ARCHIVES OF ACOUSTICS

Vol. 45, No. 2, pp. 321–328 (2020)

DOI: 10.24425/aoa.2020.133152

Research Paper

On the Consumption of Multimedia Content Using Mobile Devices:

a Year to Year User Case Study

Przemysław FALKOWSKI-GILSKI
Gdańsk University of Technology
Faculty of Electronics, Telecommunications and Informatics
Narutowicza 11/12, 80-233 Gdańsk, Poland; e-mail: [email protected]
(received June 5, 2019; accepted February 13, 2020 )

In the early days, consumption of multimedia content related with audio signals was only possible
in a stationary manner. The music player was located at home, with a necessary physical drive. An
alternative way for an individual was to attend a live performance at a concert hall or host a private
concert at home. To sum up, audio-visual effects were only reserved for a narrow group of recipients.
Today, thanks to portable players, vision and sound is at last available for everyone. Finally, thanks to
multimedia streaming platforms, every music piece or video, e.g. from one’s favourite artist or band, can
be viewed anytime and everywhere. The background or status of an individual is no longer an issue.
Each person who is connected to the global network can have access to the same resources. This paper is
focused on the consumption of multimedia content using mobile devices. It describes a year to year user
case study carried out between 2015 and 2019, and describes the development of current trends related
with the expectations of modern users. The goal of this study is to aid policymakers, as well as providers,
when it comes to designing and evaluating systems and services.
Keywords: audio coding; broadcasting; mobile devices; multimedia; signal processing; streaming services.

1. Introduction 2. Mobile streaming services

Nowadays, due to the presence of numerous multi- Over the last decades, the music industry contin-
media streaming services, e.g. YouTube, YouTube Mu- ues to adapt to constant changes in technology. The
sic, Apple Music, Google Play Music, Netflix, Spotify, breakthrough came in 2017, when streaming and down-
Twitch, etc., content streaming is a crucial application. loading revenue outweighed physical music sales, such
It should be also noted that social media, such as Face- as CDs and vinyl. Mobile streaming services are not
book, can also be used for content distribution. Their uniform themselves. They are comprised by stream-
success, together with web browsers, is the focus of at- ing sociomusical platforms (e.g. Spotify), musical so-
tention of numerous researchers interested in a variety cial systems (e.g. last.fm), music distribution services
of aspects, ranging from energy optimisation and net- (e.g. Soundcloud), as well as millions of users that gen-
work planning to recommendation systems. The main erate income, both from subscription and advertise-
reason is the process of streaming itself, performed us- ment.
ing wireless networks, since current mobile devices use What is worth mentioning, streaming takes place
a lot of power for constant decoding of multimedia at the expense of downloading music from the Inter-
content in order to present them via speaker or dis- net. Surprisingly, it was the leading driver of revenue
play. Rich content distribution among handheld de- for the U.S. music industry, whereas downloading al-
vices, such as smartphones and tables, is becoming bums, compared to 2016, decreased by approx. 15%.
more and more popular every year. With the outcome On the other hand, vinyl sales increased by 20% com-
of online streaming services, including VoD (Video-on- pared to 2016, which accounts for 10% of all physical
Demand) and other cloud based solutions, numerous media sales (Christman, 2017). The list of most popu-
service providers sometimes face bottlenecks, resulting lar terrestrial broadcasting as well as online streaming
in stalling or buffering. This paper describes a survey services, along with primary information concerning
carried out over a period of 5 years. utilised codec and bitrate, is described in Table 1.
322 Archives of Acoustics – Volume 45, Number 2, 2020

Table 1. Popular terrestrial broadcasting and online streaming services.

Type Service Codec Bitrate [kbps]
Terrestrial DAB (Digital Audio Broadcasting) MP2 64–192
broadcasting
DAB+ (Digital Audio Broadcasting plus), HE-AAC v2 48–128
DMB (Digital Multimedia Broadcasting),
DRM (Digital Radio Mondiale),
DRM+ (Digital Radio Mondiale plus)
Online Amazon Music FLAC, MP3 up to 256
streaming
Apple Music AAC-LC 128–256
Deezer HE-AAC v2, 24–128 (HE-AAC v2)
MP3 64–320 (MP3)
Google Play Music MP3 64–320
Netflix H.264 Audio: up to 768
Audio-Video: up to approx. 16 000
Pandora HE-AAC up to 192
Spotify Ogg Vorbis 96–320
Tidal MQA up to 1411
Twitch AAC-LC, Audio: up to 160, recommended 96
H.264 (AAC-LC, H.264)
Audio-Video: up to approx. 8000
(H.264)
YouTube AAC-LC, Audio: up to 192 (AAC-LC)
H.264 Audio-Video: up to approx. 16 000
(H.264)
YouTube Music HE-AAC v2 48–256
Web streaming (various) AAC-LC, 32–320 (AAC-LC, Opus)
HE-AAC v2, 24–128 (HE-AAC v2)
MP3, 64–320 (MP3, Ogg Vorbis)
Ogg Vorbis,
Opus

One must note that in the case of the majority of different layers of the Internet protocol stack, at dif-
online streaming platforms, especially those focused on ferent endpoints in client-server communication, etc.
audio content distribution, the codec and bitrate, be- These solutions, applicable to commercial consumer
ing closely linked with end user perceived quality, are mobile devices, are most often limited to networking
strictly dependent on the type of service. Most often, technologies, such as Wi-Fi, 3G, and LTE. Some of
the highest quality is reserved only for those consumers them propose application layer techniques using SVC
who pay a monthly premium subscription fee. When it (Scalable Video Coding), transcoding and content se-
comes to terrestrial broadcasting services, e.g. DAB+ lection. These solutions differ from the others, since
digital radio, quality remains the same for each con- they modify the actual multimedia content to reduce
sumer. In the case of Poland, the bitrate of audio ser- energy consumption of the mobile client (Hoque et al.,
vices ranges from 64 to 128 kbps, whereas e.g. in the 2014).
Czech Republic it ranges from 48 to 80 kbps (Zyka, In (Vallina-Rodriguez, Crowcroft, 2012),
2019). At the same time, additional value-added data the authors look at smartphone energy management
services are available at 16 kbps. It is also easy to no- techniques from the following perspectives:
tice how online multimedia streaming can affect the 1) energy-aware operating systems,
battery life of a mobile device. 2) efficient resource management,
3) impact of users’ interaction patterns with mobile
2.1. Energy consumption
devices and applications,
During the last years, a wide range of solutions has 4) wireless interfaces and sensors management,
been proposed to optimise energy consumption of mul- 5) benefits of integrating mobile devices with cloud
timedia streaming clients. They include operation at computing services.
 P. Falkowski-Gilski – On the Consumption of Multimedia Content Using Mobile Devices. . . 323

Of course, some factors can be linked with wireless out. This user-oriented study was performed over a five
communication aspect, particularly issues related with year time period, from 2015 to 2019, on a group of 50
network planning and network monitoring. people each year, respectively. Those individuals were
all students of the Gdańsk University of Technology,
2.2. Network planning aged between 19 and 25. It was interesting to learn
how their expectations changed over the time, with
In (Havinga, Smit, 2001), the authors focus the outcome of numerous portable devices as well as
on MAC (Medium Access Control) layer solutions streaming services available online, which undoubtedly
and energy-efficient error control techniques. Exten- had an impact on today’s digital society.
sive power aware mobile multimedia was surveyed by The study was divided into 4 groups, namely:
(Zhang et al., 2009), where the authors investigated
adaptive technologies for video coding and transmis- 1) consumption of multimedia content;
sion. 2) consumer devices;
In (Cao et al., 2004) the authors provided an 3) processing and storing content;
overview of network-aware applications for mobile mul-
timedia delivery. However, they excluded energy-aware 4) Internet connection.
multimedia delivery techniques. Software strategies The survey consisted of closed and opened ques-
that are applicable to portable computer energy man- tions, in order to ensure the best possible feedback
agement were surveyed in (Lorch, Smith, 1998). The from each individual and freedom of expression.
study covers all components of a portable device, in-
cluding wireless interfaces. 3.1. Consumption of multimedia content
Another paper by (Kennedy et al., 2012) also ad-
dressed energy consumption of different components The first part was devoted to the consumption of
of a mobile device during multimedia streaming. As multimedia content. Figure 1 shows the percentage
far as the networking interface is concerned, the au- of users consuming this type of content, whereas the
thors mostly focused on link layer solutions, as well popularity of online streaming platforms is shown in
as cross layer multimedia delivery mechanisms. While Fig. 2.
most studies analyse multimedia streaming in the regu-
lar client-server architecture, a survey on the research
on QoS (Quality of Service) for peer-to-peer media dis-
tribution was presented in (Xiong et al., 2011). How-
ever, that survey did not discuss energy consumption
required for multimedia streaming.
In (Počta, Beerends, 2015), the authors investi-
gate the perceived quality of current audio-based ser-
vices; (Uhl et al., 2017) and (Brachmański, 2018)
describe quality evaluations of speech signals. Biases
encountered in modern audio listening tests are dis-
cussed in (Zieliński, 2016). In (Leszczuk et al., 2013) Fig. 1. Consumption of multimedia content.
and (Uhl, Paulsen, 2014), both QoS and QoE (Qual-
ity of Experience) aspects related to video stream-
ing services are investigated. Additional information
on multimedia broadcasting and multicasting, partic-
ularly in mobile networks, may be found in (Iwacz
et al., 2008).
Another paper, by (Gilski, Stefański, 2016),
talks about the possibilities, limitations, and user ex-
pectations related to analogue and digital services.
After an extensive literature review, no surveys were
found on the topic of consumption of multimedia con- Fig. 2. Usage of streaming platforms.
tent using mobile devices. That is why this year to year
case study was carried out. As it can be seen, the majority of users are in-
terested in consumption of multimedia content. The
3. Survey percentage of active users exceeds 70%, with an ex-
ception in 2017, where a slight decrease was observed.
In order to evaluate current trends in the multi- The main cause for choosing streaming platforms, and
media content consumption, a survey had been carried not traditional terrestrial networks, including digital
324 Archives of Acoustics – Volume 45, Number 2, 2020

terrestrial television DVB-T, as well as terrestrial ana-

logue FM and digital DAB+ radio (Gilski, 2017a),
was the need for a broader programme offer.
An overall definition of online streaming services
is that they are created by separate components, in-
cluding songs, albums, metadata, etc., in a way that
helps to meet the user’s needs and expectations. After
a closer look on the characteristics and functionality of
these services, two categories can be distinguished:
1) traditional streaming services – high availability
Fig. 4. Most popular streaming platforms.
of music pieces, radio stations, music recommen-
dation, along with the ability to add friends and
watch them, create and share playlists, subscrip- cluding: Android, iOS, Windows, and BlackBerry. Spo-
tion fees, etc.; tify operates on two layers: general map of relation-
2) sociomusical services – webpages of artists, al- ships between songs and personalisation layer.
bums, songs, events, with the ability to track mu- The first one analyses all playlists of all Spotify
sic listened by other users, have discussions with users, how they choose them, which songs are played
them, even receive recommendation based on an and how they use other functions of the application.
individual’s taste, comparing one’s music taste This enables a continuous analysis inside the cloud-
with other users, providing statistical data, along based service in order to learn the interconnections
with distribution services intended mainly for de- between users and songs. In the second layer, all con-
buting artists. clusions from the first analysis are confronted with
the user’s musical preferences. This means not only
The preferred type of used streaming platforms, ei- what a particular user listens to, but what songs he or
ther payable or free, is shown in Fig. 3. she likes to combine. Processing these two sets of de-
pendencies results in a weekly playlist. On this basis,
a recommendation playlist for each individual user is
released every week. This list of albums and artists
is not a random selection of an algorithm. It results
from a careful analysis prepared by music journalists,
people who know what new and noteworthy pieces ap-
pear on the market.
Other applications include TuneIn (with 4% in 2015
and 0% after 2015), Apple Music, launched in 2015
(with less than 10% between 2017–2019), and YouTube
Fig. 3. Preferred type of streaming platform. Music, launched at 2017 (with less than 5% between
2017–2019).
Not surprisingly, users favour free services instead The dedicated platform for iOS, Apple Music, is an
of payable ones. It is worth mentioning that when it integrated system that combines streaming and Inter-
comes to quality, some of them offer only a limited net radio with a social platform in order to track the
bitrate, most often equal to 48 kbps for audio con- activity of one’s favourite artist. However, the feature
tent. Higher bitrates, ranging up to 128 kbps or even of creating and sharing playlists is not available.
320 kbps, depending on the codec used, are available Other platforms, considering those not indicated
only for those who buy a premium account or make in this study, include Deezer and Google Play Music.
a payment and/or monthly subscription. Surprisingly, Deezer is the largest library of recordings, comprising
after 2015 there was a significant increase in the num- over 35 million songs. The service differentiates be-
ber of users who wanted to consume premium high- tween subscribers. The premium version enables one
quality content. Most often this option was also related to access higher-quality music files, with bitrates rang-
with less and/or no advertisements, including commer- ing up to 320 kbps. It is available for downloading files
cials launched at the beginning, end, or after a part of and streaming recordings and can be used on e.g. TVs
material. In 2016, the percentage of payable services and car audio systems.
reached 30%, and remained at this level up till 2018. Google Play Music has more than 20 million songs.
Since 2019, a huge increase can be observed. The share Each user can create a personalised music collection,
of the most popular services is shown in Fig. 4. even up to 50 000 songs. The music recommendation
The leading streaming service, Spotify, has over 60 system itself is based on a number of factors, including
million active users and 15 million subscriptions. It is playback history, social media activity, and activity in
available on the most popular operating systems, in- the service application. This service enables users to
 P. Falkowski-Gilski – On the Consumption of Multimedia Content Using Mobile Devices. . . 325

add their own recordings and buy additional songs in

the Google Play store (Kostek, 2018).

3.2. Consumer devices

Even though listening to music using portable mul-

timedia players is opposed to listening to live music,
a trend can be observed. Currently, as shown in Fig. 5,
over 80% of users utilise this kind of devices in order
to consume content. The biggest leap was observed in
2016 and 2019, where 99% declared possessing and us- Fig. 7. Preferred type of device for content consumption
ing portable players. using streaming platforms.

Fig. 5. Usage of portable multimedia player devices. Fig. 8. Preferred type of portable device.

As indicated by the users, digital content has one

As indicated in Fig. 6, the smartphone remains
main advantage, which is annotation. Annotation is
the most popular device. Not surprisingly, the lap-
generally understood as associating any element of mu-
top comes at the second place, since it combines the
sical content, such as lyric, title, music genre, with
productivity of a desktop device with the mobility of
some additional information, i.e. comments. However,
smaller portable ones. The third device of choice is
when talking about labeling pieces of music, playlists,
a classical MP3/MP4 player. Other included e.g. a CD
radio channels, etc., the problem is still far from be-
player (with 2% in 2015 and 0% after 2015), E-book
ing solved, or even optimised. Music is very complex
reader used by people to listen to audio files (with 1%
and exhibits a significant amount of variation. Differ-
in 2015 and 2% in 2018 and 2019), as well as a tablet
ent genres or styles can overlap, making metadata ex-
(with 6% in 2015 and 3% in 2018 and 2019).
traction more difficult.

3.3. Processing and storing content

In modern devices, having the display on while de-
coding multimedia content can consume a large part of
energy. Of course, the energy required to decode audio
or video content depends on the computational com-
plexity of the codec and/or compression algorithms
used for encoding. Researchers discovered that H.263+
(Cote et al., 1998) is the least energy hungry, whereas
MPEG-4 (Koenen, 2002) and Windows Media are
Fig. 6. Most popular portable multimedia players. the most energy hungry codecs or compression tech-
niques (Lin et al., 2010). However, many research ef-
With the outbreak of numerous portable devices, forts improve battery life time of mobile devices by in-
their availability, including different operating systems, troducing different techniques while decoding, such as
as well as light weight, users tend to favour multimedia dynamic voltage scaling (Simunic et al., 2001), CPU
consumption on portable, rather than desktop devices, register, or cache optimisation (Asaduzzaman, Mah-
as shown in Fig. 7. goub, 2006), traffic concealing at the network interface
When it comes to defining the most popular (Wang et al., 2011), OS or application level optimi-
portable device, as shown in Fig. 8, after 2015, the sation (Mohapatra, Venkatasubramania, 2003).
smartphone is undeniably the first choice, the laptop Display optimisation for multimedia streaming to mo-
comes in second, whereas the tablet has a fairly stable bile devices also had been studied in (Hsiu et al.,
group of fans. 2011).
326 Archives of Acoustics – Volume 45, Number 2, 2020

According to the obtained results, as shown in

Fig. 9, users still favour lossy compression algorithms
for offline multimedia consumption. Despite the out-
break of newer and more efficient coding techniques, as
shown in Fig. 10, MP3 is still the leading format, FLAC
came in second, whereas WAV came third. Other, in-
cluding the AAC (Advanced Audio Coding) and Ogg
Vorbis format, did not exceed 12%. Quite surprisingly,
neither of the aforementioned formats exceeded 2%.
However, it should be pointed out that the results from
Figs 9 and 10 refer only to the offline multimedia con- Fig. 11. Preferred bitrate of audio content.
sumption.
3.4. Internet connection

Although display and decoding are often respon-

sible for a large part of energy consumption, wireless
interfaces can equally deplete the same amount of en-
ergy while running audio or video streaming applica-
tions on mobile devices. The type of preferred Internet
connection is shown in Fig. 12, whereas the consump-
tion of multimedia content with a mobile data plan is
shown in Fig. 13. According to the obtained results,
the biggest overbalance was observed in 2015. Since
Fig. 9. Preferred type of compression method. 2016, the network load of both fixed and wireless con-
nections, i.e. cellular or Wi-Fi, tends to balance. Cur-
rently, over 90% of users regularly consume multimedia
content using a mobile data plan.

Fig. 10. Preferred audio format for storing audio files.

A very fundamental question remains open whether

Fig. 12. Preferred type of Internet connection.
quality of music is still important. One may have an
impression that this is no longer an issue, as millions of
users download and stream music of low quality, most
often in order to cope with limitations of a mobile data
plan. On the other hand, live music events, both con-
cert performances and reinforced events, gather thou-
sands of people and this brings new technology to live
reinforced music. Not surprisingly, whenever users have
a choice, they choose the highest bitrate available, as
shown in Fig. 11. Only in 2015 users declared a bitrate
lower than 128 kbps.
The process of assigning appropriate bitrates is not Fig. 13. Consumption of multimedia content using
trivial and takes a lot of time, especially when man- a monthly mobile data plan.
aging with limited resources (Gilski, 2017b). Tagging
music data is most often carried out manually and re- Just a few years ago, as shown in Fig. 14, in 2015
quires a person with a musical background. Currently, the majority of users had a data plan limit of 2 GB
service providers lean to the conclusion that social tag- and less, with approximately a quarter even less than
ging, carried out by users themselves, will help music 1 GB. From 2016, this situation changed dramatically,
services to describe their cloud content more precisely. where the majority of people have a data plan of more
 P. Falkowski-Gilski – On the Consumption of Multimedia Content Using Mobile Devices. . . 327

As it has been shown, there is a constant change

in demand for devices and services. This forces new
means and ways to emerge, especially when it comes
to creating, processing, and delivering multimedia con-
tent, which has a significant effect not only on users
or individuals but above all device manufacturers and
network providers. The constant change for higher user
revenue is joined with continuous efforts to maintain
stable, reliable, and high-quality services.
In fact, next generation wireless technologies have
Fig. 14. Data limit of a mobile data plan. put a significant emphasis on supporting distribution
of rich media content and video-on-demand services.
than 2 GB. However, still less than 10% of users have However, energy consumption in the handheld wireless
an unlimited data plan. devices is a major bottleneck that hinders the growth
According to the studies, surprisingly, Wi-Fi can of mobile device based rich media services. The biggest
use roughly three times the energy required to decode problem today in the mobile world is that mobile de-
audio or video content, whereas 3G interface requires vices are battery driven and battery technologies are
around five times the audio decoding energy (Hoque not matching the required energy demand. It can be
et al., 2011; Chandra, Vahdat, 2002). The reason for argued whether constant staring at the screen is the
such high energy consumption is the continuous flow best way to take advantage of convenience that mo-
of network traffic, which forces these wireless radios to dern technology brings.
be powered up most of the time during streaming. Al-
though these wireless technologies operate at the phys-
ical layer (PL), their power consumption highly de- References
pends on the wireless interface usage or management at 1. Asaduzzaman A., Mahgoub I. (2006), Cache mod-
the upper layers of the Internet protocol stack, such as eling and optimization for portable devices running
at link layer (LL), network layer (NL), transport layer MPEG-4 video decoder, Multimedia Tools and Appli-
(TL), and application layer (AL). Therefore, these up- cations, 28: 239–256, doi: 10.1007/s11042-006-6145-y.
per layers should be included in the energy consum- 2. Brachmański S. (2018), Quality evaluation of speech
ption optimisation process. As it is shown, the aspect AAC and HE-AAC coding, Proceedings of Joint Con-
of network planning and management, particularly re- ference – Acoustics 2018, Ustka, Poland, Septem-
lated to energy efficiency in wireless communication, ber 11–14, 2018, pp. 1–4, doi: 10.1109/ACOUSTICS.
for both audio and video multimedia streaming toward 2018.8502335.
mobile devices, is still a wide topic, with different op- 3. Cao J., McNeill K.M., Zhang D., Nunamaker J.F.
timisation technologies still to come. (2004), An overview of network-aware applications for
mobile multimedia delivery, Proceedings of the 37th
4. Summary Annual Hawaii International Conference on System
Sciences, HICSS 2004, Big Island, HI, USA, Jan-
In order to consume multimedia content, one has uary 5–8, 2004, pp. 4663–4672, doi: 10.1109/HICSS.
to be connected to the Internet. With the outbreak 2004.1265689.
of numerous devices with access to the web, referred 4. Chandra S., Vahdat A. (2002), Application-specific
to as IoT (Internet of Things), usage to music techno- network management for energy-aware streaming of
logy enables to perceive the whole music ecosystem as popular multimedia formats, Proceedings of the Gene-
an inventory controlled by computer based technology, ral Track of the Annual Conference on USENIX An-
which includes not only music but also its user. One nual Technical Conference, ATEC ‘02, Monterey, CA,
should be aware that our music taste, mood, what we USA, June 10–15, 2002, pp. 329–342.
share, things we download or stream, are all tagged. 5. Christman E. (2017), U.S. music industry sees first
Thus we constitute an integral part of this ecosystem. double digit growth in almost 20 years as streaming
Undisputedly, new means and ways of delivering takes over, https://www.billboard.com/articles/busi-
multimedia content have changed the way we perceive ness/7744268/riaa-us-music-industry-2016-revenue-do-
information. Currently, thanks to the global web, peo- uble-digit-growth [access: 05.06.2019].
ple can easily consume and exchange digitalised data. 6. Cote G., Erol B., Gallant M., Kossentini F.
With the outcome of new mobile devices and stream- (1998), H.263+: video coding at low bit rates, IEEE
ing services, users changed their preferences. Currently Transactions on Circuits and Systems for Video Tech-
most of us do not use physical drives to store music or nology, 8(7): 849–866, doi: 10.1109/76.735381.
video files. Online streaming platforms are doing this 7. Gilski P., Stefański J. (2016), Can the digital
for us. surpass the analog: DAB+ possibilities, limitations
328 Archives of Acoustics – Volume 45, Number 2, 2020

and user expectations, International Journal of Elec- 20. Lorch J.R., Smith A.J. (1998), Software strate-
tronics and Telecommunications, 62(4): 353–361, doi: gies for portable computer energy management, IEEE
10.1515/eletel-2016-0049. Personal Communications, 5(3): 60–73, doi: 10.1109/
8. Gilski P. (2017a), DAB vs DAB+ radio broadcasting: 98.683740.
a subjective comparative study, Archives of Acoustics, 21. Mohapatra S., Venkatasubramanian N. (2003),
42(4): 157–165, doi: 10.1515/aoa-2017-0074. PARM: power aware reconfigurable middleware,
9. Gilski P. (2017b), Adaptive multiplex resource al- Proceedings of the 23rd International Conference
location method for DAB+ broadcast system, Pro- on Distributed Computing Systems, Providence, RI,
ceedings of 21st Signal Processing: Algorithms, Archi- USA, May 19–22, 2003, pp. 312–319, doi: 10.1109/
tectures, Arrangements, and Applications, SPA 2017, ICDCS.2003.1203480.
Poznań, Poland, September 20–22, 2017, pp. 337–342, 22. Počta P., Beerends J.G. (2015), Subjective and ob-
doi: 10.23919/SPA.2017.8166889. jective assessment of perceived audio quality of current
10. Havinga P.J.M., Smit G.J.M. (2001), Energy- digital audio broadcasting systems and web-casting ap-
efficient wireless networking for multimedia applica- plications, IEEE Transactions on Broadcasting, 61(3):
tions, Wireless Communications and Mobile Comput- 407–415, doi: 10.1109/TBC.2015.2424373.
ing, 1(2): 165–184, doi: 10.1002/wcm.9. 23. Simunic T., Benini L., Acquaviva A., Glynn P.,
11. Hoque M.A., Siekkinen M., Nurminen J.K. De Micheli G. (2001), Dynamic voltage scaling and
(2011), On the energy efficiency of proxy-based traf- power management for portable systems, Proceedings
fic shaping for mobile audio streaming, Proceedings of of the 38th Annual Design Automation Conference,
Consumer Communications and Networking Confer- DAC ‘01, Las Vegas, NV, USA, June 18–22, 2001, pp.
ence, CCNC 2011, Las Vegas, NV, USA, January 9–12, 524–529, doi: 10.1145/378239.379016.
2011, pp. 891–895, doi: 10.1109/CCNC.2011.5766635.
24. Uhl T., Paulsen S. (2014), The new, parameter-
12. Hoque M.A., Siekkinen M., Nurminen J.K. ized VT model for determining quality in the video-
(2014), Energy efficient multimedia streaming to mo- telephony service, Bulletin of the Polish Academy
bile devices – a survey, IEEE Communications Sur- of Sciences: Technical Sciences, 62(3): 431–437, doi:
veys & Tutorials, 16(1): 579–597, doi: 10.1109/SURV. 10.2478/bpasts-2014-0045.
2012.111412.00051.
25. Uhl T., Paulsen S., Nowicki K. (2017), New
13. Hsiu P.C., Lin C.H., Hsieh C.K. (2011), Dynamic approach for determining the QoS of MP3-coded
backlight scaling optimization for mobile stream- voice signals in IP networks, EURASIP Journal on
ing applications, Proceedings of the 17th IEEE/ACM Audio Speech and Music Processing, 1: 1–9, doi:
International Symposium on Low-Power Electronics 10.1186/s13636-016-0099-4.
and Design, ISLPED ’11, Fukuoka, Japan, August
1–3, 2011, pp. 309–314, doi: 10.1109/ISLPED.2011. 26. Vallina-Rodriguez N., Crowcroft J. (2012),
5993655. Energy management techniques in modern mobile
handsets, IEEE Communications Surveys & Tutori-
14. Iwacz G., Jajszczyk A., Zajączkowski M. (2008),
als, 15(1): 179–198, doi: 10.1109/SURV.2012.021312.
Multimedia broadcasting and multicasting in mobile
00045.
networks, Chichester, United Kingdom: John Wiley &
Sons. 27. Wang R., Tsai J., Maciocco C., Tai T.Y.C., Wu J.
(2011), Reducing power consumption for mobile plat-
15. Kennedy M., Ksentini A., Hadjadj-Aoul Y.,
forms via adaptive traffic coalescing, IEEE Journal on
Muntean G.M. (2012), Adaptive energy optimization
Selected Areas in Communications, 29(8): 1618–1629,
in multimedia-centric wireless devices: a survey, IEEE
doi: 10.1109/JSAC.2011.110911.
Communications Surveys & Tutorials, 15(2): 768–786,
doi: 10.1109/SURV.2012.072412.00115. 28. Xiong X., Song J., Yue G., Liu J., Xie L. (2011),
16. Koenen R. (2002), Overview of the MPEG-4 standard, Survey: research on QoS of P2P reliable streaming
Geneva, Switzerland. media, Journal of Networks, 6(8): 1114–1121, doi:
10.4304/jnw.6.8.1114-1121.
17. Kostek B. (2018), Listening to live music: life be-
yond music recommendation systems, Proceedings of 29. Zhang J., Wu D., Ci S., Wang H., Katsagge-
Joint Conference – Acoustics 2018, Ustka, Poland, los A.K. (2009), Power-aware mobile multimedia:
September 11–14, 2018, pp. 1–5, doi: 10.1109/ACOUS- a survey, Journal of Communications, 4(9):600–613,
TICS.2018.8502385. doi: 10.4304/jcm.4.9.600-613.
18. Leszczuk M., Janowski L., Romaniak P., Papir Z. 30. Zieliński S. (2016), On some biases encountered
(2013), Assessing quality of experience for high defini- in modern audio quality listening tests (part 2): se-
tion video streaming under diverse packet loss patterns, lected graphical examples and discussion, Journal of
Signal Processing: Image Communication, 28(8): 903– the Audio Engineering Society, 64(1/2): 55–74, doi:
916, doi: 10.1016/j.image.2012.09.006. 10.17743/jaes.2015.0094.
19. Lin C.H., Liu J.C., Liao C.W. (2010), Energy ana- 31. Zyka K. (2019), The digital audio broadcasting jour-
lysis of multimedia video decoding on mobile handheld ney from the lab to listeners – the Czech Repub-
devices, Computer Standards & Interfaces, 32(1–2): lic case study, Radioengineering, 28(2): 483–490, doi:
10–17, doi: 10.1016/j.csi.2009.04.003. 10.13164/re.2019.0483.