Robust signal processing for asynchronous distributed massive MIMO

在这个NSERC合作研究和发展基金的建议，我们建议研究集中式和分散式信号处理技术的异步分布式大规模多输入多输出（MIMO）系统。分布式大规模MIMO系统由多个基站（BS）组成，每个基站配备有多个天线。 这种系统受益于大规模MIMO系统提供的优势（快速衰落和不相关噪声的影响消失），同时允许分布式处理和利用用户分集。在我们的研究中，我们专注于分布式大规模MIMO系统的三个方面，即能源效率，延迟和计算复杂度。在大规模MIMO通信系统的传统模型中，通常假设网络中的不同用户/基站在符号级上是时间同步的。然而，由于不同用户位于距大规模MIMO基站不同距离处的事实，该假设可能不现实。据我们所知，已发表的分布式大规模MIMO结果忽略了两个事实。经常被忽略的第一个事实是，由一个用户设备发送的信号将很可能以不同的延迟到达不同的大规模MIMO BS。也被忽略的第二个事实是，由不同的大规模MIMO BS发送的信号也以不同的延迟到达一个接收器。例如，在具有许多基站并且在每个基站处可能具有许多天线的协作多点（CoMP）方案中，需要考虑这些前述延迟。由于这些延迟，端到端信道不能再被建模为频率平坦链路。实际上，端到端信道最好用多径（频率选择性）脉冲响应来建模。端-端-端信道的这种观点使得该信道的均衡不可避免，无论它将是单载波均衡方案还是多载波均衡方案。 **

In this NSERC Collaborative Research and Development Grant proposal, we propose to study centralized and de-centralized signal processing techniques for asynchronous distributed massive multiple input multiple output (MIMO) systems. Distributed massive MIMO systems are comprise of many base stations (BSs) each of which is equipped with multiple antennas. Such systems benefit from the advantages offered by massive MIMO systems (where the effects of fast fading and uncorrelated noise are vanished) while allowing distributed processing and exploiting user diversity.****In our study, we focus on three aspects of distributed massive MIMO systems, namely energy efficiency, latency, and computational complexity. In traditional models for massive MIMO communication systems, it is often assumed that different users/base stations in the network are time-synchronized at the symbol level. This assumption, however, may not be realistic due to the fact that different users are located at different distances from the massive MIMO base stations. To the best of our knowledge, the published results on distributed massive MIMO ignore two facts. The first fact which is often ignored is that the signals transmitted by one user device will very likely arrive at different massive MIMO BSs with different delays. The second fact which is also ignored is that signals transmitted by different massive MIMO BSs also arrive at one receiver with different delays. For example, in coordinated multi-point (CoMP) scheme with many base stations and with possibly many antennas at each base station, these aforementioned delays need to be taken into account. Due to these delays, the end-to-end channel can no longer be modeled as a frequency-flat link. Indeed, the end-to-end channel is best modeled with a multi-path (frequency selective) impulse response. Such a view point on the end-end-end channel makes the equalization of this channel inevitable, whether it will be a single-carrier equalization scheme or a multi-carrier one. **