论文中有用的句子：
1.Compared to small-scale MIMO [6], MaMi systems with linear detection schemes provide superior performance and hardware efficiency.
与小规模MIMO [6]相比，具有线性检测方案的MaMi系统可提供卓越的性能和硬件效率。
2.The array and spatial multiplexing gains in MaMi require handling large matrices, however, the reported higher hardware efficiencies, and low complex PAR scheme makes it very promising for future deployments. Comparison with state-of-art MIMO implementations is shown in Fig. 3.6.6.
MaMi中的阵列和空间复用增益需要处理大型矩阵，然而，报告的更高的硬件效率和低复杂的PAR方案使其在未来的部署中非常有前途。与现有技术MIMO实现的比较如图3.6.6所示。
3.
The array and spatial multiplexing gains in MaMi require handling large matrices, however, the reported higher hardware efficiencies, and low complex PAR scheme makes it very promising for future deployments. Comparison with state-of-art MIMO implementations is shown in Fig. 3.6.6.
下行链路预编码采用QRD单元，其性能和能量效率分别为34.1M-QRD / s / kGE和6.56nJ / QRD，成本低，与[3-4]相比是有利的。
4.
The analytic and numeric analysis in the previous sections have focused on cellular networks where the BSs and UEs are equipped with ideal transceiver hardware, which can radiate any waveform without distortions and which can receive any waveform with infinite resolution. However, practical transceivers inevitably operate with non-linearities in ampli- fiers, clock drifts in local oscillators, finite-precision ADCs, I/Q imbalance in mixers, and non-ideal analog filters [33]– [36]. In this section, we provide a prediction of how these hardware impairments affect the achievable SEs in multi-cell massive MIMO systems. We notice that it was recently shown in [12], using impairments models developed and evaluated in [33]–[35], that the hardware impairments caused by the BS array are negligible in massive MIMO systems, since the desired signals are amplified by the array gain from coherent processing while the distortions add non-coherently. Hence, the hardware impairments in the UE hardware are expected to be the main hardware limitation [12] and henceforth we only consider those impairments in this section.
前面部分中的分析和数值分析侧重于蜂窝网络，其中BS和UE配备有理想的收发器硬件，其可以辐射任何波形而没有失真并且可以接收具有无限分辨率的任何波形。然而，实际的收发器不可避免地在放大器中非线性，本地振荡器中的时钟漂移，有限精度ADC，混频器中的I / Q不平衡以及非理想模拟滤波器[33] - [36]。在本节中，我们提供了这些硬件损伤如何影响多小区大规模MIMO系统中可实现的SE的预测。我们注意到最近在[12]中显示，使用[33] - [35]中开发和评估的损伤模型，由于所需信号被放大，所以BS阵列引起的硬件损伤在大规模MIMO系统中可以忽略不计。阵列从相干处理中获得，而失真增加非相干性。因此，预期UE硬件中的硬件损伤是主要的硬件限制[12]，此后我们仅考虑本节中的那些损伤。
画一个二维分布图~两个参数对应不同的信息论的值~
标题：The discussion of the HPO-MIMO system’s achievable rate under the different distribution
HPO-SKEW-NORML
在传统MIMO中，香农论证了当信号是高斯分布时，可以达到信息论的上限。但是在非线性MIMO中，经过了取相位操作后，是否高斯信号仍然是使得互信息最大的分布呢？这个问题将在本章回答。
信息理论：
本节中，我们的意图是探索HPO-MIMO的可达速率界限。据我们有限的知识可知，这部分内容是首次被讨论出来。we first investigate the information theoretic limits of the HPO-MIMO system. 其次我们发现，在HPO-MIMO系统中，与传统MIMO不同，高斯信号并不能使得系统容量最优。取代的是，SKEW-NORMAL分布作为信令时，性能要比高斯分布以及QAM分布要高。
In this section, our intention is to explore the reachability rate limits of HPO-MIMO. According to our limited knowledge, this part of the content was first discussed. Secondly, we found that in the HPO-MIMO system, unlike traditional MIMO, the Gaussian signal does not optimize the system capacity. Instead, the SKEW-NORMAL distribution as a signalling performance is higher than the Gaussian distribution and the QAM distribution.
摘要：
第一句：MIMO是提高RATES的好办法，但是价格贵，功耗大。
第一句：multiple input multiple output (MIMO) communication systems are popular for achieving high rates in further generation, but at the same time they suffer from high implementation cost.
或者：
The wide bandwidth and large number of antennas used in further generation (millimeter wave) systems put a heavy burden on the power consumption at the receiver.
第二句：
In this letter, the uplink performance of HPO-MIMO systems where the base station (BS) equip with low power consumption phase-detector and only one analog-to-digital converters (ADCs) on each antenna is analyzed.
摘要：
This paper proposes a multiuser nonlinear MIMO (NL-MIMO) scheme whose base station (BS) acquires $\pi$-periodic phase measurements of the complex envelop signals.
The radio-frequency (RF) circuits are designed by phase detectors, low-resolution analog-to-digital converters (ADCs) and low-precision low-noise-amplifiers (LNAs).
Consequently, their power and cost are much lower than their inphase-quadrature (IQ)-structured correspondences.
However, the BS has to recover user data based on imprecise observations suffering from nonlinear distortions including quantization losses, nonlinear LNA distortions, and magnitude missing. The channel estimation (CE) and multiuser detection (MUD) are firstly categorized as a generalized linear mixing problem with phase observations, and then resolved by generalized approximate message passing (GAMP) with sampling integration and message censoring mechanisms.
