Multiplex communications – Communication over free space – Combining or distributing information via time channels
Reexamination Certificate
2001-08-22
2004-05-18
Jung, Min (Department: 2663)
Multiplex communications
Communication over free space
Combining or distributing information via time channels
C370S465000
Reexamination Certificate
active
06738370
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to wireless communication, such as provided by systems as specified in 3GPP (Third Generation Partnership Project) Wideband Code Division Multiple Access (WCDMA) release 5, High Speed Downlink Packet Access (HSDPA), but also as provided by other kinds of wireless communications systems. More particularly, the present invention relates to retransmitting via a wireless communication system a portion of a signal when the portion is received with an error; the invention is of use in applications in which both forward error correction and automatic retransmission request are implemented.
BACKGROUND OF THE INVENTION
To provide for higher data throughput in wireless communication systems, adaptive modulation and coding schemes (MCSs) are used in which both the modulation complexity and (channel) coding complexity are varied in response to changing channel conditions. In some communication systems such as systems implementing HSDPA (high speed downlink packet access), the number of channelization codes (and so the number of channels) can also be varied in response to changing channel conditions. Modulation complexity and channel coding are changed based on rather rapidly changing channel conditions, whereas the number of channels are varied on the basis of a longer-term average, and depending on how much data is to be transmitted. Varying modulation complexity means varying the number of bits that are communicated per symbol (a given modulation complexity provides a set or constellation of symbols, with each symbol used to convey a bit string, the greater the number of symbols in the constellation, the longer the bit string conveyed by each symbol). Varying the coding complexity means, for example, varying the amount of redundancy included in forward error correcting the data to be transmitted. Varying the number of channelization codes means changing the number of channels multiplexed together by use of a code tree (ensuring that all channels remain orthogonal even while the number of channels is varied). The modulation complexity and the number of channelization codes can be adaptively optimized for instance as shown in copending, commonly-owned U.S. Provisional Application Serial No. 60/301,078 filed Jun. 26, 2001. Because the conditions of a wireless communication channel tend to change more often and more substantially than the conditions of a hard-wired channel, errors in communication are more likely. To address the problem of higher error rates (both bit error rates and symbol error-rates), wireless communication systems have implemented various coping mechanisms. One coping mechanism for non-real time data is so-called automatic repeat (retransmission) request (ARQ) protocol, whereby, if a received symbol is determined to have an error, the receiving system automatically requests retransmission of the symbol.
The Problem Addressed by the Invention
Higher-order modulation complexities (higher-order compared to binary systems) include what are generally known as N-QAM (quadrature amplitude modulation) systems (such as e.g. 16-QAM and 64-QAM). N-QAM systems (and other higher-order complexities) convey multiple bits per transmitted symbol. It is inherent in any amplitude modulation system involving more than two symbols (including any N-QAM system for N greater than 2) that the symbol error probabilities are not all the same, i.e. the probability that a receiver will conclude that a symbol was received other than the actually transmitted symbol is different for different symbols. (See e.g.
Introduction to Communication Systems
, Third Edition, by Ferrel G. Stremler, Addison Wesley Publishing Co., 1990, section 9.5.) Depending on how bits are assigned to the modulation symbols of a coding scheme, the bit error probabilities may vary too, i.e. the probability of a receiver concluding that a 1 was received when a 0 was transmitted (i.e. the bit error probability for a zero) may be different than the probability of a receiver concluding that a 0 was received when a 1 was transmitted.
For instance, in the symbol constellation diagram provided as
FIG. 1
, showing the constellation currently proposed for high speed downlink packet access (HSDPA), it can be seen that the first two bits are the same for each of the four symbols in any of the four quadrants; in the first quadrant, for example, the first two bits for each symbol are 00. On the other hand, in all of the four quadrants, the last two bits are always 11 in the corner symbols, while they are 00 in the innermost symbols. Consequently, a corner symbol being incorrectly mistaken for an innermost symbol occurs with a different frequency than an innermost symbol being mistaken for a corner symbol. Therefore, a bit having a value of 1 being incorrectly detected as having a value of 0 occurs with a different frequency than a bit having a value 0 being incorrectly detected as having a value of 1. Thus, the bit error probabilities for this constellation are different for 0′s and 1′s.
The notations i
1
, i
2
, q
1
, q
2
in
FIG. 1
represent the bits in the group constituting a modulation symbol; the bits are in the order i
1
q
1
i
2
q
2
. A bar under or beside one of the notations (either i
1
, i
2
, q
1
, or q
2
) indicates where in the constellation diagram the bit indicated by the notation has the value “1” (i.e. the bar indicates all or part of the set of constellation points from which the modulation symbol is chosen if a particular bit is 1). For instance, if the bit q
1
=1, then the symbol must be chosen from the set of points indicated by the bar beside q
1
in
FIG. 1
, and if q
1
=0, then the symbol must be chosen from the complementary set of points.
A radio receiver has a limited dynamic range. If the modulation symbols have different amplitudes (of the same sign or phase), which indeed is the case in N-QAM modulation, a radio receiver will respond differently to the different symbols on account of their different amplitudes. For instance, the highest amplitude symbols might saturate the receiver, so that the receiver clips those symbols. On the other hand, because the smallest amplitude symbols might be smaller than the smallest quantization level, those symbols might be interpreted by a receiver as having a zero amplitude in the A/D-conversion. Clipping and zeroing are particularly noticeable in a fading channel, where the amplitude of the signal might vary from +10 dB (because of multipath constructive interference) to −40 dB compared to an unfaded signal.
In order to provide both a high data rate (in the downlink) and also reliability, for HSDPA it is proposed that what is termed H-ARQ (for hybrid automatic repeat requests) protocols be employed (at least for data transmission). In H-ARQ, a data packet that has been determined to be in error is retransmitted (when the receiver detects an error in the packet, based for example on some form of simple parity check). The retransmitted packet is combined with the original packet prior to forward error correction (FEC) decoding (in the user terminal receiver, prior to decoding the convolutional or Turbo code), thereby increasing the reliability of the (downlink) transmission.
In a system using either ordinary automatic repeat request (ARQ) or H-ARQ, if the symbols constituting the higher-order modulation system are generated identically in the retransmission and the original transmission, the bit error probabilities in the retransmission are identical to the bit error probabilities in the original transmission. The probability of an error recurring is therefore the same with each retransmission, all other things being the same as when the error first occurred.
Prior Art Solutions
Several H-ARQ techniques have been proposed in HSDPA to improve the likelihood that in case of an error in a packet, a retransmission of the packet will be error-free. The most straightforward is Chase combining, where the same data packet is retransmitted a number of times, and prior to decoding, the repeated
Jung Min
Lee Andy
Nokia Corporation
Ware Fressola Van Der Sluys & Adolphson LLP
LandOfFree
Method and apparatus implementing retransmission in a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus implementing retransmission in a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus implementing retransmission in a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3242300