Remote Display Solutions for Mobile Cloud Computing


Mobile devices have become an essential part of our daily life, with smartphone sales now surpassing desktop system sales.1 As mobile device popularity grows, end-user demands to run heavier applications are also increasing.
Although advances in miniaturization continue, the desire to preserve the advantages mobile devices have over desktop systems in weight, size, and device autonomy will always impose intrinsic limits on processing power, storage
capacity, battery lifetime, and display size. Researchers must redesign conventional desktop applications to operate
on mobile hardware platforms, thereby often reducing functionality, whereas more demanding applications typically
require specific hardware resources that are unlikely to be available on mobile devices.

MOBILE CLOUD COMPUTING CHALLENGES

Essentially, mobile cloud computing physically separates
the user interface from the application logic. The mobile device executes only a viewer component, operating
as a remote display for the applications running on distant servers in the cloud. As Figure 1 shows, a remote display framework has three components: a server-side component that intercepts, encodes, and transmits the application graphics to the client; a viewer component on the client; and a remote display protocol that transfers display updates and user events between both endpoints.

DEVICE BATTERY LIFETIME

The operational time of mobile devices is often limited when they are extensively used. These battery capacity shortcomings result in short recharge cycles and prevent users from relying completely on their mobile device. During the past decade, advances in nominal battery capacity have been modest. Kostas Pentikousis5 observed that technological improvements are currently stagnating
because of the lack of a major battery technology breakthrough comparable to the advent of rechargeable Li-ion batteries.

WIRELESS BANDWIDTH AVAILABILITY

Compared to fixed access networks, modern broadband mobile and wireless technologies offer limited and variable bandwidth availability. Universal Mobile Telecommunications System (UMTS) users typically receive up to 384 kilobits per second, while Krishna Balachandran and colleagues9 reported practical throughputs of 347 Kbps for Long Term Evolution (LTE) and up to 6.1 Mbps for WiMAX. Actual throughput
depends on user mobility, interference, and fading effects.

Downstream data peak reduction

Interactive applications only update their display when instructed by the user. These display updates usually involve sending a large amount of data to the client in a short interval, which requires an instantaneous bandwidth much higher than the average bandwidth requirement. Furthermore, this bursty traffic pattern is unfavorable in wireless network environments, as it might induce additional
collisions on the wireless channel.