esa-sceva/satcom-mcqa

---
license: apache-2.0
language:
- en
tags:
- satellite
- communictions
pretty_name: satcom-mcqa
size_categories:
- n<1K
---

# 🛰️ Satcom Multiple Choice Question-Answer Dataset

## Dataset Summary

**SATCOM-MCQA** is a **human-generated multiple-choice evaluation dataset** for Satellite Communications (SatCom).

* Around 1,000 questions with 3-4–5 answer options each
* Questions cover technical SatCom topics (e.g., link budget, BER, antenna gain, propagation)
* ~30% math-focused questions requiring quantitative reasoning
* Designed for evaluation

---

## Key Features

* **Human-written multiple-choice questions**
* Covers core SatCom/NTN domains:

* Link Budget
* Bit Error Rate (BER)
* Antenna & Beamforming
* Propagation & Frequency bands
* Multi-layer NTNs (satellites, HAPS, UAVs)
* 6G integration

* Includes **reference links** where relevant
* No synthetic content or model-generated text

---

## Intended Use

* Benchmarking LLMs or classifiers on technical SatCom MCQ reasoning
* Evaluating knowledge in satellite engineering, NTN design, and 6G research
* Testing quantitative reasoning in engineering tasks

---

## Data Structure

### Fields

* `question` *(string)* — the question text
* `options` *(list of strings)* — multiple-choice options
* `answer` *(string)* — the correct option (e.g., `"Option 2"`)
* `topic` *(string)* — the domain/topic of the question
* `mathematical_reasoning_required` *(bool)* — if numeric/math reasoning is needed
* `raw_latex` *(string or null)* — optional LaTeX formulas

### Example

```json
{
"question": "Antenna gain increases when the effective area of the antenna is:",
"options": ["Decreased", "Increased", "Constant", "None of these"],
"answer": "Option 2",
"topic": "Link Budget",
"mathematical_reasoning_required": false,
"raw_latex": null
}
```

---

## Source Material

Questions were crafted using academic and technical literature on SatCom:

* Link budget fundamentals
* Bit error rate calculations
* Antenna design and beamforming papers
* Publicly available academic reports

---

## Example Usage in Python

```python
from datasets import load_dataset
ds = load_dataset("esa-sceva/satcom-mcqa")
test_ds = ds["train"]
sample = test_ds[0]
print("Question:", sample["question"])
print("Options:", sample["options"])
print("Answer:", sample["answer"])
```

---
license: Apache-2.0
language:
- 英语
tags:
- 卫星
- 通信
pretty_name: satcom-mcqa
size_categories:
- 样本量不足1000
---

# 🛰️ 卫星通信多项选择题问答数据集

## 数据集概述

**SATCOM-MCQA**是一款由人工生成的卫星通信（Satellite Communications，简称SatCom）多项选择题评测数据集。

* 共包含约1000道题目，每道题配备3至5个备选答案
* 题目涵盖卫星通信核心技术议题（如链路预算、误比特率（Bit Error Rate，简称BER）、天线增益、电波传播等）
* 约30%的题目为数学导向型，需要进行定量推理
* 专为评测场景设计

## 核心特性

* **人工编写的多项选择题**
* 覆盖卫星通信/非地面网络（Non-Terrestrial Networks，简称NTN）核心领域：
* 链路预算
* 误比特率（BER）
* 天线与波束赋形
* 电波传播与频段
* 多层非地面网络（卫星、高空平台站（High Altitude Platform Stations，简称HAPS）、无人机（Unmanned Aerial Vehicle，简称UAV））
* 6G集成
* 相关题目附带参考链接
* 无合成内容或模型生成文本

## 适用场景

* 用于大语言模型（Large Language Model，简称LLM）或分类器在卫星通信技术类多项选择题推理任务中的基准测试
* 用于评测卫星工程、非地面网络设计以及6G研究领域的知识掌握情况
* 用于测试工程任务中的定量推理能力

## 数据结构

### 字段说明

* `question`（字符串类型）：题目文本
* `options`（字符串列表类型）：多项选择题备选答案
* `answer`（字符串类型）：正确答案（例如：`"Option 2"`）
* `topic`（字符串类型）：题目所属领域/议题
* `mathematical_reasoning_required`（布尔类型）：是否需要进行数值/数学推理
* `raw_latex`（字符串或空值类型）：可选的LaTeX公式内容

### 示例

json
{
"question": "天线增益会在天线有效面积____时提升",
"options": ["减小", "增大", "保持不变", "以上均不对"],
"answer": "Option 2",
"topic": "链路预算",
"mathematical_reasoning_required": false,
"raw_latex": null
}

## 来源材料

题目基于卫星通信领域的学术与技术文献编写而成：

* 链路预算基础
* 误比特率计算
* 天线设计与波束赋形相关论文
* 公开可获取的学术报告

## Python示例用法

python
from datasets import load_dataset
ds = load_dataset("esa-sceva/satcom-mcqa")
test_ds = ds["train"]
sample = test_ds[0]
print("Question:", sample["question"])
print("Options:", sample["options"])
print("Answer:", sample["answer"])