On Embeddings between Bv and W
On Embeddings between Bv and W
复制标题
关于 Bv 和 W 之间的嵌入
DOI:
--
复制
发表时间:
--
期刊:
影响因子:
--
通讯作者:
Kuk Bv
中科院分区:
文献类型:
--
作者:
B. J. Schmitt;M. Winkler;R. Aachen;Kuk Bv
Let R N be a nonempty open set. A function u 2 L 1 (() belongs to BV (() ii its derivative Du (in the sense of distributions) is a (vector-valued) Radon measure on with nite total variation. BV (() therefore is a canonical extension of W 1;1 ((), or, vice versa, W 1;1 (() is embedded continuously into BV ((). In fact we have (1) cf. 4, sect. 5.1]. A diierent way of extending W 1;1 (() consists in decreasing the order of derivatives and considering fractional order Sobolev spaces (also known as Sobolev-Slobodeckii spaces) W s;p ((), s 2 (0; 1) and certain p 2 1; 1) (cf. 2, sect. 6.8]). Therefore one might ask whether there is, for suitable choices of s and p, a relationship between BV (() and W s;p (() in the sense that one space contains the other. In Section 1 we shall demonstrate that relaxing the diierential order always produces spaces W s;p (() being too large to lie in BV ((), while Section 2 will show that on the other hand, though being considerably larger than W 1;1 ((), BV (() already embeds into W s;p (() if only W 1;1 (() does.