利用Wozencraft系综构造新的WOM码
New constructions of WOM codes using the Wozencraft ensemble
摘要:
本文给出了几种新的WOM码的构造方法。在我们的建设的新颖性是使用所谓的Wozencraft合奏的线性代码。具体来说,我们得到以下结果。我们给出了一个明确的两写写一次存储器(简称WOM)的代码,接近容量,在二进制字母表的建设。更正式地说,对任意ε>0,0<p<1和n =(1/ε)^{O(1/pε)},我们构造了一个长度为n,容量为H(p)+1-p-ε的二写WOM码.由于两次写入WOM码的容量为max_p(H(p)+1-p),因此我们得到ε-接近容量的码。此外,编码和解码可以分别在时间O(n^2.poly(log n))和时间O(n.poly(log n))以及对数空间中完成。我们得到了一种新的三写WOM码的编码方案。当块长度为exp(1/ε)时,我们的方案达到了1.809-ε的速率。这给出了比使用先前技术可以实现的速率更好的速率。我们强调一个连接到线性种子提取器位固定的来源。特别是,我们表明,获得这样一个提取器的种子长度为O(log n),可以导致改进的参数为2写WOM代码。然后,我们给出了现有的提取器的构造的应用程序的问题,设计编码方案的内存缺陷。
英文摘要:
In this paper we give several new constructions of WOM codes. The novelty in our constructions is the use of the so called Wozencraft ensemble of linear codes. Specifically, we obtain the following results. We give an explicit construction of a two-write Write-Once-Memory (WOM for short) code that approaches capacity, over the binary alphabet. More formally, for every ε>0, 0<p<1 and n =(1/ε)^{O(1/pε)} we give a construction of a two-write WOM code of length n and capacity H(p)+1-p-ε. Since the capacity of a two-write WOM code is max_p (H(p)+1-p), we get a code that is ε-close to capacity. Furthermore, encoding and decoding can be done in time O(n^2.poly(log n)) and time O(n.poly(log n)), respectively, and in logarithmic space. We obtain a new encoding scheme for 3-write WOM codes over the binary alphabet. Our scheme achieves rate 1.809-ε, when the block length is exp(1/ε). This gives a better rate than what could be achieved using previous techniques. We highlight a connection to linear seeded extractors for bit-fixing sources. In particular we show that obtaining such an extractor with seed length O(log n) can lead to improved parameters for 2-write WOM codes. We then give an application of existing constructions of extractors to the problem of designing encoding schemes for memory with defects.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。