Telephonic communications – Diagnostic testing – malfunction indication – or electrical... – Testing of subscriber loop or terminal
Reexamination Certificate
2000-01-26
2001-07-17
Kuntz, Curtis (Department: 2743)
Telephonic communications
Diagnostic testing, malfunction indication, or electrical...
Testing of subscriber loop or terminal
C379S026020, C379S008000, C379S030000
Reexamination Certificate
active
06263048
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates in general to communication systems, and is particularly directed to a new and improved scheme for testing and estimating the performance of digital subscriber (copper) loops through the transmission, reception and analysis of a multi-tone power ratio waveform, that contains a plurality of discrete tones having equal spectral spacings and spectral notches periodically distributed among the tones.
BACKGROUND OF THE INVENTION
The advent of the use of discrete multi-tone (DMT) modulation for digital subscriber line (DSL) modem technology applications, and its potential for mass deployment by local exchange carriers for high speed internet access, has presented telecommunication service providers with a unique challenge for testing digital subscriber (copper) loops. As diagrammatically illustrated in
FIG. 11
a DSL modem
10
is intended to work over an existing copper pair
11
between a customer premises (CP)
12
and central office (CO)
14
. Unlike traditional dial up modems, DSL modem signals do not pass through a PCM conversion process in the voice switch
16
of the central office equipment. Instead, the modem signal is demodulated at the central office and sent to a separate digital subscriber line access multiplexer (DSLAM)
17
, which couples the digital data stream onto the network
18
for transmission to a remote site
19
.
Although traditional dial-up modem performance is limited primarily by the PCM conversion process in the voice switch
16
, rather than the characteristics of the copper loop
11
, DSL modem performance and thus DSL service is directly dependent on the quality of the copper loop. Loop quality, in turn, is dependent on loop length, interference from sources within the cable plant and outside (radio stations, etc.), as well as bridge taps, load coils, splicing etc. It will be readily appreciated, therefore, that being able to sort cable pairs from an available bundle, based on their ability to provide a specific grade of service (data rate,) is essential to the economic deployment of DMT DSL.
As diagrammatically illustrated in
FIG. 2
, a DMT waveform provides multiple discrete tones (sub-carriers)
21
. An illustrative example is Asymmetric Digital Subscriber Line (ADSL), having a 4.3125 KHz spacing between adjacent sub-carriers 21-i and 21-(i+1), over a spectrum from about 30 KHz to about 1.1 MHz. The effective or composite data rate of the ADSL DMT waveform is the sum of the data rate on each discrete sub-carrier. The data rate for each sub-carrier is ultimately governed by the signal-to-noise and distortion ratio (SINAD) for each 4.3125 KHz channel between 30 KHz and 1.1 MHz. Given an accurate estimate of SINAD for each sub-carrier channel, a maximum theoretical data rate or payload, that is independent of modem implementation, can be calculated for the loop.
One method to conduct loop testing would be to couple actual DMT modems to each end of the copper loop and perform a bit error rate (BER) measurement to grade the line. Unfortunately, BER measurements are directly dependent on the modem manufacture's receiver implementation. A different manufacturer may have more margin over some impairments and less over others. Thus, loop grading may be artificially skewed by the receiver implementation used in the conducting the BER test.
SUMMARY OF THE INVENTION
In accordance with the present invention, this modem-dependency skew problem customarily encountered in the course of testing a telecommunication loop (e.g., a copper twisted pair used for digital subscriber line communications) is effectively obviated by taking advantage of a methodology for testing the internal analog circuitry of DMT DSL modems, which allows measurement of SINAD of the loop at various points in the DMT spectrum (between 30 KHZ and 1.1 MHz). The waveform for the test is a DMT waveform that contains a plurality of discrete tones having equal spectral spacings between sub-carriers. In addition, spectral notches are distributed periodically among the tones, for example, at every tenth sub-carrier position. The waveform is transmitted over the loop and measurements on the received signal are conducted.
In particular, the noise floor at missing sub-carrier spectral location is integrated and compared to the signal power of an adjacent existing sub-carrier within the MTPR test vector waveform. From these two values, a power ratio, termed the multi-tone power ratio (MTPR), is determined for various carrier locations across the spectrum, so that a maximum theoretical payload for the loop under test can be calculated. The MTPR takes into account various loop parameters governing ultimate modem performance simultaneously (e.g., loop loss, noise, distortion, interference, etc.) and also provides a figure of merit (FOM) for the loop under test.
In order to generate a figure of merit (FOM) for the loop, a test device, such as a data modem, may be placed at each end of the metallic line pair. Because the figure of merit obtained from MTPR analysis is channel-based, rather than equipment-based, it is independent of receiver implementation. The MTPR waveform may be transmitted over the loop in both directions and a respective FOM for each of the upstream and downstream channels computed. Then, prior to placing a new customer in service, the MTPR FOMs for a number of loops may be compared to bit rate information in a database, which correlates MTPR with a bit rate for a specific modem. Given the terminal equipment, MTPR FOMs, and the customer's service grade (data rate), one or more cable pairs that satisfy or exceed the FOM within an available cable bundle may be selected.
Advantageously, a single technician may control the operation of the MTPR test mechanism of the present invention using a feedback mechanism from a far end test device, such as a cellular data modem. Alternatively, a test waveform may be injected from a far end site in the return stream to the near end test site. Because generation of the MTPR waveform is relatively straightforward and the waveform is easy to process by conventional digital signal processing schemes, the test sets need not process the data in real time, thereby minimizing the required hardware in the test sets.
The basic functionality of a multitone power ratio (MTPR) for testing subscriber lines is a modification of frequency division multiplexing (FDM) analog technology for testing system linearity. This test consists of a transmitting and monitoring the channel response to a waveform that consists of a large number of frequency domain impulses, uniformly spaced over the bandwidth of interest, but having the feature that, periodically, a frequency impulse is missing, thereby giving the appearance of a spectral notch in the comb spectrum of the waveform.
The frequency impulses are mutually separated by a prescribed DMT carrier spacing, and each carrier of the comb has a controlled starting phase to constrain the peak to average ratio (PAR) associated with the channel waveform due to possible subcarrier instantaneous summation. Each tone's starting phase is adjusted to establish a desired PAR and the average signal level of the test vector is adjusted for a certain backoff level below full scale. This MTPR test waveform is transmitted over the twisted copper pair and the depth of the notches in the signal at the receive end of the loop is monitored with respect to the level of their adjacent carriers. Factors that contribute to raising the floor or filling-in the received signal notches include loop loss, noise, distortion, interference, etc.
The architecture of a digital signal processor-based test unit in accordance with the present invention includes a test vector buffer memory that stores a MTPR waveform pattern that has been digitally synthesized by a supervisory microcontroller. The contents of the test vector buffer memory are converted into analog format by a high speed digital-to-analog converter (DAC) and applied to a low output impedance, line driver, which amplifies the an
Nelson George Rodney
Roberts Richard D.
Squires Ronald S.
Allen Dyer Doppelt Milbrath & Gilchrist, P.A.
Barnie Rexford
Harris Corporation
Kuntz Curtis
LandOfFree
Testing of digital subscriber loops using multi-tone power... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Testing of digital subscriber loops using multi-tone power..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Testing of digital subscriber loops using multi-tone power... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2555434