We propose fully distributed multi-group multicast precodingdesigns for cell-free massive multiple-input multiple-output (MIMO) systems with modest training overhead. We target the minimization of the sum of the maxi...
详细信息
We propose fully distributed multi-group multicast precodingdesigns for cell-free massive multiple-input multiple-output (MIMO) systems with modest training overhead. We target the minimization of the sum of the maximum mean squared errors (MSEs) over the multicast groups, which is then approximated with a weighted sum MSE minimization to simplify the computation and signaling. To design the joint network-wide multi-group multicast precoders at the base stations (BSs) and the combiners at the user equipments (UEs) in a fully distributed fashion, we adopt an iterative bi-directional training scheme with UE- and/or group-specific precoded uplink pilots and group-specific precoded downlink pilots. To this end, we introduce a new group-specific over-the-air uplink training resource that entirely eliminates the need for backhaul signaling for the channel state information (CSI) exchange. The precoders are optimized locally at each BS by means of either best-response or gradient-based updates, and the convergence of the two approaches is analyzed with respect to the centralized implementation with perfect CSI. Finally, numerical results show that the proposed distributed methods greatly outperform conventional cell-free massive MIMO precodingdesigns that rely solely on local CSI.
In this letter, we address the problem of distributed multi-antenna cooperative transmission in a cellular system. Most research in this area has so far assumed that base stations not only have the data dedicated to a...
详细信息
In this letter, we address the problem of distributed multi-antenna cooperative transmission in a cellular system. Most research in this area has so far assumed that base stations not only have the data dedicated to all the users but also share the full channel state information (CSI). In what follows, we assume that each base station (BS) only has local CSI knowledge. We propose a suboptimal, yet efficient, way in which the multicell MISO precoders may be designed at each BS in a distributed manner, as a superposition of so-called virtual SINR maximizations: a virtual SINR maximizing transmission scheme yields Pareto optimal rates for the MISO Interference Channel (IC);its extension to the multicell MISO channel is shown to provide a distributedprecoding scheme achieving a certain fairness optimality for the two link case. We illustrate the performance of our algorithm through Monte Carlo simulations.
The trend of developing distributed multiple-input multiple-output (MIMO) cooperation has been growing for future wireless networks due to its potential of capacity improvement through network-level precoding. In mass...
详细信息
ISBN:
(纸本)9781665435406
The trend of developing distributed multiple-input multiple-output (MIMO) cooperation has been growing for future wireless networks due to its potential of capacity improvement through network-level precoding. In massive MIMO applications, the overhead of channel state information (CSI) exchange among distributed transmitters is too large to make it possible in practical implementations. In this paper, we consider a cooperative multicell massive MIMO network with distributed regularized zero-forcing (RZF) precoding at each base station (BS), where a novel CSI exchange scheme is devised to reduce the interactive overhead. As a key finding of this work, we theoretically prove that it suffices to share the Gram matrix of local CSI among the cooperative BSs in order to achieve the same performance as a centralized cooperative MIMO network using the RZF precoding with global CSI sharing. The CSI exchange from each BS is thus reduced to a symmetric matrix that has a much smaller size than the full CSI and the amount of CSI exchange does NOT grow with the large number of antennas in massive MIMO. Specifically, based on the exchanged Gram matrices, we derive a decentralized RZF precodingdesign at each BS and develop both the optimal and suboptimal cooperative power allocation strategies, which achieve different performance and complexity tradeoffs. A virtual centralized power allocation is accomplished at each BS and the performance achieved by the proposed decentralized precoding is the same as the centralized benchmark scheme with full CSI exchange. These superiorities of the proposed schemes are verified through simulation results.
暂无评论