POMDP-Based Decision Making for Fast Event Handling in VANETs
Authors: Shuo Chen, Athirai Irissappane, Jie Zhang
AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results demonstrate that our model can effectively balance the decision quality and response time while still being robust to sophisticated malicious attacks. |
| Researcher Affiliation | Academia | Shuo Chen, Athirai A. Irissappane, Jie Zhang School of Computer Science and Engineering, Nanyang Technological University, Singapore University of Washington, Washington, USA chen1087@e.ntu.edu.sg, athirai@u.washington.edu, zhangj@ntu.edu.sg |
| Pseudocode | Yes | Algorithm 1: Learning the Observation Function |
| Open Source Code | No | The paper does not provide an explicit statement or link to its own open-source code. It links to a third-party tool (Cluster 3.0 software) used in the experiments, but not the authors' own implementation. |
| Open Datasets | No | The paper describes a simulation setup where data is generated internally: "We evaluate our POMDP model using the Veins simulator..., which combines OMNe T++ for vehicle-to-vehicle communication simulation, and SUMO for road traffic microsimulation." It does not mention using a publicly available dataset with concrete access information (link, DOI, or specific citation to a benchmark). |
| Dataset Splits | No | The paper describes simulation periods and data collection intervals but does not specify explicit train/validation/test dataset splits (e.g., percentages or sample counts) needed for reproduction in a traditional machine learning sense. The learning of the observation function occurs dynamically over time in the simulation, rather than on fixed splits. |
| Hardware Specification | No | The paper mentions using simulators (Veins, OMNeT++, SUMO) but does not provide any specific hardware details (e.g., CPU, GPU models, memory, or cloud instance types) on which these simulations were run. |
| Software Dependencies | Yes | K-means clustering is performed by Cluster 3.0 software5. 5http://bonsai.hgc.jp/~mdehoon/software/cluster/software.htm |
| Experiment Setup | Yes | The values for rewards used in our experiments are: R(r = yes, RE) = R(r = yes, IG) = 30, R(r = no, IG) = R(r = no, RE) = 20 and R(r, Q) = 1. The ground truth update delay from centralized map service is set to 60s. The freshness is fresh if the trustworthiness is updated within 30 seconds (s) and old otherwise. The time window W is set to 60s. |