In this paper, we study a multi-agent game between $N$ agents, which solve a consensus problem, and receive state information through a wireless network, that is controlled by a Base station (BS). Due to a hard-bandwidth constraint, the BS can concurrently connect at most $R_{d}