Multiplex communications – Communication over free space – Combining or distributing information via time channels
Reexamination Certificate
2002-05-29
2003-10-07
Hsu, Alpus H. (Department: 2665)
Multiplex communications
Communication over free space
Combining or distributing information via time channels
C370S332000, C370S230000, C714S701000, C455S069000, C455S067150
Reexamination Certificate
active
06631127
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to communication systems and in particular to dynamically selecting an automatic repeat request method for the retransmission of a data frame.
BACKGROUND OF THE INVENTION
In a Hybrid Automatic Repeat Request (H-ARQ) system, retransmissions of data frames are used to obtain a desired level of data quality at the receiving end of a communication system. However, excessive retransmissions can lead to a reduction in system throughput. In systems that employ Adaptive Modulation and Coding (AMC), the selection of AMC is based on channel conditions at the time the selection is made. Taking into account measurement error, feedback delay, and mobile station velocity, the AMC selection based on then current channel conditions is often not the best choice at the time that the data frame is transmitted. The system then relies on ARQ methods to correct situations where the AMC was poorly chosen. To reduce ARQ throughput loss, an adaptive ARQ system is necessary to adjust the ARQ method to best match actual channel conditions.
Three methods of H-ARQ will now be described using a channel encoder with a rate of one fifth (i.e., for every bit input into the encoder, five bits are produced) for illustration, as shown in FIG.
1
.
FIGS. 2-4
depict the over the air slot format of data from the encoder operating at a rate of ½ coding of the initial transmission. Referring to
FIGS. 2 and 3
, H-ARQ utilizing a repeat and max ratio combine (Chase) technique is shown. Chase combining (as discussed herein) is based on an IEEE paper: D. Chase, “Code Combining—A Maximum-Likelihood Decoding Approach for Combining an Arbitrary Number of Noisy Packets,” IEEE Trans on Communications, Vol. 33, pp. 385-393, May 1985. The basic idea is to retransmit data packets that are not received (i.e., not decoded properly). The retransmitted packets are weighted by their corresponding signal amplitude-to-noise power ratios and are added to the originally transmitted packets—in other words, they are max ratio combined (MRC). Adding the retransmitted packets to the originally transmitted packets improves the signal-to-noise ratio (signal-to-noise ratios are summed due to MRC). Thus, packets that are not decoded at the first trial will eventually decode after several repeats, since the SNR continually increases with repeats.
Referring to
FIGS. 2 and 3
, H-ARQ utilizing the Chase combining technique operates as follows: outputs from the g
1
and g
3
polynomials are punctured (denoted by part
1
) and transmitted along with the systematic bits as shown in FIG.
2
. Note that all packet retransmissions are identical to the first packet. Thus, they can be max ratio combined or alternatively they can be decoded without any additional information (i.e., they are self-decodable). Referring to
FIGS. 1 and 3
, a Partial Incremental Redundancy (IR) combining technique is illustrated. For this case, the first transmitted packet is identical to that shown in FIG.
3
. However, the first re-transmission includes the second half of the g
1
/g
3
punctured parity bits (denoted as part
2
) that were not transmitted with the first packet. Max ratio combining can be exploited on the systematic bits. The partial IR method provides a higher coding gain, especially in fast fading channels, for example, in vehicular channels for which perfect signal-to-noise (SNR) tracking is impossible. Also, it should be noted that this scheme is self-decodable as in FIG.
2
. Referring to
FIGS. 1 and 4
, a Full IR combining technique is illustrated. For this case, the first re-transmission does not repeat the systematic bits. Instead, parity bits are continuously sent in each next transmission. This process then repeats. The full IR method provides a higher IR gain relative to the partial IR method, but re-transmitted packets are not self-decodable (except when the systematic bits are again transmitted).
When the AMC system tracks properly, as in the case of a slow moving mobile station, there are many instances when only a small SNR is required in the first retransmission to correctly receive a given frame of data. This assumes that the two transmissions are combined using Chase combining or Full IR. By providing a small SNR, saved resources such as transmit power and Walsh codes, can be utilized elsewhere to increase system throughput. The Full IR combining scheme provides significant channel coding gain with respect to Chase combining. However, when an initial transmission is received with a degraded SNR and the second transmission is received at or above threshold, Full IR performance degrades relative to Chase combining. In such a case, the Chase combining method works properly. However, the Full IR method requires further retransmissions because the critical systematic bits were lost in the initial transmission. Thus, the second transmission is not ‘self-decodable’ and relies heavily on information received in the first transmission.
A method for selecting Chase, Partial Chase, and Partial IR combining is known. However, the method does not utilize Full IR, which provides the best performance. Given a combining technique, the prior art method determines how many retransmissions are necessary via the accumulated energy in the user equipment (UE) communicated to the base station in a two bit soft acknowledgment. Thus, the prior art method has the disadvantages of loss in throughput relative to full IR and the expense of extra ACK/NACK bits transferred in the uplink. In addition, the prior art does not provide a method of selecting between ARQ combining methods based on the SNR of the communication channel at the UE.
Thus, there is a need for an improved method of dynamically selecting the best H-ARQ method for transmitting data frames.
REFERENCES:
patent: 5229767 (1993-07-01), Winter et al.
patent: 5822315 (1998-10-01), de Seze et al.
patent: 6247150 (2001-06-01), Niemela
patent: 6308294 (2001-10-01), Ghosh et al.
patent: 6456605 (2002-09-01), Laakso et al.
patent: 2001/0034209 (2001-10-01), Tong et al.
patent: 2002/0080719 (2002-06-01), Parkvall et al.
patent: 2003/0081576 (2003-05-01), Kim et al.
patent: 2003/0095532 (2003-05-01), Kim et al.
patent: 2003/0118031 (2003-06-01), Classon et al.
Martins and Alves, “ARQ Protocols with Adaptive Block Size Perform Better Over a Wide Range of Bit Error Rates”, Jun. 6, 1990, IEEE Transactions on Communications, vol. 38, No. 6, pp. 737-739.
Ahmed Mansoor
Laird Kevin M.
Rohani Kamyar
Hsu Alpus H.
Molinari Michael J
Motorola Inc
Williams Lalita P.
LandOfFree
Apparatus and method for dynamically selecting an ARQ method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for dynamically selecting an ARQ method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for dynamically selecting an ARQ method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3142118