A hybrid strategy for cellular data offloading to wifi access point based on type of application
The cellular networks are increasingly facing the challenge of data explosion due high demands of data by users. Several forecasts and analysis indicates that, in the near future the technology cannot cater for the users’ data demands. Many attempts to upgrade the technology resulted into inefficien...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2014
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The cellular networks are increasingly facing the challenge of data explosion due high demands of data by users. Several forecasts and analysis indicates that, in the near future the technology cannot cater for the users’ data demands. Many attempts to upgrade the technology resulted into inefficient or expensive solution. Cellular data offloading to Wi-Fi is the most promising solution. The two main strategies for cellular data offloading are delay-tolerance and on-the-spot. The on-the-spot offloads less significant percentage of the traffic while the delay strategy achieves higher offloading percentage but with high rate of packet loss and high latency which affect the performance of real-time services. This research proposed an enhancement over the existing delay-algorithm. On receiving the user data request the strategy checks the availability and signal strengths of both the Wi-Fi and cellular and then classified the applications into real-time and non-real-time. The strategy always connects to Wi-Fi network if available. But when the Wi-Fi signal falls below the minimum threshold during the transfer, the strategy will complete the transfer using cellular immediately for real-time applications and wait till the end of the delay deadline in case of nonreal- time applications before it switch to cellular. The simulation result shows the enhanced algorithm achieved 59% of offloading efficiency similarly to the Wiffier delay-algorithm but with 28% reduction in packet loss and 30% less completion time for real-time traffics compares to original delay-algorithm. |
|---|