| 研究課題/領域番号 |
23K16877
|
|---|
| 研究種目 |
若手研究
|
| 配分区分 | 基金 |
|---|
| 審査区分 |
小区分60060:情報ネットワーク関連
|
|---|
| 研究機関 | 国立情報学研究所 |
|---|
研究代表者 |
劉 佳 国立情報学研究所, ストラテジックサイバーレジリエンス研究開発センター, 特任助教 (10813420)
|
|---|
| 研究期間 (年度) |
2023-04-01 – 2025-03-31
|
| 研究課題ステータス |
交付 (2023年度)
|
| 配分額 *注記 |
4,550千円 (直接経費: 3,500千円、間接経費: 1,050千円)
2024年度: 2,080千円 (直接経費: 1,600千円、間接経費: 480千円)
2023年度: 2,470千円 (直接経費: 1,900千円、間接経費: 570千円)
|
|---|
| キーワード | Mobile crowdsensing / Knowledge acquisition / Incentive mechanisms / Game modeling / Equilibrium / Mobile Crowdsensing / Incentive Mechanism / Economic Feasibility / Internet of Things / Game Theory |
|---|
| 研究開始時の研究の概要 |
This research proposes a novel one-stop mobile crowdsensing (MCS) ecosystem in IoT. By elaborately devising the architecture and operating rules as well as verifying the economic feasibility, the MCS ecosystem will promote the integration of physical and information worlds for Japan Society 5.0.
|
| 研究実績の概要 |
In this fiscal year, we have proposed a one-stop mobile crowdsensing (MCS) ecosystem for knowledge acquisition that covers the whole process from upper-layer knowledge trading to underlying knowledge generation. We resort to blockchain technology and provide a series of tailored operating rules and functions to protect the truthfulness of data gathering and the fairness of knowledge trading. In addition, we have designed incentive mechanisms to stimulate selfish and rational entities in the ecosystem to participate in knowledge acquisition. To analyze the strategic interactions among entities theoretically, we have developed a nested hierarchical game model, where the upper-layer knowledge trading is evaluated based on the Contract Theory, and the lower-layer knowledge generation is formulated as a two-stage Stackelberg game. By solving the nested hierarchical game in a backward inductive way, we have identified the optimal strategy for each entity in closed form.
Except for the above research, we have also conducted studies on relevant wireless communication underlying technologies, such as signal processing, interference control, spectrum allocation, etc., in order to provide a stable and reliable information infrastructure for the MCS ecosystem.
The research achievements include publications of 5 journal papers and 3 international conference papers. Moreover, a paper has been accepted by an international journal, and 2 papers have been accepted by international conference proceedings, and these papers will be published soon.
|
| 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
In this fiscal year, we have proposed a one-stop MCS ecosystem for knowledge acquisition, which contains a complete process from the underlying data sensing, aggregating, and knowledge training, to the upper-layer knowledge trading. Leveraging emerging blockchain technology, we have designed a secure and truthful data aggregating scheme based on the PoC consensus mechanism, as well as a reliable and fair knowledge trading scheme based on smart contracts. Incentive mechanisms have also been incorporated to stimulate selfish and rational entities to participate in knowledge acquisition works. To identify the strategic interactions in the ecosystem, we have developed a nested hierarchical game model, where the upper layer is modeled by the Contract Theory and the lower layer is modeled as a two-stage Stackelberg game. By solving the hierarchical game in a backward inductive way, we obtained the optimal strategies of different entities. These progresses have exceeded the original research plan for this fiscal year.
This research is expected to produce at least 3 journal papers and 3 conference papers in two years. Fortunately, only in the first year, we have exceeded this target by producing 5 journal papers and 3 conference papers. The reason for progressing more smoothly than the initial plan is due to our solid research base accumulated from previous studies on game theoretic-based wireless network performance modeling and network economics, as well as the efforts devoted to promoting the research progress in this project.
|
| 今後の研究の推進方策 |
The follow-up research plan is as follows:
1. In a practical mobile crowdsensing ecosystem, the quality of data provided by workers varies significantly, and in order to protect their own privacy, they may introduce intentional noise into the raw data. To address these challenges, our next step will be to consider developing intelligent worker selection algorithms with learning capabilities, as well as payment strategies that ensure good economic properties in the market.
2. Similarly, the diversity of raw perception data and human disturbances pose serious challenges to the accuracy of mobile crowdsensing. Our next step will also involve considering the introduction of the ground truth discovery mechanism to optimize the data aggregation process and enhance the accuracy of the entire mobile crowdsensing system.
|
