Voting system for improving the performance of single-user decoders within an iterative multi-user detection system
View Patent ↗A system is presented that provides real-time performance for iterative multi-user detectors, such as Turbo MUDs, which are used to separate simultaneous transmissions on the same frequency, by permitting the MUD to use a less computationally intense, fast-processing algorithm and to correct for errors caused by the fast processing. In order to reduce the errors, a voting system is coupled to the output of the multi-user detector within the iterative system. The voting system provides confidence values on a bit-by-bit basis for the estimates made by the multi-user detector, with the confidence values then being utilized as soft inputs to a bank of conventional single-user decoders.
1. In a system to separate incoming interfering signals, apparatus for improving the performance of a multi-user detector coupled to the incoming signal and coupled to a single-user detector in which said multi-user detector provides as an output an ordered number of best guesses as to what a particular incoming bit is, comprising:
a voting unit connected between said multi-user detector and said single-user detector for averaging for one user the best guesses for said incoming bit and for applying said average to said single-user detector, thus to provide said single-user detector with a soft input for improving the performance thereof.
2. The apparatus of claim 1 , wherein said single-user detector provides on a bit-by-bit basis a confidence value and further including means for coupling said confidence value to said multi-user detector to allow said multi-user detector to recompute estimates of said incoming signal from which said best guesses are derived.
3. The apparatus of claim 1 , wherein said multi-user detector is an iterative multi-user detector.
4. A method for reducing the computational complexity of an iterative multi-user detection system in which an incoming data stream having a number of bits from different users is applied to a multi-user detector coupled to a single-user detector, comprising the steps of:
providing the multi-user detector with a fast algorithm which examines less than the total number of possibilities for each of the bits of data in the incoming data stream, the algorithm resulting in prioritized best guesses as to what an incoming bit is;
for a given bit, averaging a predetermined number of best guesses for a predetermined user to provide a number corresponding to the average of the best guesses; and,
coupling the number to the single-user detector as a soft input thereto, thus to correct for inaccuracies in the output of the single-user detector due to examining less than the total number of possibilities for an incoming bit, whereby real-time performance can be achieved through the use of said fast algorithm.
5. The method of claim 4 , wherein the single-user detector has as an output a confidence value for the bit, with the confidence value being derived from the soft input to the single-user detector, and further including the step of coupling the confidence value to the multi-user detector to improve the accuracy of the best guesses therefrom.
6. In an iterative multi-user detection system, a method for increasing the number of users accommodated by a multi-user detector coupled to a single-user detector in which a number of best guesses of what a particular bit is for a given user is established as the output of the multi-user detector, comprising the steps of:
averaging the best guesses for a given user for a given bit; and,
applying the average as a soft input to the single-user detector, whereby less computationally complex algorithms may be used in the multi-user detector to increase the number of users that can be accommodated in a given time interval.
7. The method of claim 6 , wherein the multi-user detector utilizes algorithms which for each bit of incoming data analyzes less than the total number of possibilities so as to decrease processing time, such that the increased number of users is accommodated in real time.