smartphone users and 47 percent of Dutch Carrier Application Handset Content Don’t know provider manufacturer providersmartphone users hold the carrier responsible forinternet connectivity problems.Coverage and data speed drive satisfaction A competitive advantage canCoverage where people need it and a fast and be generated by improvingreliable connection to the internet are “must haves”for smartphone users and are the strongest drivers consumer network satisfactionof satisfaction. This reflects the need to constantlybe connected and to have a positive smartphone Figure 24:  orrelation between satisfaction with features Cexperience. and overall satisfactionProviding an excellent network with great coverage 0.6 Netherlandsand data speed where apps run smoothly, is Finlandessential to satisfying users and standing out from 0.5the competition. Source: Ericsson ConsumerLab (2012) 0.4Satisfied users make recommendationsWhereas half of all users are very satisfied with 0.3their carriers, there is an opportunity to generatecompetitive advantage by improving consumernetwork satisfaction for users who experience 0.2internet access issues. If problems occur morethan once a week, then overall satisfaction with the 0.1network significantly declines. Analysis shows thathighly satisfied subscribers will generate good word 0of mouth, so this may be a worthwhile opportunity. 3G mobile High Price Clarity Data Content network speed 3G plan of voice allowance and apps coverage mobile calls available network JUNE 2012 TRAFFIC AND MARKET REPORT  17
  • Accelerating Accelerated content loads up tomobile content 70%fasterdeliveryAs content delivery over mobile networks is rapidly One way of enhancing the user experience is togrowing, it is becoming increasingly important to accelerate content delivery through a combinationsecure quality of experience end-to-end, from the of internet optimization and mobile prioritization.content server across the fixed internet, and themobile network to the mobile device. The effect of this is threefold; it provides shorter average page load times, a more consistent userCoverage and speed are the biggest drivers of experience and page load failures are largelynetwork satisfaction (see page 17). Improving avoided. Being able to provide a consistently goodthese will have the most positive impact on user user experience has a direct effect on a contentsatisfaction and improve both the operator and the provider’s business value through increasedcontent provider’s business. For a content provider, a conversion rates and brand perception.one second delay in page load time can result in lostconversions4, fewer page views and a decrease in Figure 25 shows measurements from a commercialcustomer satisfaction. mobile broadband (HSPA) network. The dark blue line represents page load times for content downloaded from a content provider’s web server (origin) to a mobile device, and the light blue line shows page load times for the same content when the delivery isContent Device server accelerated end-to-end. Page load times for non-accelerated content varies greatly over the five day period due to variation in radio network load, while the accelerated content is less affected. Page Internet acceleration Mobile prioritization load times are, for the accelerated content, shorter on average and a more consistent experience is provided. In this measurement, the accelerated QoE measured and secured end-to-end content downloads up to 70 percent faster. Quality of Experience is secured end-to-end through a combination of content acceleration through the fixed 4 The conversion rate is the ratio of visitors who convert internet and content prioritization in the mobile network. website visits into desired actions e.g. purchase of product.Figure 25: Page load time with/without end-to-end acceleration 60 Source: Ericsson & Akamai (2012) –– With Mobile Cloud Accelerator 50 –– Without Mobile Cloud Accelerator 40 Page load time 30 20 10 0 Day 1 Day 2 Day 3 Day 4 Day 518   TRAFFIC AND MARKET REPORT  JUNE 2012
  • Traffic variationsbetween networks Traffic pattern analysis New devices and applications affect mobile networks. Having a deep and up-to-date knowledge of the traffic characteristics of different devices and applications is important when designing, testing and managing mobile networks. Ericsson regularly performs detailed traffic measurements in all major regions of the world. The measurements in this section were made in a selected number of live commercial WCDMA/HSPA networks in Asia, Europe and the Americas.Variations observed in mobile traffic patterns are >  The overall quality of the network, speeddependent on many, often interacting, factors. and latency.Some of the most significant ones are: >  Price plans and penetration of fixed broadband>  The data plan (price and allowance) affects user networks. When the primary access to the internet behavior. Low prices drive up user numbers and is via mobile access it drives usage upwards. a larger data allowance typically drives higher volume per user. Ericsson traffic>  Video often represents the largest share of total measurements show that traffic. The availability of video content in the respective market has a major impact on traffic there is no typical network patterns in all networks. Common examples In the coming sections traffic characteristics are are Video On Demand services, such as Netflix examined one factor at a time. These measurements and Hulu, as well as a multitude of TV channels clearly show that there are major differences between (typically local to the market) available free on the networks depending on these factors. the internet.>  Device type, screen size and resolution.>  raffic management – this is regulated by each T country’s authorities, however, it also represents a strategic decision for the operator. For example, operators can manage P2P file sharing in different ways by either allowing it, forcing it to low traffic hours, or not allowing it. Traffic variation figures Largest average value, measured in one of the networksThe results show that there is a big difference between the individual Average value of the measured networks’ average valuesnetworks measured in this chapter. The graphs therefore reflect thisbroad spread of data by showing the average values of the differentmeasured networks. Smallest average value, measured in one of the networks JUNE 2012 TRAFFIC AND MARKET REPORT  19
  • Traffic variations – devices Figure 26:  enetration and share of total mobile traffic volume P per device in different networks, measured in bytesThere are major differences in howmuch traffic various devices generate in 100different mobile networks. Source: Ericsson (2011, 2012) Device penetration 80Figure 26 shows penetration and share Trafficof total traffic volume for different volume sharesdevice types (see legend at the 60bottom of the page for information %on how to read the graph). In some 40European networks, traffic frommobile PCs dominate, while in North 20America smartphone traffic is typicallypredominant. The penetration of 0mobile PCs differs between around 3G router Mobile PC Tablet Mobile phone M2M1-16 percent in the measurednetworks. Mobile PCs users createapproximately 8-88 percent of totaltraffic in the measured networks. One Figure 27:  verage monthly mobile traffic volume per A subscription and device in different networksreason for the variation is the differentfocus operators have had on PC andsmartphone segments. Operators that 18 Source: Ericsson (2011, 2012)launched mobile broadband early, 16typically successful in selling dongle 14subscriptions, often have a large shareof PC subscribers. In some markets 12it is still the dominant segment due to GB/month 10late proliferation of smartphones and ageneral strategy to target households 8with DSL replacement offerings. 6Figure 27 shows the spread of average 4monthly data traffic per subscription 2and device type in the different 0measured networks. The largest spread 3G router Mobile PC Tablet Mobile phone M2Mcan be observed in 3G routers (1-16 GBper month). For PCs, the variation isalso large (1-7 GB per month) followed 3G router: WLAN router with built-in HSPA Tablet: portable tablet computers withby tablets (300-1600 MB per month) and uplink interface rather than a connection to touch screen display (e.g. iPad, Galaxy Tab). a fixed network. Mobile phone: any mobile phone, includingmobile phones (30-230 MB per month). Mobile PC: laptop or desktop PC both smartphones and feature phones. devices with built in 3G modem or M2M: machine to machine devicesThe observed M2M traffic is very external USB dongle. (e.g. vehicle tracking, fleet management,small, with an average volume security surveillance, remote monitoring).below 10 MB per subscription in allmeasured networks. The M2M share Largest average value, measured in one of the networksof mobile networks often include lowdata applications such as security Average value of the measured networks’ average valuessurveillance, fleet management, andPoint Of Sale terminals. Smallest average value, measured in one of the networks20  TRAFFIC AND MARKET REPORT  JUNE 2012
  • Traffic variations – mobile phone Figure 28:  verage monthly traffic volume per subscription for different mobile A phone operating systems in different networksoperating systemsFigure 28 shows the spread in average 1,600 Source: Ericsson (2011, 2012)traffic volume per subscription between 1,400measured networks. iPhone5 and 1,200 MB/month/subscriptionAndroid smartphones represent thelargest traffic volumes per subscription. 1,000One reason for the wide spread is 800the difference in data plans offeredto the users (see page 23). Android 600models have a greater variance due 400to a larger diversity of device models.In networks where high-end models 200dominate, average usage on Android 0devices can exceed average iPhone iPhone Android Windows Symbian Blackberry Feature smartphone smartphone smartphone smartphone phoneusage. However, when operators focuson the low-end Android segment theaverage usage is usually smaller thanfor iPhones. Figure 29:  enetration and share of total mobile phone traffic volume for P different operating systems in different networks, measured in bytesThere are large differences between 100traffic volume of phones with Source: Ericsson (2011, 2012) Operatingdifferent versions of Windows OS, System 80 penetrationhence average values might increase Trafficsignificantly in the future when Windows volume 608 becomes dominant. shares %Figure 29 shows penetration and 40share of total traffic volume for thedifferent mobile phone OS types. 20iPhones represent, on average, nearly50 percent of the total mobile phone 0traffic in the measured networks. The iPhone Android Windows Symbian Blackberry Featurereason is the relatively high average smartphone smartphone smartphone smartphone phoneusage per subscription coupled withhigh penetration. The variance betweenmeasured networks is high, rangingfrom a few percent to nearly 80 percentof the total traffic volume for iPhones.5 The iPhone OS is called iOS. Largest average value, measured in one of the networks Average value of the measured networks’ average values Smallest average value, measured in one of the networks JUNE 2012 TRAFFIC AND MARKET REPORT  21
  • Traffic variations – applications There are also large differences between networks and regions regarding P2P file sharing. ExplanationsFigure 30 shows average monthly traffic volume include cultural differences, availability of otherper subscriber for mobile PC devices and how it content distribution types and legal constraints.varies between the measured operators in Asia and Differences between data plans and trafficEurope. HTTP video represents the largest volume management functions implemented by operatorsper subscriber. There are differences in how HTTP can have an effect.video applications behave and load the networks.Applications like Hulu and Netflix use adaptive bit Video represents the largestrates, which means they use as much bandwidth aspossible in order to maintain the highest possible data traffic volumequality at any given moment. YouTube and some P2P file sharing is typically used by rather smallsimilar applications use the bit rate that the user groups of users, but can account for a large share ofselects, starting by default at the lowest possible the total traffic volume. These measurements werebit rate. made in mobile networks, but P2P file sharing is more prevalent on fixed networks, where speeds areThe biggest difference between Europe and Asia normally higher and data plans less restrictive.is in the usage of peer-to-peer (P2P) TV, which hasbecome a common form of TV distribution in someAsian countries.Figure 30: Average monthly mobile application traffic volume per subscription  for mobile PC, in different networks 2,000 Source: Ericsson (2011, 2012) Europe Asia 1,500MB/month/subscription 1,000 500 0 Web browsing HTTP video P2P TV P2P file sharing Largest average value, measured in one of the networks Average value of the measured networks’ average values Smallest average value, measured in one of the networks Video definitionBoth HTTP video and P2P TV are types of online video where the user watches the film while it is being downloaded (streamed). In the case ofP2P file sharing the user first downloads the entire file, most often a movie, and watches it offline. Examples of applications that use HTTP areYouTube, Hulu, Netflix, and BBC iPlayer. An example of P2P TV is PPStream.22  TRAFFIC AND MARKET REPORT  JUNE 2012
  • Traffic variations: data plan As described in the previous section, there are What is often overlooked is that a high volume user several factors which have an impact on traffic will tend to switch to the newest device model upon patterns. Data plans are one of the most important. its release. Therefore this increase in usage can be This section focuses on the measureable effects attributed to both the new device and the subscriber relating to data plans in a selected mature segment it has attracted. 3G network. It is not uncommon for average usage to decrease Variations by phone model and data plan over time for any given device model, as proliferation in the market increases and more regular users Figure 31 shows how the average monthly volume transfer to the device. Usage patterns are also per subscription varies by data plan as well as by related to the data plan that comes with a device. mobile phone model for a specific brand. Figure 31 shows that the traffic generated from an existing mobile phone model can be higher than When a new device is released, there are sometimes that of a brand new one, due to differences in the dramatic headlines in the media focusing on the subscriber’s data plan. The user’s choice of specific enormous increase in usage that some new devices phone model can also be dependent on the voice seem to generate. plan. As a result, the proliferation of different devices is also dependent on the voice plan connected to the device at the point of sale. The data plan is one of the most important factors determining level of traffic usage Figure 31: Average monthly traffic volume per data plan  per smartphone model, from the same vendor 800 500 MB Source: Ericsson (2011, 2012) 2 GB data plan 700 1 GB 600 2 GB 1 GB data plan data planMB/month/subscription 500 500 MB 400 300 200 100 0 Model A Model B Model C Model D released released released released 2008 2009 2010 2011 JUNE 2012 TRAFFIC AND MARKET REPORT  23
  • Traffic variations by application and data plan Both web browsing and P2P file sharing usage remain fairly constant within the same data plan Figure 32 shows the variation of a subscriber’s between Q1 and Q4, while online video has increased average monthly traffic volume for different dramatically. The explanation for this can be found applications for mobile PC devices. It is measured by looking at time spent online against the amount at two different occasions during 2011 in one of data generated in a specific time period. There are selected network. limitations on the time an individual can spend on each application. With low bandwidth applications, Video traffic is still increasing such as web browsing, this limitation quickly becomes visible within the package data allowance. As an example, a user on a data plan of around 15 GB consumed on average 2.3 GB per month of Therefore, we observe a saturation of traffic from that online video in Q1, while in Q4 this figure was type of application. Video has yet to reach its peak. 3.5 GB per month. An example of this is online video with higher bit rates becoming available from YouTube. Another reason The monthly subscriber traffic from applications is that improved video quality, and increased screen such as online video and web browsing both reach sizes facilitate looking at longer clips or watching for a saturation points, although at different levels. At the longer time. The increasing availability of content also same time, P2P file sharing keeps increasing with the has an impact. monthly data plan allowance. Figure 32: Average monthly traffic volume per application, per subscription, as a function of data plan 5 Source: Ericsson (2011, 2012) 4GB/month/subscription 3 2 1 0 0 5 10 15 20 25 User monthly data plan (GB) HTTP video (2011 Q1) Web browsing (2011 Q1) P2P file sharing (2011 Q1) HTTP video (2011 Q4) Web browsing (2011 Q4) P2P file sharing (2011 Q4) In these measurements, the majority of users are on medium data plans between 2-5 GB/month. In Q4 higher data plans were available. 24  TRAFFIC AND MARKET REPORT  JUNE 2012
  • In co-operation withThe signatureof humanity The top 9 categories in descending order are video playback, web browsing, file sharing, file download, INTRODUCING MIT social networking, media playback, software update, The MIT Senseable City Lab studies audio playback and email. the relationship between cities, new technologies and people. The Lab draws From the 12 datasets provided, 8 were selected that on diverse fields to deliver research and spanned at least 7 consecutive days, and displayed applications that empower citizens to make the aggregated traffic consumption from this period. choices and create a more liveable urban condition. Carlo Ratti is the Director of the Each graph follows a classic stacked area chart. Senseable City Lab, Pedro Cruz PhD Visiting The height aligns with symmetrical categories; other Student and Prudence Robinson Partner applications stack alternatively above or below (figure Strategist and Research Fellow. A). This primitive visualization style is adequate for this concrete type of data since it visually depicts a flow, resulting in an almost asymmetrical form. A different color is assigned to each category with the following stacking order shown in figure B.Pedro Cruz, Prudence Robinson, Carlo Ratti Figure B: Application stacking orderGlobal data from cell phone networks providesunprecedented glimpses into the fabric of societyand pulse of our planet. The MIT Senseable CityLab received 12 anonymous datasets from Ericsson Audio playbackcorresponding to 12 networks from 3 different Media playbackcontinents and consisting of 33 GB of data. These File downloadsets refer to data traffic over the networks andconsist of connection time per user, amount of traffic Web browsingconsumed during that period and, where available, Video playbackthe software used and the type of activity. File sharingThe Senseable City Lab studied how data is Social networkingconsumed over these networks, while establishing Software updatea time-based narrative. Additionally, the inclusion Emailof the activity type shows how the traffic is used.Figure A: Single operator in Asia SAT SUN MON TUE WED THU FRI JUNE 2012 TRAFFIC AND MARKET REPORT  25
  • In co-operation withThe noon and midnight of each day are Figure C: Traffic consumption profiles of Northern Europemarked with a sun and moon respectively, (top) and South America (bottom)as well as the corresponding week day. WED THU FRI SAT SUN MON TUEEach vertical line marks one hour as seenin figures C and D.Figure C has highly distinguishable featureswhere the differences between the maximaand minima hours are clearly noticeable. Itis also interesting to observe how the maincategories of traffic oscillate; video playback THU FRI SAT SUN MON TUE WEDand web browsing are good examplesof this. Software updates can generateconsiderable traffic at night. The trafficprofiles do not undergo major changesduring weekends.The graphs consist of dots of varyingbrightness. Each dot represents 200 users,consuming traffic of a certain category Figure D: Close-up of a specific network showing differences in texture densityat that moment. This approach enables acomparison to be made between the ratio of SAT SUNconnected users and the traffic consumedper category over time. This is clearly shownin figure D.It can be argued that the density of webbrowsing is higher than the density of filesharing which is higher than the density ofvideo playback. This means, for example,that a user consuming video playbackconsumes more in quantity than a webbrowsing user.The final artifact visualizes the consumptionpatterns of all 8 networks combined. Thescale has been modified in order to maintainlegibility, each dot represents 800 usersand the vertical consumption scale has Figure E: Stacked area chart of collated networksbeen quadrupled. When the distinctiveconsumption patterns of each artifact aremerged, their differences dissolve andcreate a cyclical pattern that demonstratesan almost constant periodicity in time andvolume: the signature of humanity.26  TRAFFIC AND MARKET REPORT  JUNE 2012
  • Key figuresMobile subscription essentials 2011 2012 2017 CAGR 2011-2017 UnitWorldwide mobile subscriptions 6,000 6,700 8,900 7% – Smartphone subscriptions 700 1,000 3,100 30% – HT smartphones 350 millions – Mobile PC & tablet subscriptions 200 250 650 25% – Mobile broadband subscriptions 1,000 1,400 5,100 30% Mobile traffic essentials 2011 2012 2017 CAGR 2011-2017 Unit – Monthly traffic/smartphone 250 350 1,100 30% – Monthly traffic/HT smartphone 450 – Monthly traffic/PC 2,000 2,500 8,000 25% MB/month – Monthly traffic/tablet 650 850 3,200 30% Monthly traffic/fixed broadband connection 35,000 50,000 140,000 25% Traffic growth Multiplier 2011-2017 CAGR 2011-2017 All mobile data 15 60% – Smartphones 20 65% – PC 10 50% – Tablets 75 100% Fixed broadband 5 30% Glossary2G: 2nd generation mobile networks M2M: Machine-to-Machine3G: 3rd generation mobile networks MB: MegaByteAPAC: Asia Pacific Mbps: Megabits per secondCAGR: Compound Annual Growth Rate Mobile PC: (See page 6)CDMA: Code Division Multiple Access NA: North AmericaCEMA: Central and Eastern Europe, Middle East and Africa OS: Operating SystemDSL: Digital Subscriber Line P2P: Peer-to-PeerEDGE: Enhanced Data Rates for GSM Evolution PetaByte: 1015 BytesGB: GigaByte PSTN: Public Switched Telephone NetworkGDP: Gross Domestic Product TD-SCDMA: Time Division Synchronous Code Division Multiple AccessGSM: Global System for Mobile Communications VoIP: Voice over IP (Internet Protocol)HSPA: High Speed Packet Access WCDMA: Wideband Code Division Multiple AccessHT: High Traffic WE: Western EuropeLA: Latin America xDSL: Various technologies for DSLLTE: Long-Term Evolution JUNE 2012 TRAFFIC AND MARKET REPORT  27
  • Ericsson is the world’s leading provider of communicationstechnology and services. We are enabling the Networked Societywith efficient real-time solutions that allow us all to study, work andlive our lives more freely, in sustainable societies around the world.Our offering comprises services, software and infrastructure withinInformation and Communications Technology for telecom operatorsand other industries. Today more than 40 percent of the world’smobile traffic goes through Ericsson networks and we supportcustomers’ networks servicing more than 2.5 billion subscribers.We operate in 180 countries and employ more than 100,000people. Founded in 1876, Ericsson is headquartered in Stockholm,Sweden. In 2011 the company’s net sales were SEK 226.9 billion(USD 35.0 billion). Ericsson is listed on NASDAQ OMX, Stockholmand NASDAQ, New York stock exchanges.The content of this document is subject to revision withoutnotice due to continued progress in methodology, design andmanufacturing. Ericsson shall have no liability for any error ordamage of any kind resulting from the use of this document.EricssonSE-126 25 Stockholm, SwedenTelephone +46 10 719 00 00 198/287 01-FGB 101 220Fax +46 8 18 40 85 Revision Bwww.ericsson.com © Ericsson AB 2012