Chapter 26: Multiuser Information Theory

Learning Objectives

• Derive the capacity region of the Gaussian multiple access channel (MAC) using random coding with successive interference cancellation, and prove the converse via Fano inequality
• Prove the capacity region of the degraded Gaussian broadcast channel (BC) using superposition coding, and state the dirty-paper coding theorem for the MIMO BC
• Classify interference channel regimes (strong, weak, moderate) and state the Han-Kobayashi achievable rate region and the Etkin-Tse-Wang approximate capacity result
• Derive the cut-set upper bound and decode-forward lower bound for the relay channel, and describe noisy network coding for general relay networks
• Define degrees of freedom (DoF) for multiuser channels and prove the Cadambe-Jafar result that the K-user interference channel has K/2 DoF via interference alignment
• State the finite blocklength normal approximation theorem, define channel dispersion, and analyse its implications for URLLC system design