Lucent Technologies Inc.

Alcatel-Lucent USA Inc.

David Garrett

Bertrand Martyn Hochwald

Bing Xu

Stephan Ten Brink

Linda Mary Davis

Geoff Knagge

Christopher J Nicol

A decoding scheme for determining the information within a coded transmission includes a spherical decoder candidate search that reduces the number of candidate symbol combinations from a much larger number of possibly transmitted combinations. In one example, the spherical decoder uses an upper triangular matrix representation of the channel matrix as part of a recursive, cumulative cost function that provides the results of the candidate search. The candidate search includes considering symbols in a hierarchy or tree structure that represents the possibly transmitted combinations. The cost function incrementally accumulates a cost of a possible combination as the search progresses through the hierarchy. As soon as a symbol is determined to not be a valid candidate, all possible combinations including that symbol are discarded without considering them, directly.

Spherical decoder for wireless communications

A symbol vector received from a multiple antenna transmitting array is detected using a sphere decoder. The sphere decoder is used to select from a set of candidate binary strings that string which is most likely to have been transmitted. Soft information for use in a turbo decoder or the like is obtained by considering cost functions for a set of candidate strings, including strings derived from the most likely string by flipping one or more bits thereof.

Method of sphere decoding with low complexity and good statistical output

A method and an apparatus is provided for producing branch metrics in a LogMAP turbo decoding operation. During a forward recursion of a trellis, a set of primary branch metrics is generated. The primary branch metrics are stored in receiver form in a relatively small memory cache module and corresponding secondary branch metrics are produced by negating the primary branch metrics. The primary branch metrics and the secondary branch metrics constitute all possible branch metrics for a given state in the trellis. During a backwards recursion of the trellis, the stored primary branch metrics are retrieved from the memory cache module and the secondary branch metrics are regenerated by negating the retrieved primary branch metrics.

Turbo decoder with reduced-size branch metric cache

A communication device comprising an ML-APP detector coupled to at least two antennas. The ML-APP detector comprises at least one Hx unit coupled to at least one LLR unit. The Hx unit generates a portion of all possible symbols that can be received and transfers each generated symbol candidate to the LLR unit which performs a conversion operation on the transferred symbol candidate to generate another symbol candidate that is not part of the special portion. In this manner all of the possible symbol candidates are obtained by the LLR unit. The LLR unit compares the symbol candidates to a received symbol to perform a cost calculation. The symbol candidate yielding the lowest cost from the cost calculations of all possible symbol candidates is selected as the best candidate. APP decoding is then performed on the selected candidate using soft information associated with the selected candidate which soft information is generated by the LLR unit.

Maximum likelihood a posteriori probability detector

A method and apparatus for performing add-compare-select processing using carry-save arithmetic. Data compressors that operate based upon carry-save principles are utilized to render the correct result without requiring intermediate results to be resolved. Intermediate results are truncated to ensure that the dynamic range of the add-compare-select unit is not exceeded, whilst ensuring that the resolution of the intermediate results is not adversely affected. The computation of two competing paths is delayed and only the difference is computed directly, resulting in a reduction of the propagation path through the add-compare-select unit.

High speed add-compare-select processing

Path metric normalization in calculating trellis-based algorithms is improved by normalizing the path metric with an average value (rather than a maximum or a minimum) of the path metrics. Using an average rather than a maximum or a minimum provides various advantages without adversely affecting the execution of the trellis-based algorithm. Due to the relatively faster computation time available when an average is computed, average value normalization can be applied every cycle, because its critical path is less than that of the ACS unit.

Path metric normalization

An error convergence measurement circuit for providing convergence of a filter includes a terminating component configured to terminate filter convergence independent of an operating environment of the filter. Selectable termination threshold levels are provided.

Error convergence measurement circuit for providing convergence of a filter

In the multiplier, a partial product circuit generates a partial product based on a multiplicand operand and outputs of a Booth recoder circuit, which operates on a multiplier operand. The partial product circuit ANDs the multiplicand with a zero Booth recoded output, which indicates whether to zero out the multiplicand. An enable circuit selectively enables the multiplier circuit, and more particularly, disables the multiplier circuit by making the zero Booth recoded output indicate to zero out the multiplicand.

Growing community of inventors

Pyrmont, Australia

David Garrett