Network topology inference is one of the important applications of network tomography.Traditional network topology inference may impact network normal operation due to its generation of huge data traffic.A unicast net...Network topology inference is one of the important applications of network tomography.Traditional network topology inference may impact network normal operation due to its generation of huge data traffic.A unicast network topology inference is proposed to use time to live(TTL)for layering and classify nodes layer by layer based on the similarity of node pairs.Finally,the method infers logical network topology effectively with self-adaptive combination of previous results.Simulation results show that the proposed method holds a high accuracy of topology inference while decreasing network measuring flow,thus improves measurement efficiency.展开更多
We propose a novel optical time-to-live （TTL） processing scheme using asymmetric Mech-Zehnder inter-ferometer （AMZI） and Fabry-Perot semiconductor optical amplifier （FP-SOA）. AMZI transfers M TTL pulses into M-1...We propose a novel optical time-to-live （TTL） processing scheme using asymmetric Mech-Zehnder inter-ferometer （AMZI） and Fabry-Perot semiconductor optical amplifier （FP-SOA）. AMZI transfers M TTL pulses into M-1 pulses and two residual pulses with 6-dB power difference. FP-SOA enhances the power difference between the M - 1 pulses to the residual pulses to more than 10 dB. A numerical model is established for verifying the feasibility of this scheme.展开更多
基金supported by the National Natural Science Foundation of China (Nos.61373137,61373017, 61373139)the Major Program of Jiangsu Higher Education Institutions (No.14KJA520002)+1 种基金the Six Industries Talent Peaks Plan of Jiangsu(No.2013-DZXX-014)the Jiangsu Qinglan Project
文摘Network topology inference is one of the important applications of network tomography.Traditional network topology inference may impact network normal operation due to its generation of huge data traffic.A unicast network topology inference is proposed to use time to live(TTL)for layering and classify nodes layer by layer based on the similarity of node pairs.Finally,the method infers logical network topology effectively with self-adaptive combination of previous results.Simulation results show that the proposed method holds a high accuracy of topology inference while decreasing network measuring flow,thus improves measurement efficiency.
基金This work was supported by the National Natural Science Foundation of China under Grant No. 60572008.
文摘We propose a novel optical time-to-live （TTL） processing scheme using asymmetric Mech-Zehnder inter-ferometer （AMZI） and Fabry-Perot semiconductor optical amplifier （FP-SOA）. AMZI transfers M TTL pulses into M-1 pulses and two residual pulses with 6-dB power difference. FP-SOA enhances the power difference between the M - 1 pulses to the residual pulses to more than 10 dB. A numerical model is established for verifying the feasibility of this scheme.
