Pulse or digital communications – Systems using alternating or pulsating current – Plural channels for transmission of a single pulse train
Reexamination Certificate
2000-02-17
2003-09-30
Ghayour, Mohammad H. (Department: 2634)
Pulse or digital communications
Systems using alternating or pulsating current
Plural channels for transmission of a single pulse train
Reexamination Certificate
active
06628726
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to digital transmission and in particular to concepts for reducing the out-of-band radiation of digital transmitters.
BACKGROUND OF THE INVENTION
In general a digital transmitter comprises an information source, which e.g. might be an MPEG audio encoder for digital radio or an MPEG video encoder for digital television. The output data of the source, which exist in the form of a digital bit stream, are then typically encoded with the aid of a channel encoder so as to introduce the bit stream redundancy which enables transmission errors to be overcome in the receiver. Following this the channel-coded digital bit stream is fed into a so-called “interleaver”, which alters the sequence of the data according to an algorithm which is known to the receiver so that so-called burst errors in the transmission channel do not lead to the loss of a fairly large contiguous part of the message but only to smaller short losses which are spread over a larger time interval. The interleaved bit stream at the output of the interleaver is then imaged in the form of modulation symbols With the aid of a so-called mapper according to the type of modulation used.
If modulation is not employed, and the digital bit stream is effectively transmitted directly without modulation, the mapper and also the modulator which succeeds it can be dispensed with.
If a modulation method is employed, however, e.g. a multicarrier modulation method, the mapper is succeeded by a modulator, which modulates the modulation symbols onto the carrier.
Recently the OFDM method for digital radio applications has become increasingly popular. In this method a plurality of subcarriers is employed onto which the modulation symbols formed by the mapper are modulated. Here the modulation method is an inverse discrete Fourier transform, which, as is known, is used to generate a discrete time signal from the many modulated carriers. The discrete, normally complex time signal exists in the form of complex sampled values or “samples”, which are then fed into an interpolation low-pass in order to remove the periodically repeating spectral contributions. The signal at the output of the interpolation low-pass is typically modulated onto a HF carrier frequency by means of a complex IQ modulator to obtain a HF signal which is fed into a transmitter amplifier which feeds the amplified signal to an antenna which radiates the signal.
In radio applications tube amplifiers such as klystrons or travelling wave tubes are typically used in view of the high power requirements at the output of the amplifier. If smaller powers are acceptable, as in mobile radio e.g., where there is a dense network of transmitters, transistor amplifiers can also be used.
A property which transistor amplifiers and tube amplifiers have in common is that they are linear only over a certain input power range and have an output power curve which falls off at larger input powers and finally levels off at a constant value when the amplifier reaches complete saturation. Put another way, since its characteristic is non-linear, the amplifier causes non-linear distortions of the input signal at higher input powers.
In situations where a certain frequency band has been allocated for a particular transmission application, e.g. through a public licensing authority, regulations stipulate that the transmission signal for the particular licensed transmission application may only carry power within a prescribed band and that there must be no, or very little, power outside the allotted band. The power outside the allotted band is also called out-of-band radiation.
The non-linear characteristic of the amplifier at higher input powers leads, as already mentioned, to non-linear distortions, whose nature is such that the amplifier generates higher harmonics which no longer lie within the allotted band but outside the band and which can be measured as out-to-band radiation.
It is known that these non-linear distortions produce a relatively white spectrum. If the input signal in the amplifier is also band-limited, which is to be expected e.g. for OFDM modulation, the output signal will then possess power outside this band.
To avoid this, i.e. to comply with the licenser's regulations for the tolerance schemes specified for the frequency bands, i.e. how much out-of-band radiation outside the allotted band is still acceptable, the input voltage into the amplifier should exceed the maximum input voltage for a distortionless amplification as seldom as possible and preferably not at all. In other words this means that the maximum input voltage actually occurring should be as small as possible. If the maximum input voltage is always smaller than the maximum voltage at which the amplifier still operates linearly or only amplifies non-linearly to such a small extent that its out-of-band radiation lies below the permitted value, distortions which lead to an out-of-band radiation which exceeds the permitted value never occur.
A disadvantage of the OFDM method described at the outset is the typically large ratio of peak power to average power, also referred to as PAR (PAR=Peak to Average Power Ratio). Large peaks in the time signal, i.e. in the modulation symbol after the IDFT, can occur if the occupation of the carriers happens to be so unfavourable that e.g. all the 256 OFDM subcarriers overlap constructively at a particular instant. In this case there will be a large signal peak which may easily lie 10 to 20 dB above the average signal power. To comply with the permitted out-of-band radiation despite this, a high power margin, also referred to as “power back-off”, is typically maintained in the transmitter amplifier. Put another way, the amplifier is operated at a working point which is fixed low enough that even a high power peak still falls within the linear range of the amplifier.
This operating mode of an amplifier is an extremely inefficient operating mode in which the amplifier, while requiring considerable supply power, only delivers a relatively small output power. The requirement of low out-of-band radiation in conjunction with high peak values in the time signal, which occur not only for an OFDM modulation but also e.g for a single-carrier method through a pulse shaper, i.e. which can occur generally when filtering, means that expensive amplifiers are needed, which must be operated with a high power margin and which have a low efficiency. However, for smaller battery-operated systems in particular the efficiency is also a point of ever increasing concern, especially with regard to mobile radio and the accumulators with limited storage capacity which are employed there.
DESCRIPTION OF PRIOR ART
WO 98/10567 relates to a method for reducing the peak value factor for digital transmission methods. The basic idea consists in already taking measures on the digital side to ensure that signal peaks which are too high do not occur in the time signal, so that smaller power margins suffice for the transmitter amplifier without resulting in out-of-band radiation which exceeds the permitted limit. The known concept is generally referred to as selected mapping. Selected mapping or SLM really means just that, from a message for transmission, i.e. an information word or more generally speaking a vector of data bits, in some way or other U different possible signals are generated, which can also be termed representatives or candidate transmit sequences. Not all of these signals are transmitted, however. Instead a special signal is selected as the transmission signal. In particular, each transmission signal has a peak value, which is measured. The candidate transmission signal with the lowest peak value is then chosen as the actual transmission signal and is transmitted.
At the receiver the object is now to discover a) what the message is and b) which of the U possible representatives was transmitted per message. There are two possible ways in which the receiver can do this. First, by means of side information, which is communicated to the receiver
Fraunhofer-Gesellschaft zur Foerderung, der Angewandten Forschun
Ghayour Mohammad H.
Glenn Michael A.
Glenn Patent Group
Wong Kirk D.
LandOfFree
Device and method for generating a transmit sequence and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Device and method for generating a transmit sequence and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Device and method for generating a transmit sequence and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3010745