Analysis of crosstalk and modeling of possible number of routing stages for optically transparent wavelength-division multiplexing nodes of generalized multiprotocol label-switched networks

Patil, Suvarna S.; Chaudhari, Bharat S.; Li, Baojun

doi:10.1117/1.3509193

Cited by 2 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The connection request is placed from the initiator to the respective target, with an assumption that, the connection request arrival to be a random process, modeled as Poisson distribution using well-known Erlangs traffic Model. Considering its mean 'm' and holding time of the connected path with mean 'μ' then the probability of getting path connection is given as [6] ( )…”

Section: Analytical Modelingmentioning

confidence: 99%

Passive Micro-resonator based GMPLS Router for Optical Network-on-Chip

Patil

Chaudhari²

2011

NPA

Self Cite

View full text Add to dashboard Cite

Generalized multiprotocol label switching (GMPLS) with wavelength division multiplexing (WDM) technology is a very promising protocol technology for future optical networks. GMPLS technology interconnects new and legacy networks by automating connection provisioning and traffic engineering. The present electrical interconnects for the system on chip (SoC) are unable to satisfy the multiple design requirements of bandwidth, data rate and latency. In this paper, for the first time we have proposed a micro resonator based GMPLS router for optical network on chip (ONoC). The evolution of such a router node on integrated circuit technology will cause the system design to move towards a communication-based architecture. The concept of an integrated GMPLS optical interconnect will be a potential technological solution, alleviating some of the more pressing issues involved in exchanging data between cores in SoC architectures. The investigation for ONoC was carried out at the physical level, where the system performance on the basis of crosstalk, blocking probability, offered traffic load and packet error rate (PER) was analyzed.

show abstract

Section: Analytical Modelingmentioning

confidence: 99%