Chapter Summary

Chapter 18 Summary

Key Points

• 1.

  When the number of users $K$ exceeds the number of transmit antennas $n_t$, user scheduling is essential. Multiuser diversity provides a $\log K$ SNR enhancement, but the sum-rate gain scales only as $n_t \log\log K$ — diminishing returns that nevertheless justify the scheduling overhead in practical systems.

• 2.

  The ISAC deterministic-random tradeoff (Liu/Caire 2023, IEEE Joint Paper Award) provides the information-theoretic foundation for integrated sensing and communications. The transmitted signal is decomposed into a deterministic sensing component and a random communication component, and the Pareto boundary of the rate-CRB region is traced by optimizing the power split.

• 3.

  Even the random communication signal contributes to sensing (through its covariance structure), but a dedicated deterministic component is needed for optimal sensing performance. The tradeoff severity depends on the angular separation between the communication user and the sensing target.

• 4.

  Topological interference management achieves non-trivial DoF using only knowledge of the interference graph topology, not the channel coefficients. The topological DoF is determined by graph-theoretic quantities (independence number) and is connected to the index coding problem from computer science.

• 5.

  The hierarchy of CSIT — no CSIT (TDMA), topological CSIT (TIM), statistical CSIT, full CSIT (IA/DPC) — reveals a spectrum of interference management strategies with increasing complexity and performance. Practical systems typically operate between topological and statistical CSIT.