Method and apparatus for identification of an access network by means of 1-port measurements
View Patent ↗Method and apparatus for modeling a network are described in which a 1 port measurement is made on the network by inputting an excitation signal at one port of the network and recording the results reflected back to the port. A 1 port parametric model of the network is generated whereby as much information about the topology of the network is included in calculating initial values of the parameters. Then the values of the parameters of the 1 port parametric model are optimized by reducing the difference between the results of the measurement step and results calculated using the 1-port parametric model and the excitation signal.
1. A method of modeling a network, comprising the steps of:
carrying out a 1 port measurement on the network by inputting an excitation signal at one port of the network and recording the results at the port;
generating a 1 port parametric model of the network;
optimizing the values of the parameters of the 1 port parametric model by reducing the difference between the results of the measurement step and results calculated using the 1-port parametric model and the excitation signal; and
generating a 2-port parametric model using the optimized parameter values of the 1 port parametric model to provide a 2-port model of the network.
2. The method according to claim 1 , wherein the reducing step is carried out by minimizing a cost function by adapting the values of the parameters of the 1 port parametric model, the cost function representing a quantification of the difference.
3. The method according to claim 1 , further comprising a step of limiting the 2-port model to a transfer function.
4. The method according to claim 3 further comprising the steps of:
measuring a first noise PSD at one port of the network and estimating or measuring a second PSD at the other port; and
using the transfer function and the first and second noise PSD to estimate the capacity of the network.
5. A system for modeling a network, comprising:
means for carrying out a 1 port measurement on the network by inputting an excitation signal at one port of the network and recording the results at the port;
means for generating a 1 port parametric model of the network;
means for optimizing the values of the parameters of the 1 port parametric model by reducing the difference between the results of the measurement step and results calculated using the 1-port parametric model and the excitation signal; and
means for generating a 2-port parametric model using the optimized parameter values of the 1 port parametric model to provide a 2-port model of the network.
6. The system according to claim 5 , wherein the means for reducing comprises means for minimizing a cost function by adapting the values of the parameters of the 1 port parametric model, the cost function representing a quantification of the difference.
7. The system according to claim 5 , further comprising: means for limiting the 2-port model to the transfer function.
8. The system according to claim 7 further comprising:
means for measuring a first noise PSD at one port of the network and estimating or measuring a second PSD at the other port; and
means for using the transfer function and the first and second noise PSD to estimate the capacity of the network.
9. A data carrier medium carrying one or more computer readable code segments for controlling a processing device to carry out the method in accordance with claim 1 .
10. A computer program product for execution of the method in accordance with claim 1 on a computer system.
11. A method for modeling a representation of network, said method comprising:
transmitting a description of the network from a near location to a remote computing system via a telecommunications network;
executing on the remote computer system the method of claim 1 ; and
transmitting to a near location information relating to the response of the network to an excitation signal.