MAC for Efficient IoT connectivity

Timely delivery of data is essential for Internet-of-Things (IoT)-based critical services, but so is energy efficiency at the device side to maintain persistent connectivity and uninterrupted functioning. In this thesis, we investigate both reliable critical data transmission to ensure timely delivery of critical data and goal-oriented communication to secure energy efficiency at the device side. First, we focus on the IoT alarm scenario and propose two methods to elevate the communication success rate, thereby ensuring a timely delivery of the event-triggered/critical data. The first method is an on-device intelligence-based distributed random access scheme, while the second is a central entity-based channel scheduling method; both tailored for networks with spatially correlated device activation. Next, we shift our focus to goal-oriented communication and propose an edge intelligence-based goal-oriented sensor reporting scheduling method that prioritizes the transmission of data that is pertinent to the receiver's goal, rather than transmitting the entire observed data. This selective transmission approach thereby leads to energy efficiency improvements at the device side. The results demonstrate that all three proposed methods outperform their respective benchmark methods, where the random access scheme and channel scheduling method increases the success rate of critical data transmission, while the proposed goal-oriented scheduling method minimizes the number of device transmissions.