5G URLLC: Ultra-Reliable Low-Latency Communications in 5G systems

One of the key differences between 5G networks and currently used 4G networks is the ability to provide a fundamentally new service called Ultra-Reliable Low-Latency Communications (URLLC). The usage of URLLC means that the network provides data delivery with low latency and extremely high reliability. Such transmission parameters are necessary for several existing applications (for example, industrial automation and robot control), as well as applications that are expected to appear in the near future (intelligent transportation systems, control of autonomous vehicles, remote surgery, Tactile Internet, etc.).

The specifications for 5G systems introduce several mechanisms and technical solutions for URLLC service support. However, there are many open problems associated with the development of algorithms for adaptive mechanism selection and parameter setting to provide URLLC services. These problems depend on the scenario, the number of devices in the network, the traffic generated by them, and the requirements of specific applications, as well as the delay and reliability of data delivery.

Within the Grant of the Russian Science Foundation (РНФ) No. 21-79-10158, “Providing Ultra-Reliable Low-Latency Communications in 5G systems and beyond,” the Wireless Network Lab developed a number of algorithms and solutions that are vital to provide the URLLC service. WNL proposed an algorithm for selecting a precoder and MCS based on a vector autoregressive channel prediction model, reducing channel resource consumption by up to 5 times. Algorithms for selecting MCS, scheduling radio resources, and dividing users were developed, increasing network capacity for multicast URLLC traffic by up to three times. An adaptive algorithm for selecting uplink transmission parameters using Grant-Free channel access method was developed, increasing network capacity by up to two times compared to existing solutions. A neural network-based algorithm called ALPACA was proposed to predict the future state of a channel, increasing the capacity of URLLC traffic by up to 40%. Algorithms for multiplexing URLLC and broadband traffic using MU-MIMO spatial multiplexing in the downlink were developed, increasing URLLC coverage by up to 50% and broadband throughput by up to 60%. A similar dynamic multiplexing solution for the uplink was developed, doubling broadband throughput while maintaining URLLC quality of service. Finally, a hybrid scheme for channel resource allocation to serve URLLC traffic with variable packet sizes was proposed and analytically optimized, increasing network capacity by up to two times.

List of relevant papers:


      2025
    1. Evgeny Khorov, Artem Krasilov, Alexey Shashin, Nikolay Nikolaev. Analysis of Hybrid Radio Resource Allocation Scheme for Video Traffic in 5G V2X Systems. // Problems of Information Transmission, Vol. 61, No. 3, pp. 245–256, 2025. doi: 10.1134/S0032946025030032.

      2. 2024
    2. Evgeny Khorov, Artem Krasilov, Alexey Shashin. Initial Parameters Selection for Serving Uplink URLLC Traffic of High-Mobility Users. //Journal of Communications Technology and Electronics. – 2023. – Т. 68. – №. 12. – С. 1515-1522. https://doi.org/10.1134/S1064226923120185.
    3. Evgeny Khorov, Artem Krasilov, Ruslan Yusupov, Irina Lebedeva. Resource-Efficient Multicast URLLC Service in 5G Systems. //Sensors. – 2024. – Т. 24. – №. 8. – С. 2536. https://doi.org/10.3390/s24082536.

      4. 2023
    4. Evgeny Khorov, Aleksei Kureev, Kirill Glinsky. ALPACA: An Asymmetric Loss Prediction Algorithm for Channel Adaptation Based on a Convolutional-Recurrent Neural Network in URLLC Systems. // IEEE Access, 2023. https://doi.org/10.1109/ACCESS.2023.3344317.
    5. Evgeny Khorov, Andrey Belogaev, Artem Krasilov, Alexey Shashin. Adaptive Parameters Selection for Uplink Grant-Free URLLC Transmission in 5G Systems. //Computer Networks. – 2023. – Т. 222. – С. 109527. https://doi.org/10.1016/j.comnet.2022.109527.
    6. Evgeny Khorov, Artem Krasilov, Aleksei Kureev, Kirill Glinsky. Performance of ML-Based Channel Prediction Algorithms for URLLC: Channel Model Matters. //2023 IEEE International Black Sea Conference on Communications and Networking (BlackSeaCom). – IEEE, 2023. – С. 306-311. https://doi.org/10.1109/BlackSeaCom58138.2023.10299788.

      7. 2022
    7. Evgeny Khorov, Artem Krasilov, Ruslan Yusupov, Irina Lebedeva. Efficient Multiplexing of Downlink eMBB and URLLC Traffic with Massive MU-MIMO. //2022 IEEE International Black Sea Conference on Communications and Networking (BlackSeaCom). – IEEE, 2022. – С. 185-190. https://doi.org/10.1109/BlackSeaCom54372.2022.9858204.
    8. Evgeny Khorov, Artem Krasilov, Aleksei Kureev, Kirill Glinsky. PABAFT: Channel Prediction Approach Based on Autoregression and Flexible TDD for 5G Systems. //Electronics. – 2022. – Т. 11. – №. 12. – С. 1853. https://doi.org/10.3390/electronics11121853.
    9. Evgeny Khorov, Artem Krasilov, Ruslan Yusupov, Irina Lebedeva. Multiplexing of URLLC and eMBB Traffic in a Downlink Channel with MU-MIMO. //Journal of Communications Technology and Electronics. – 2022. – Т. 67. – №. 12. – С. 1506-1512. https://doi.org/10.1134/S1064226922120129.
    10. Evgeny Khorov, Andrey Belogaev, Artem Krasilov, Alexey Shashin. Algorithm for Transmission Parameters Selection for Sporadic URLLC Traffic in Uplink. //Journal of Communications Technology and Electronics. – 2022. – Т. 67. – №. 12. – С. 1492-1499. https://doi.org/10.1134/S1064226922120191.