First draft for content review
Pre-release
Pre-release
Comments:
- I have marked typos and such into the paper version - I can collect my
copy for corrections - you write as if this is a report of Tuxera achiements - it is not.
This is your thesis and you must make it clear what did you do
personally. This applies to several Chapters e.g. conclusions and
description of the implementation and testing. - structure: Now you go to 4 levels of headers but under-use numbering
of Chapters. My suggestion would be to try to keep 3 levels of headers.
What could help is to have more chapters by turning some 2nd level
Sections into Chapters. For example, 2.2. could be a Chapter, if you
keep 2.3 in this Chapter, get rid of 2.3.3 but keep the 2.3.3.1 and 2.3.3.2 - For comparison, clear criteria would be helpful, e.g. functionality,
quality under loss and capacity constraints, power consumption of the
Mobile device (if you did not measure it, may be there are qualitative
statements that are possible? OR are they all the same in terms of
power consumption? Possibly also diversity leading to incompatibility? - when you present each standard in Section 2.2. a Figure similar to
Figure 1 showing also possible mapping to equipment would make
everything more concrete. In particular, it is not always clear for some
of the standards, how is the support for cloud based content organised
(YouTube/Netflix etc). - Figures 12... better use kBytes or Mbytes instead of bytes - makes the
figures more readable. Why is AirPlay volume < DLNA in Fig 12a but later
it turns out that volumes are: AirPlay > DLNA because AirPlay uses
non-compressed format? - Conclusions: HTTP vs RTP - it is possible to make RTP more adaptive to
network conditions by adding support for RTCP for feedback, some
heuristics that for example try to adjust sending rate to available
capacity and finally an adaptive playout algorithm. For this approach to
work alternative coding rates most likely must be supported. OR the
playout algorithm must be able to stop playout for a while and start
again once there is enough content. Is this approach not considered on
any of the forums?