Short-Packet URLLCs for MIMO Underlay Cognitive Multihop Relaying with Imperfect CSI

In this work, we investigate short-packet communications for multiple-input multiple-output underlay cognitive multihop relay networks with multiple primary users, where transceivers transmit and receive short packets to provide ultra-reliable low-latency (uRLLCs). For performance evaluation, the closed-form expressions of end-to-end (E2E) block error rate (BLER) considered systems are derived in a practical scenario under imperfect channel state information interference channels, from which E2E throughput, energy efficiency (EE), latency, reliability, asymptotic analysis also studied. Based on analytical results, adapt some state-of-the-art machine learning (ML)-aided estimators predict system terms EE, reliability real-time configurations. We obtain optimal power-allocation relay-location strategies minimize BLER proportional tolerable power uRLLC constraints, require negligible computational complexity offer significant savings. Furthermore, ML-based evaluation achieves equivalent while significantly reducing execution time compared conventional simulation methods. Among ML frameworks, extreme gradient boosting model is demonstrated be most efficient estimator future applications.