| 研究課題/領域番号 |
17K12635
|
|---|
| 研究機関 | 東北大学 |
|---|
研究代表者 |
Korman Matias 東北大学, 情報科学研究科, 助教 (80732718)
|
|---|
| 研究期間 (年度) |
2017-04-01 – 2021-03-31
|
| キーワード | Computational Geometry / Algorithms / Sensor Networks / Computer Science / Routing / Time-Space Trade-off / Combinatorial Geometry / Ad-Hoc Networks |
|---|
| 研究実績の概要 |
2017 was a very successful year. We ended up publishing over 15 research papers related to the project. In the following I describe the most important results:
-We have designed two algorithms for routing in the presence of obstacles. Naturally, each of the algorithms have their own advantages and drawbacks. However, one of them is particularly interesting because it is the FIRST one that can work even if the underlying graph is not plane. This was posed explicitly as one of the milestones of the project.
-We also established the first results in the dynamic coloring problem. That is, how to color a large graph so that small changes in the topology have only local impact. Although it is a very natural problem, no serious work on this topic existed. Previously, a small change in a graph would need a whole restructuring of the coloring of the graph.
-We have successfully managed to extend the concept of mean in the presence of polygonal obstacles. We gave a formal definition that extends the one for the unconstrained (the general plane). We are now looking into algorithmic aspects of that problem.
In short, all milestones planned for year 2017 have been successfully achieved and the milestones for 2018 are nicely underway.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
1: 当初の計画以上に進展している
理由
All milestones for year 2017 have been achieved. In most cases, the papers have been written and have even appeared at conferences (one is scheduled to appear). We are preparing the journal version of these papers.
Meanwhile, we started to focus on the milestones for 2018. One of them involves making a small-scale implementation for experimenting. This process is a bit more lengthy, so we decided to start it in late 2017. The implementation is well underway. The core of the program is finished. We are preparing some experiments to make sure that the program does not have any errors.
We have also planned meetings with other researchers to work on other milestones (not related to the implementation). These meetings will be held along 2018.
|
| 今後の研究の推進方策 |
The main focus of 2018 will be the polishing of the implementation. We have focused in two topics. First, we want an implementation of memory constrained algorithms. That is, algorithms that can run with small amount of memory (such as embedded devices) at the cost of increasing the runtime. We will make a special focus in the time-space trade-off. The second problem that we will consider is trajectory clustering. Imagine you have a bunch of sensors moving along some predesignated zone (such as cell phones when they move around a city). We will look for ways of clustering the trajectories and detect hotspots in which many users gather
|
| 次年度使用額が生じた理由 |
Most of the money is budgeted toward travel expenses of the main researcher. We also have saved a small fraction to pay for conference registration, fungible materials and/or research material but this will be a relatively small amount.
|
