Pulse or digital communications – Receivers – Particular pulse demodulator or detector
Reexamination Certificate
1998-07-17
2001-06-26
Pham, Chi (Department: 2631)
Pulse or digital communications
Receivers
Particular pulse demodulator or detector
C375S346000, C714S759000
Reexamination Certificate
active
06252917
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to decoding of received signals and particularly, although not exclusively to decoding of signals received over a satellite link, or a terrestrial microwave link.
BACKGROUND OF THE INVENTION
In microwave communications links, for example between a satellite and one or a plurality of earth stations, or between a radio base station of a terrestrial wireless communications system, transmitted signals become corrupted by noise due to a variety of factors, for example background noise, noise introduced through transmitter and receiver components, noise introduced through atmospheric transmission conditions, and interference introduced from other transmitters operating interfering carrier frequencies. In each case, for overlapping, but slightly different reasons, there is an advantage in being able to deal with a signal having as low a carrier to interference ratio, or as low a signal to noise ratio as possible. Toleration of a lower C/I or SNR ratio enables, use of lower power transmitter and receiver equipment, thereby in a satellite system, reducing the weight of equipment which needs to be launched, and reducing the power supply requirements of the equipment. In the case of terrestrial wireless systems, tolerating a lower carrier to interference ratio and signal to noise ratio enables mobile handsets having lower power transmitter apparatus, thereby reducing the size and power requirements of the handset, and possibly increasing the capacity of the overall terrestrial wireless system.
In order to recover transmitted signals which are received with a relatively low level of carrier to interference ratio and/or signal to noise ratio, it is known, prior to transmitting the signals to encode the signals with redundant bits of information according to known encoding algorithms. On receipt of the coded signals, known decoders are able to reconstruct parts of a signal which have been irretrievably corrupted due to noise or interference, and reconstruct the original signal from the redundant information contained in the coding. Such systems are known as forward error correction coded systems. Although the forward error correction codes add redundant bits to a signal to be transmitted, and effectively decrease the bandwidth available for data transmission, benefits are achieved in being able to decode signals which would otherwise be unable to be decoded, and improving the range, power consumption, and weight of transmitter and receiver equipment which can be used. The coding overhead of transmitting extra coding bits using forward error correction systems enable improvements in the effective bit error rate (BER) of a transmission link. In particular for satellite systems, the bit error rate of a satellite link is a limiting factor in performance of a satellite, since it has a direct impact on the power requirements of the launched transmitter/receiver equipment, and hence on the is cost of communications equipment. In the case of mobile wireless systems, the bit error rate of a link has a direct impact on the size and power requirements of the handsets which can be used.
A known forward error correction system comprises a convolutional coder and a Viterbi decoder, as is well known in the art, such as are available from Qualcom Incorporated. More recently, there have emerged in the prior art, a set of concatenated recursive codes known as “turbo codes”, which may be used for the same purpose as Viterbi forward error correction codes, i.e. improving the bit error rate of transmission links, but which have improved performance compared with Viterbi coded systems. Turbo codes offer an approximate 3 dB improvement over Viterbi codes, which has the practical implication of allowing an approximate halving of transmitter power for a transmission link having a same bit error rate, as compared with a Viterbi coded link. Parallel concatenated systematic recursive codes (otherwise called “turbo codes”) are described in “Near Shannon Limit Error-Correcting and Decoding: Turbo Codes (1)” by C Berrou, proceedings ICC May 1993. However, a problem with turbo code decoders compared with prior art forward error correction decoders is that turbo code decoders require increased data processing power. Typically, a known turbo code requires ten times as much processing power as a known Viterbi forward error correction decoder. Processing powers of the order of Giga instructions per second may be required. Digital signal processing apparatus having an ability to handle these volumes of instructions may comprise commercially available digital signal processing chip sets, or alternatively custom made decoder chip sets. Thus, the use of turbo codes instead of Viterbi codes, whilst potentially improving the transmitter/receiver equipment with respect to required transmission power and size of equipment, incurs the penalty of requiring around ten times as much signal processing power as Viterbi decoders. In the case of satellite systems, in addition to the increased cost of the extra signal processing power, the increased power consumption requirement is of concern. In the case of terrestrial wireless base stations, which operate in a cost competitive market, there is the disadvantage of the increased cost of the extra DSP chip sets required for providing the required signal processing power.
SUMMARY OF THE INVENTION
One object of the present invention is to obtain the advantages of turbo code decoders applied to a receiver equipment, for example a satellite equipment or a base station of a terrestrial wireless system, whilst minimizing the power supply requirement and additional cost incurred by the provision of increased digital signal processing power.
Another object of the invention is to optimize the use of signal processing power applied for decoding of turbo code encoded signals, by taking advantage of a statistical analysis of corruption of signals received from a plurality of transmitters.
A further object of the present invention is to reduce an overall required processing power for decoding a plurality of turbo encoded receive signals from a plurality of channels, each channel characterized by having its own corresponding respective signal to noise ratio and/or carrier to interference ratio, by optimizing the use of signal processing capabilities.
According to one aspect of the present invention there is provided a decode apparatus for decoding a plurality of encoded messages each having an associated corruption level, said decoding means comprising:
analyzer means for determining a level of corruption of each said message and generating data describing a signal processing requirement for decoding each said message;
at least one decoder for decoding said encoded messages; and
scheduler means for assigning said messages to individual ones of said at least one decoder depending on said signal processing requirement data generated by said analyzer means.
Preferably, said analyzer means comprises:
means for determining a carrier to noise ratio metric of a said message;
means for assigning a number of turbo code decoding iterations to said message.
Preferably said means for determining a number of turbo code iterations comprises a data storage means containing data describing carrier to noise ratio and number of decoding iterations and a relationship between said carrier to noise ratio and said number of iterations.
In the best mode presented, each said decode processor sends signals to said scheduler concerning a status of said decoder. A said status signal may comprise data describing a current number of iterations performed on a specified message packet, or may comprise data describing a convergence rate of bit error rate of a said message.
Preferably, the apparatus further comprises a plurality of buffers, for storing said messages at said decoders, wherein said scheduler sends said message packets to selected individual ones of said buffers for storage prior to decoding.
According to a second aspect of the present invention, there is provided a method
Bayard Emmanuel
Lee Mann Smith McWilliams Sweeney & Ohlson
Nortel Networks Limited
Pham Chi
LandOfFree
Statistically multiplexed turbo code decoder does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Statistically multiplexed turbo code decoder, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Statistically multiplexed turbo code decoder will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2463694