1: \begin{abstract} 

2: This paper investigates the interference nulling capability of reconfigurable

3: intelligent surface (RIS) in a multiuser  environment where

4: multiple single-antenna transceivers communicate simultaneously in a shared

5: spectrum. From a theoretical perspective, we show that when the channels

6: between the RIS and the transceivers have line-of-sight and the direct paths

7: are blocked, it is possible to adjust the phases of the RIS elements to null

8: out all the interference completely and to achieve the maximum $K$

9: degrees-of-freedom (DoF) in the overall $K$-user interference channel, provided that the number of RIS elements exceeds some finite value that depends on $K$. Algorithmically, for any fixed channel realization we formulate the interference nulling problem as a feasibility problem, and propose an

10: alternating projection algorithm to efficiently solve the resulting nonconvex

11: problem with local convergence guarantee.  Numerical results show that the

12: proposed alternating projection algorithm can null all the interference if the

13: number of RIS elements is only slightly larger than a threshold of $2K(K-1)$.

14: For the practical sum-rate maximization objective, this paper proposes to use

15: the zero-forcing solution obtained from alternating projection as an initial

16: point for subsequent Riemannian conjugate gradient optimization and shows that

17: it has a significant performance advantage over random initializations.  For

18: the objective of maximizing the minimum rate, this paper proposes a subgradient

19: projection method which is capable of achieving excellent performance at low complexity.

20: \end{abstract}

21: