Machine Learning based Channel estimation in V2V communication

Machine Learning based Channel estimation in V2V communication

Internship Description

The data rates provided by the prevalent dedicated short-range communication (DSRC) standard in vehicle-vehicle (V2V) communication do not support sharing of the large amount of data generated by sensors in modern vehicles. The solution lies in the exploitation of the large bandwidths available in the millimeter wave (mmWave) spectrum (30 - 300 GHz). Fortunately, contemporary automotive radars already operate in mmWave band and therefore, their hardware can be reused for V2V communication. The sparsity in angle and delay domains of mmWave channels could be utilized for efficient channel estimation in V2V communication. As the sparsity structure in mmWave channels is dictated by the locations of scatterers, we expect the structure to change rapidly in highly mobile environment of V2V communications. However, the quick variation in channels is expected to be systematic as the location of scatterers will change in a systematic manner. This effect will be more prominent in the vehicle-to-infrastructure (V2I) scenario where the fixed location of the antenna and surrounding objects results in a fixed sparse component in addition to a varying sparse component. In this project, our goal would be to use radar estimates to quickly predict and track the expected pattern in sparsity structure using machine learning algorithms. Specifically, we aim to propose sparsity aware channel estimation methods that could predict and track the fast changing sparsity pattern to assist in channel estimation. ​

​​

Deliverables/Expectations

1. Understanding channel estimation in mmWave communication.
2. Understanding the blend of automotive radar operations and communication in V2V scenarios.
3. Algorithms (MATLAB code) for channel estimation in mmWave based V2V communication.
4. Final report summarizing and explaining all project work and reporting results of evaluation tests performed. 

Faculty Name

Tareq Al-Naffouri

Field of Study

Electrical Engineering