Motivation:

There has been a growing interest in semantic and goal-oriented communication systems in recent years. This interest is mainly driven by two reasons: First, due to emerging applications such as autonomous driving, smart cities, and so on the amount of data being transmitted through the wireless network is increasing unprecedentedly which can possibly saturate the network capacity. Second, with the advancements in human-to-machine (H2M) and machine-to-machine (M2M) communications, there is no interest in following the current content-delivery type of communication. It is actually the extraction and delivery the task-relevant information that has received huge attention recently. Thus, by semantic communication we aim to make communication systems intelligent, meaning that it needs to understand the relevance, urgency, and meaning of the data traffic in conjunction with the underlying task.

Objectives:

The project aims at investigating distributed semantic encoding in a multi-user network where users collaborate together to perform a single task. The primary objectives include:

• Developing an efficient scheme for transmitting task-relevant information collaboratively
• Comparing the proposed method with existing distributed encoding schemes

Related Topics:

• Task-Oriented Communication
• Distributed Information Bottleneck (D-IB) 
• Deep Neural Network (DNN)

Requirements:

• Information Theory Background
• Programming Skills (Python)
• Basic Knowledge of Machine Learning
• Being Self-Motivated and Committed

For More Information:

Please do not hesitate to contact me if you are interested: halimi@ant.uni-bremen.de