宇宙微波背景（CMB）是过去几十年天文领域最精确的观测之一。它帮助解决了宇宙学领域很多悬而未决的谜题。我们期待下一代 CMB 卫星能够回答的问题中，暴涨属于最令人感兴趣的一类。如果暴涨理论的大致图景是正确的，那么暴涨就是导致宇宙间不均匀性的原因。有了不均匀性才有今天丰富多彩的结构。随着标量场粒子迹象不断显现，暴涨理论的基础变得更坚固，因此对暴涨模型参数的限制也变得更有意义。在本文中，我们利用 Fisher 矩阵方法预测未来 CMB 数据，能告诉我们哪些关于暴涨的新信息。为了做到这一点，需要先对CMB 限制宇宙学参数时，各参数之间的简并性有了解。通过计算，我们发现极化信号，尤其是 B 模式极化，可以帮助获取比仅有温度信号显著精确的暴涨参数误差范围。如果 Planck 卫星能够探测到 B 模式信号，将能有效区分暴涨模型。否则的话，则需要有对极化信号更为敏感的观测设备。尽管主体计算默认在 FRW 宇宙学框架下进行，并假设了标准模型，本文也对不均匀宇宙学，特别是 LTB 模型，进行了一些讨论。我们的计算结果暗示，有一大类LTB模型，即EdS空洞模型，在拟合不同观测数据时，可能存在普遍的不协调性。这就提供了批量排除 LTB 模型的可能。虽然对于这一现象目前还不能作定论，但在一定意义上这使得前述的标准模型假设更加可信。

The Comic Microwave Background (CMB) is one of the most precise observations the astronomy community has obtained in the past decades. It help answer many a long-lasting question in the field of cosmology.Among all the pending problems the next generation CMB observatories are going to solve, the existence of inflation is one of the most extraordinary ones. If the working inflation theory is schematically correct, then one should give the credit of the splendid complexity of the universe, hence the human, to inflation.As the traces of the scaler field (Higgs) accumulate, the foundation of inflation become smore solid, which makes the constraints on inflationary parameters wellmotivated.We tried in this work figuring out what the coming observational data might be able to tell us about inflation, with the Fisher Matrix method. In order to do this, one had better to firstly have a clue of the degeneracies between the parameters.Through calculation, we found that the polarization observation greatly increase the constraining power of CMB. But to differentiate more effectively among various inflation models, the Planck Surveyor might not be surficient; especially when Planck fails todetect the B-mode polarization. In that case, instruments with even higher polarization-sensitivity will be needed.Although the basic line is following an FRW cosmology and assuming the standardcosmological model, we discussed along the way a bit about the LTB cosmology. There is sign, according to our calculation, that a big familly of the LTB models, namely the EdS voids model, are not consistent in fitting different observational data. This is not conclusive, but to some extent solidify the said assumption.