Preamble collision in the random access channel (RACH) is a major bottleneck in massive machine-type communication (mMTC) scenarios, typical of cellular IoT (CIoT) deployments. This work proposes a machine learning-based mechanism for early collision detection during the random access (RA) procedure. A labeled dataset was generated using the RA procedure messages exchanged between the users and the base station under realistic channel conditions, simulated in MATLAB. We evaluate nine classic classifiers -- including tree ensembles, support vector machines, and neural networks -- across four communication scenarios, varying both channel characteristics (e.g., Doppler spread, multipath) and the cell coverage radius, to emulate realistic propagation, mobility, and spatial conditions. The neural network outperformed all other models, achieving over 98\% balanced accuracy in the in-distribution evaluation (train and test drawn from the same dataset) and sustaining 95\% under out-of-distribution evaluation (train/test from different datasets). To enable deployment on typical base station hardware, we apply post-training quantization. Full integer quantization reduced inference time from 2500 ms to as low as 0.3 ms with negligible accuracy loss. The proposed solution combines high detection accuracy with low-latency inference, making it suitable for scalable, real-time CIoT applications found in real networks.

翻译：在随机接入信道（RACH）中的前导码碰撞是蜂窝物联网（CIoT）部署中典型的大规模机器类通信（mMTC）场景的主要瓶颈。本文提出了一种基于机器学习的机制，用于在随机接入（RA）过程中早期检测碰撞。通过在MATLAB中模拟真实信道条件，利用用户与基站之间交换的RA过程消息生成了一个带标签的数据集。我们评估了九种经典分类器——包括树集成、支持向量机和神经网络——在四种通信场景下的性能，这些场景变化了信道特性（如多普勒扩展、多径）和小区覆盖半径，以模拟真实的传播、移动性和空间条件。神经网络在所有模型中表现最佳，在分布内评估（训练和测试数据来自同一数据集）中实现了超过98%的平衡准确率，并在分布外评估（训练和测试数据来自不同数据集）中保持了95%的准确率。为了在典型基站硬件上部署，我们应用了训练后量化。全整数量化将推理时间从2500毫秒降低至最低0.3毫秒，且准确率损失可忽略不计。所提出的解决方案结合了高检测精度和低延迟推理，使其适用于真实网络中可扩展的实时CIoT应用。