Pulse or digital communications – Transmitters – Antinoise or distortion
Reexamination Certificate
1999-01-28
2001-12-11
Chin, Stephen (Department: 2734)
Pulse or digital communications
Transmitters
Antinoise or distortion
Reexamination Certificate
active
06330288
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to communications and, more particularly, to wireless systems.
BACKGROUND OF THE INVENTION
A number of present day wireless data networks transmit data between their base stations and mobile units in a series of fixed-length, physical layer blocks (hereafter simply referred to as a “block”. Each block comprises a number of payload bits and parity bits, which are generated by a forward error correction code or coding scheme. In general adding more parity bits per block increases the number of airlink errors that can be detected and corrected.
However, having a large number of parity bits per block has an obvious downside—it decreases the number of available payload bits. As a result, wireless data networks such as General Packet Radio Service (GPRS) networks use more than one coding scheme to transmit data over the airlink. When the received signal-to-noise ratio (SNR) is high, airlink bit error rates are low. As a result, a coding scheme with a small number of parity bits may offer adequate protection. At low SNR, “stronger” codes may be needed to protect data against airlink errors since stronger codes add more parity bits to each block.
The error performance of a cellular airlink varies as mobile units move within a cell. To make most efficient use of the airlink, coding schemes are dynamically selected in response to changes in the quality of the cellular airlink. Current coding scheme selection algorithms are a function of a channel quality metric (CQM). A CQM is, e.g., a function of soft bit or soft symbol information, block or bit error rate estimates, received signal strength, and/or the carrier-to-interference ratio (C/I).
For example, with respect to C/I, for a given coding scheme, the portion of transmissions which result in block errors decreases as the C/I value of a received signal increases. Using simulation or analytical techniques, it is possible to estimate the rate at which payload bits are carried over the airlink as a function of C/f. Plots of throughput-versus-C/I curves for all coding schemes available in a wireless data network show at which values of C/I it is advantageous to switch coding schemes. An illustrative throughput-versus-C/I plot is shown in
FIG. 1
for three coding schemes I, II, and III, where coding scheme I is the strongest and coding scheme III is the weakest. C/I switch points are often hard-coded at the transmitter. Based on C/I measurements, a transmitter switches to a coding scheme offering the best performance (highest throughput/lowest delays) at the estimated received C/I level.
Similar selection techniques are used in other systems, e.g. Enhanced GPRS and North American TDMA Packet Data Channel. For example, instead of, or in addition to, the channel coding rate, the modulation scheme (signal constellation size) is varied to achieve a similar tradeoff for variable C/I. Thus, in the North American TDMA Packet Data Channel, the channel coding rate is fixed (at 5/6), while the modulation scheme is switched between 4-level (DQPSK) and 8-level (coherent 8 PSK) with a possible extension to 16-level (not yet specified). In this case, schematically the same performance tradeoffs as shown in
FIG. 1
apply, if we now refer to the three formats I, II, and III as three modulation schemes. (As such, as used herein, the term “coding/modulation scheme” refers to either a coding scheme, modulation scheme, or coding and modulation scheme used to transmit a signal.)
SUMMARY OF THE INVENTION
We have observed that selecting a coding/modulation scheme based on a CQM alone (e.g., C/I) does not take advantage of any spare bandwidth in a block. As such, data protection is not being maximized for a particular block. For example, payload bits are always sent in an integral number of blocks. A transmitter wishing to send one byte of payload will have to send one block regardless of which coding/modulation scheme is used. However, if selection of a coding/modulation scheme is based on CQM alone, a coding/modulation scheme with the fewest number of parity bits may be used notwithstanding the fact that spare bandwidth is available in the block. Consequently, using the code with the smallest number of parity bits both offers the lowest level of data protection and inefficient use of the airlink.
Therefore, and in accordance with the invention, a coding/modulation selection scheme takes into account CQM measurements and the volume of payload bits to be sent in a block. As a result, the strongest coding/modulation scheme for a given volume of payload bits is used.
In an embodiment of the invention, a transmitter uses one of k coding/modulation schemes in transmitting data over a wireless data network. The transmitter initially selects a coding/modulation scheme, C, as a function of C/I measurements. The transmitter then calculates the number of blocks, B, required to transmit a number of data packets, D, using the coding/modulation scheme C. In addition, the transmitter calculates the number of blocks required to transmit the number of data packets, D, for each coding/modulation scheme that is stronger than the selected coding/modulation scheme C. The transmitter finally selects that coding/modulation scheme that results in transmitting the number of data packets D in B blocks using the strongest coding/modulation scheme. As a result, each block is transmitted using the strongest coding/modulation scheme available. Thus, there will be fewer retransmissions, packet transmission delays will be lower and less variable, and achievable maximum throughputs will be higher.
REFERENCES:
patent: 4744083 (1988-05-01), O'Neill et al.
patent: 5577087 (1996-11-01), Furuya
patent: 5982813 (1999-11-01), Dutta et al.
patent: 6122293 (2000-09-01), Frodigh et al.
Budka Kenneth Carl
Nanda Sanjiv
Schefczik Hans-Peter
Chin Stephen
Jiang Lenny
Lucent Technologies - Inc.
Opalach Joseph J.
LandOfFree
Coding/modulation scheme selection technique does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Coding/modulation scheme selection technique, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Coding/modulation scheme selection technique will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2579183