Reliable Interference Prediction and Management with Time-Correlated Traffic for URLLC

• In designing ultra-reliable low-latency communication (URLLC) services in 5G-and-beyond systems, link adaptation (LA) plays a vital role in adjusting transmission parameters under channel and interference dynamics. Without capturing such dynamics (e.g., relying on average estimates), the LA algorithms fail to simultaneously meet the strict reliability and latency bounds of mission-critical applications. To this end, this paper focuses on interference prediction-based adaptive resource allocation of one-shot URLLC transmission, wherein our solution deviates from the conventional average-based interference estimation schemes. We predict the next interference value based on the interference distribution estimation using a discrete-time Markov chain (DTMC). Further, to exploit the time correlation of each interference source, we model the correlated interference variations as a second-order DTMC to achieve higher prediction accuracy. While accounting for the risk sensitivity of interference estimates, the prediction outcome is then used for appropriate resource allocation of a URLLC transmission under link outage constraints. We evaluate the complete solution, given in the form of an algorithm, using Monte-Carlo simulations, and compare it with the first-order baseline counterpart. The analysis shows that the second-order interference estimate can fulfill the target outage as low as 10-7and improve the outage probability more than ten times in some scenarios compared to the baseline scheme while keeping the same amount of resource usage. 

Subject headings

TEKNIK OCH TEKNOLOGIER  -- Elektroteknik och elektronik -- Kommunikationssystem (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Electrical Engineering, Electronic Engineering, Information Engineering -- Communication Systems (hsv//eng)

Keyword

correlated traffic

discrete-time Markov chain

interference prediction

link adaptation

URLLC

Publication and Content Type

ref (subject category)

kon (subject category)