A Fuzzy-Based Approach to Remove Clock Skew and Reset From One-Way Delay Measurement  

Oleh:

Lin, Yu ; Wang, Hongbo ; Kuo, Geng-Sheng ; Cheng, Shiduan ; Li, Qi

Jenis: Article from Journal - ilmiah internasional
Dalam koleksi: IEEE Transactions on Neural Networks vol. 16 no. 5 (Sep. 2005), page 1125-1135.
Topik: Clock; fuzzy - based approach; remove clock; skew and reset; one - way; delay measurement

Ketersediaan

Perpustakaan Pusat (Semanggi) Nomor Panggil: II36.12 Non-tandon: 1 (dapat dipinjam: 0) Tandon: tidak ada

Lihat Detail Induk

Isi artikel

One - way delay (OWD) traces are important measurements for analyzing end - to - end performance on the Internet. It is still a great challenge to provide a scalable solution for large-scale OWD measurement. Because the clocks at end systems are usually not synchronized, the OWD measurements are often inaccurate. For the more challenging case with clock resets to some reference times during the measurement, the OWD measurements are more inaccurate. Furthermore, the measurement data often exhibit considerable network - induced noise when the network is heavily loaded. All the existing OWD measurement techniques, such as linear programming and convex - hull approach (CHA), try to solve this problem by deterministic mathematics model. However, they often fail to distinguish clock resets from temporary Internet congestion. Based on the fuzzy - clustering analysis, this paper proposes a new algorithm to estimate and remove the clock skews and resets from measurement results. This algorithm has been implemented as a tool called fuzzy - based OWD corrector (FOC). The paper then presents OWD measurements of several Internet paths using FOC. Numerical experiments demonstrate that FOC is more accurate and robust than the existing techniques. FOC s computation complexity O (N) is similar to that of CHA and its computing time is much less than that of convex - hull technique.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali