机器学习信号处理工程师实习-自动化方向

职位描述
该岗位包含两个部分的内容，一部分包括通过FPGA和verilog 进行开发眼动仪相关功能，该部分功能的实现作为单独的岗位职责
一分部开发一套基于脉冲波形库与机器学习回归分析的智能测量系统，用于溶液离子浓度和类型的自动识别与定量预测。
岗位主要负责脉冲库的创建与可视化、双通道数据采集、信号特征提取、以及模型训练与性能评估等。
岗位职责：
• 使用verilog对FPGA进行开发
• 使用PicoScope 4000 系列实现化学溶液信号的采集与记录；
• 参与脉冲波形库的创建、优化与可视化；
• 通过编程控制PicoScope实现自动化数据采集；
• 对实验信号进行处理与特征提取，构建机器学习数据集；
• 训练与优化回归模型（如 XGBoost 等）实现离子类型与浓度预测；
• 对结果进行可视化分析、性能评估，并撰写项目技术报告。
• 学习电路搭建，电机控制，PCB相关知识和技能
任职条件：
1.电子、信号、通信、计算机或相关理工科专业硕士及以上学历；
2.熟练掌握python编程及常用数据处理库；
3.熟悉信号处理理论与方法（傅里叶变换、小波变换、滤波与特征分析）；
4.理解实验仪器控制与数据采集（如AWG、PicoScope）；
5.熟悉机器学习回归模型及模型优化；
6.具备良好的科研习惯、数据分析及文档撰写能力；
7.具备较强的独立思考、问题解决与团队协作能力；
8.具备良好的英语书面和口语沟通能力，能够使用英文进行工作交流。
加分项：
-有嵌入式开发背景或相关项目经验（如ESP32、STM32）；
-熟悉Git/Github代码管理
Signal Processing Engineer
Job Description
This project aims to develop an intelligent measurement system based on a pulse
waveform library and machine learning regression analysis, designed for
automatic identification and quantitative prediction of ion concentration and
type in chemical solutions.
The position primarily involves creating and visualizing the pulse library,
controlling the PicoScope device, conducting dual-channel data acquisition,
extracting signal features, and training and evaluating regression models.
Responsibilities:
• Use the PicoScope 4000 series to acquire and record signals from chemical
solutions;
• Participate in the creation, optimization, and visualization of the pulse
waveform library;
• Implement automated data acquisition through PicoScope programming control;
• Process and extract features from experimental signals to build machine
learning datasets;
• Train and optimize regression models (e.g., XGBoost) for ion type and
concentration prediction;
• Perform result visualization, performance evaluation, and prepare technical
project reports.
Qualifications:
1. Master’s degree or above in Electronics, Signal Processing, Communications, Computer Science, or other related engineering fields;
2. Proficiency in Python programming and common data processing libraries;
3. Solidunderstanding of signal processing theories and methods (Fourier Transform, Wavelet Transform, filtering, and feature analysis);
4. Familiarity with experimental instrument control and data acquisition (e.g., AWG, PicoScope);
5. Knowledge of machine learning regression models and model optimization techniques;
6. Strong research habits, data analysis, and technical documentation skills;
7. Excellent independent thinking, problem-solving, and teamwork abilities;
8. Good written and spoken English communication skills for technical collaboration.
Preferred Qualifications:
• Experience using PicoScope or related projects;
• Background or experience in embedded development (e.g., ESP32, STM32);
• Familiarity with Git/GitHub for code management.