P19 - An Efficient and Low Complexity Greedy Power Allocation Algorithm for URLLC Links
- Event
- iCCC2024 - iCampµs Cottbus Conference
2024-05-14 - 2024-05-16
Cottbus - Band
- Poster
- Chapter
- Kommunikation und HF-MEMS
- Author(s)
- N. Odhah, E. Grass, R. Kraemer - IHP Leibniz-Institut für innovative Mikroelektronik, Frankfurt/Oder
- Pages
- 175 - 178
- DOI
- 10.5162/iCCC2024/P19
- ISBN
- 978-3-910600-00-3
- Price
- free
Abstract
In this paper, an efficient and low-complexity power allocation algorithm for a URLLC-link is proposed and analyzed based on a previously published formula for calculating the effective channel capacity. The underlying optimization problem differs from standard communication links because the well-known Shannon formula cannot be applied, and two additional rate penalties must be considered. The first penalty results from the short block length of the transmitted URLLC data packets, and the second penalty comes from the additional delay caused by the queueing process at the link layer. Two mathematical methods are applied to deal with these penalties: the Finite Block Length Information Theory (FBL-IT) and the Effective Capacity (EC) calculation. We explain this formula for calculating the effective channel rate, including the two mentioned constraints for URLLC communication. Furthermore, to approach a solution for the optimization problem, an efficient greedy algorithm for allocating the total transmit power over the available channel uses of an OFDM-based URLLC system is proposed and studied. Since the URLLC service will be shared with other communication services on the same link, the greedy algorithm will be used to allocate the total allowed power of the link to the different individual communication channels in such a way that their communication requirements can be fulfilled. For this purpose, the link is divided into groups of channels with the same properties, and the greedy algorithm is applied to each group individually. Simulation results show the efficiency and reliability of the proposed algorithm for both eMBB and URLLC links.