Wireless power control schemes and optimization
Abstract/Contents
- Abstract
- Wireless communication network users are subject to a number of physical and systemic constraints. In particular, low power and mobile users such as those in the Internet of Things (IoT) spectrum are under significant pressure to provide reliable communication with significant endurance while being supported by finite power sources. The need for such efforts are becoming increasingly paramount as rapid user adoption and growth are leading to congested spectrum in a distributed environment. Intelligent power control and optimization schemes can improve device efficiency while simultaneously increasing network capacity through judicious and opportunistic use of network resources. The benefits can be had in any wireless network, but are particularly applicable to heterogenous networks, such as those containing 802.11, 802.15.4, as well as other protocols. This dissertation first considers the problems of Transmission Control Protocol (TCP) in a wireless environment. It is a known problem that TCP congestion control is suboptimal in that it underutilizes channel bandwidth in wireless environments. This is driven by an inappropriate response to non-congestive wireless losses. We develop a simplified model of the TCP congestion control window evolution process and estimate the effect of channel conditions on producing non-congestive losses and the subsequent long term effect on the congestion window and by proxy the link throughput. We introduce a pausing mechanic, and a transmission power control metric to minimize non-congestive losses in poor channel conditions. These metrics are balanced by introducing a penalty for the undesirable outcome of delaying packet transmissions using dynamic programming. The resultant algorithm, CR-TCP is adapted to readily available consumer hardware and tested in a residential/consumer network setting. The experimental results validate the promise of CR-TCP by showing improvements in throughput in poor channel conditions while maintaining performance levels in good conditions. We then consider the problem of wireless channel selection in systems that have reactive users, a trait that is most typically a product of using transmission power control schemes. In the case of traditional passive sensing methods may make suboptimal channel selections when compared to active probing that incorporates a determined channel reactivity. Our method uses non-invasive power levels so as to not excessively perturbate existing channel users. A number of metric parameters including probing steps, evolution profile, and interference metric are presented. Simulations comparing passive sensing with non-invasive power probing show cases in which optimal channel selection can result in lower system power use as well as increased stability. Finally, we analyze data aggregation for low power wireless sensor applications, considering the overhead of infrequent transmission over a 802.15.4 framework using current generation SoC solutions. In using dynamic programming as a heuristic that balances user data delay with power savings, we are able to show that power savings and reduction in channel accesses outweigh the increases in data delay.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2017 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Schubert, Kevin |
|---|---|
| Associated with | Stanford University, Department of Electrical Engineering. |
| Primary advisor | Bambos, Nicholas |
| Thesis advisor | Bambos, Nicholas |
| Thesis advisor | Apostolopoulos, John G |
| Thesis advisor | Özgür, Ayfer |
| Advisor | Apostolopoulos, John G |
| Advisor | Özgür, Ayfer |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Kevin Schubert. |
|---|---|
| Note | Submitted to the Department of Electrical Engineering. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2017. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2017 by Kevin Schubert
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...