Pulse or digital communications – Transmitters – Antinoise or distortion
Reexamination Certificate
2000-02-17
2003-07-22
Corrielus, Jean B. (Department: 2631)
Pulse or digital communications
Transmitters
Antinoise or distortion
C375S285000, C708S300000
Reexamination Certificate
active
06597746
ABSTRACT:
TECHNICAL FIELD
The present invention generally relates to communication systems, and more particularly, to a system and method for peak to average power ratio reduction in communication systems using time domain techniques for data signal transmission. The present invention is particularly suited to transmitters that use pulse amplitude modulation (PAM) and Tomlinson preceding.
BACKGROUND OF THE INVENTION
In recent years, telephone communication systems have expanded from traditional plain old telephone system (POTS) communications to include high-speed data communications as well. As is known, POTS communications include the transmission of voice information, control signals, PSTN (public switched telephone network) information, as well as, information from ancillary equipment in analog form (i.e. computer modems and facsimile machines) that is transmitted in the POTS bandwidth.
Prompted largely by the desire of businesses to reliably transfer information over a broadband network, telecommunications service providers have employed DSL systems to provide a plethora of interactive multi-media digital signals over the same existing twisted-pair copper lines. The provision of asynchronous digital subscriber line (ADSL) systems, using discrete multi-tone (DMT) line coding, to customer premises has proliferated over recent years due to the increasing demand for high speed Internet access.
ADSL systems can be designed to operate over the same copper lines with either the POTS or a Basic Access Integrated Service Digital Network (ISDN-BA) service. ADSL systems are designed to transmit significantly more data in the downstream direction, that is, from the telecommunication service provider to the remote customer, rather than sharing the bandwidth as in a dual-duplex communication system. The ADSL configuration permits the concurrent transmission of multiple digital signals from the telecommunications service provider to a remote location, while providing an adequate bandwidth for the limited capacity required for data transmissions originating at the remote site.
High-Bit-Rate digital subscriber lines (HDSL) can be considered an extension of the DSL and ADSL systems. HDSL was developed as a dual-duplex repeaterless T1 technology. HDSL was designed to serve non-load telephone subscriber loops meeting the Carrier Serving Area (CSA) guidelines. Unlike ADSL systems, HDSL systems are designed to take advantage of the full data transmission capacity of a cable pair in both directions. It is important to note that ADSL systems use DMT modulation for data transmission. Conversely, HDSL systems use PAM, a broadband carrierless method, for data transmission.
In a basic HDSL DMT implementation, a HDSL transmission unit—central office (HTU-C) is configured to modulate and convert digital data for analog transmission from a central office downstream to a remotely located HDSL transmission unit—remote (HTU-R). Each end of the HDSL link configured at the end points of a CSA loop is provided a similarly configured HDSL transmission unit. It is important to note that the central office and remote designators for the two HDSL transmission units is for descriptive purposes only as both units are interchangeable with the remote (HTU-R) unit configured to receive the analog transmission sent from the HTU-C from the telephone line. The HTU-R demodulates the signals and applies error correction before delivering each of the reconstructed digital data signals to its intended device. Concurrently, the HTU-R transmits data from the remote location back to the HTU-C.
In HDSL, incoming data bits are first encoded in PAM symbols, then subjected to Tomlinson preceding to better match the exact characteristics of the line. The encoded and modified PAM symbols are then passed through a bandwidth shaping filter.
Constraints on the average transmitted power in communication systems vary according to the specific application. In DSL applications, constraints are imposed to limit the amount of interference, or crosstalk, radiated into neighboring receivers. Because crosstalk is frequency dependent, the constraint on average power may take the form of a spectral mask that specifies the magnitude of allowable transmitted power as a function of frequency. For example, crosstalk in POTS communication systems is generally caused by capacitive coupling and increases as a function of frequency. Consequently, to reduce the amount of crosstalk generated at a particular transmitter, the pulse shaping filter generally attenuates higher frequencies more than lower frequencies.
In addition to constraints on the average transmitted power, a peak power constraint is often imposed as well. This peak power constraint is important for the following reasons:
1. The dynamic range of the transmitter is limited by its line driver and its digital to analog converter (DAC). In particular, saturation of the line driver will “clip” the transmitted waveform. Similarly, signal magnitudes that exceed the dynamic range of the DAC will also “clip” the transmitted waveform.
2. Rapid fades can severely distort signals with high peak to average power.
3. The transmitted signal may be subject to other non-clipping induced non-linearities.
4. A major component of the total power consumption within a transmitter is consumed within the line driver, as line driver power consumption overwhelms the power consumed by digital signal processing. As a result, PAR reduction is important to decrease power consumption in the output line driver.
Peak to Average Power Ratio
The peak to average power ratio is defined as the ratio of the magnitude of the instantaneous peak power to the time averaged value of the power. In HDSL
2
applications, the PAR is defined by the digital transmit shaping filter. The shaping filter is added in the modulation stage within the HTUs to shape the spectrum of the transmitted signal. The effect of a digital finite impulse response (FIR), often used as the shaping filter, can be analyzed by calculating the mean, the variance, and the peak value of the FIR filter output using random data as the FIR filter input. It can be shown that the worst case PAR is uniquely defined by the filter coefficients, H
i
, as
PAR
=
Σ
&LeftBracketingBar;
H
i
&RightBracketingBar;
Σ
H
i
2
Eq
.
1
This theoretical maximum is in fact almost never achieved, so it is typical to define the PAR at a given level of significance by defining an effective peak value Pe(s). Pe(s) is such that the probability that the absolute value of the voltage exceeds Pe is 10
−5
. It is also usual to present the negative cumulative probability of the absolute voltage Q(v), so that Q(Pe(s))=10
−5
. Once Pe(s) is known, PAR(s) is simply Pe(s)/RMS.
Frequency Domain Techniques
Two frequency domain techniques have been proposed to reduce PAR in transmitted signals in communications systems, see e.g. J. Tellado and J. M. Cioffi. “Further Results on Peak-to-Average Ratio Reduction”, ANSI T1E1.4 contribution number 98-252, San Antonio, Tex., Aug. 1998 and ETSI/ANSI TM6 contribution number TD15, Vienna, Austria, Sep. 21-25, 1998. The proposed techniques are repeated for each frame, or in other words, for each IFFT operation. The techniques may be summarized as follows:
1. Reserving a set of tones whose amplitudes are calculated either by linear programming or a simpler iterative procedure to minimize the PAR. It is significant to note that this proposal reduces the data rate which is unacceptable under HDSL communication standards.
2. Modifying the modulation scheme by permitting more than the minimal set of points in each QAM constellation. Typically, a constellation is replicated at integer multiples of a translation vector, so that one input data symbol may be mapped at different positions in the QAM plane. Under this proposal, constellation positions are selected to minimize the PAR either via an iterative or a linear programming procedure. It is significant to note that this proposal does not reduce the data rate, but resul
Amrany Daniel
Delvaux Marc
Gut Richard
Scholtz William H.
Corrielus Jean B.
Globespanvirata, Inc.
Thomas Kayden Horstemeyer & Risley
LandOfFree
System and method for peak to average power ratio reduction does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for peak to average power ratio reduction, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for peak to average power ratio reduction will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3095415