Symbol reliability determination and symbol pre-selection based on reliability criteria

Abstract

A method and apparatus for processing demodulated data comprising received symbol data is disclosed. A decoder is used to compute estimated symbols and corresponding reliability metrics. The reliability metrics are transformed into reliability weights. Optionally, residuals relating to the difference between the received symbol data and the estimated symbols are computed. Output data are generated comprising any combination of the following: estimated symbols, reliability weights, residuals, and received symbol data. The residuals may be weighted by the reliability metrics and used by demodulation or error compensation loops to instantaneously reduce or increase the bandwidth of these loops.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates to the field of communication data, and more particularly to methods and apparatus for determining symbol estimates and reliability weights, and for pre-selecting symbols based upon reliability criteria.

2. Description of Related Art

Assessing the reliability of received data is important in communication systems. For example, data reliability assessment is important when applying decision-assistance to demodulation tasks, such as carrier phase tracking and equalization. This is particularly true when the demodulator precedes a turbo decoder, and the system is operating at very low symbol-to-noise ratios. In this case the symbol error rates can become so large that the demodulation process is severely degraded, and consequently the decision-directed demodulation process becomes undesirable.

Under such conditions, symbol pre-selection becomes desirable. When a symbol decision is highly likely to be error-free, it is desirable to use the associated datum for further processing. When a symbol decision is unlikely to be error-free, it is desirable not to use the associated datum. It is desirable to avoid using errored data because harm caused by bad data may exceed the benefit provided by good data.

Decision-assisted data processing can be enhanced if the reliability assessment of symbol data is provided in a soft format, such as a linear probability with a range of values between 0 and 1, inclusive. This enables the decision-assisted process to “weight”, or equivalently, to “pre-select” symbol data prior to subsequent processing.

Therefore, a need exists for a method and apparatus for providing soft reliability determination of received data, and for pre-selecting symbol data based on reliability criteria.

SUMMARY OF THE INVENTION

The present invention relates to apparatus and methods of estimating symbol data and determining symbol reliability weights. The present invention also relates to apparatus and methods of pre-selecting symbol data according to symbol reliability criteria. Pre-selecting symbol data may involve either “hard” or “soft” reliability determination, responsive to reliability weights. In an example of a hard reliability determination, a reliability weight may be either 1 or 0, and the corresponding symbol data are used or discarded, respectively. In an example of a soft reliability determination, a reliability weight ranging between 0 and 1, inclusive, may be used as a factor for subsequent signal processing using corresponding symbol data. Pre-selection techniques involving a combination of both hard and soft determination is also used in accordance with the present invention.

In one embodiment, a log-MAP decoder is used to compute the log-likelihood (“LL”) values for all possible decoded symbols corresponding to received symbol data r_k. The maximum log-likelihood LL_k(ŝ_k) corresponding to the most likely estimated symbol ŝ_k is determined. A linear probability P_k is computed in accordance with the expression: p_k=e^LLk(ŝk). The linear probability p_k is compared to a selected threshold T corresponding to a minimum acceptable data reliability. A reliability weight R_k is determined depending on whether p_k is greater than T. If p_k is not greater than T, then R_k is set equal to 0. If p_k is greater than T, then R_k is set equal to a value or expression other than zero. Examples include, without limitation, the following values and expressions: R_k=1; R_k=p_k; and R_k=(p_k−T)/(1−T). Optionally, a residual z_k value, relating to the difference between ŝ_k and r_k, may be computed. Output data are provided that comprise any combination of the following: ŝ_k, R_k, z_k, r_k.

In another embodiment, data comprising symbols r_k are decoded by any device capable of producing soft symbol estimates, including, without limitation, a MAP decoder, a log-MAP decoder, a Soft-Input, Soft-Output decoder, a Soft-Output Viterbi Algorithm decoder, a max-log MAP decoder, or the like. Decoding may also be performed by a Viterbi decoder where a reliability metric is derived from the best path metric, minus the second-best path metric. An estimated symbol ŝ_k and a reliability metric RM_k are determined for a received symbol r_k. The reliability metric may, for example, be a maximum log-likelihood value. A probability metric, p_k, corresponding to the estimated symbol ŝ_k, is computed by a method that may, for example, include steps for computing a probability metric, p_k, taught herein above. Alternatively, other methods for computing a probability metric, p_k, may be employed. A reliability weight, R_k, is determined, depending on whether p_k is greater than T. If p_k is not greater than T, then R_k is set equal to 0. If p_k is greater than T, then R_k is set equal to a value or expression other than zero. Examples include, without limitation, the following values and expressions: R_k=1; R_k=p_k; and R_k=(p_k−T)/(1−T). Optionally, a residual value, z_k, relating to the difference between ŝ_k and r_k, may be computed. Output data are provided that comprise any combination of the following: ŝ_k, R_k, z_k, r_k.

In yet another embodiment of the present invention, data comprising symbols r_k are decoded by any device including, without limitation, a MAP decoder, a log-MAP decoder, a Soft-Input, Soft-Output decoder, a Soft-Output Viterbi Algorithm decoder, a max-log MAP decoder, a Viterbi decoder, or the like. An estimated symbol, ŝ_k, is determined for a received symbol r_k. The estimated symbol, ŝ_k, is compared to an uncoded symbol estimate, û_k. The term “uncoded symbol estimate” is used here to denote a symbol estimate determined without reference to the data code structure. For example, an uncoded symbol estimate, û_k, may be determined by selecting the symbol corresponding to a point in the signalling constellation that most closely approximates the received symbol data, r_k. If ŝ_k=û_k, then a reliability weight R_k=1 is assigned corresponding to the estimated symbol ŝ_k. If ŝ_k≠û_k, then a reliability weight R_k=0 is assigned. Optionally, a residual value, z_k, relating to the difference between ŝ_k and r_k, may be computed. Output data are provided that may comprise any combination of the following: ŝ_k, R_k, z_k, r_k.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of a receiver circuit incorporating a symbol reliability device.

FIG. 2 is a flow diagram showing an exemplary method for determining symbol estimates and reliability probability metrics.

FIG. 3 is an example of a calculation of a forward state probability.

FIG. 4 is an example of a calculation of a reverse state probability.

FIG. 5 is an example of a calculation of a forward transition probability.

FIG. 6 is a flow diagram showing an embodiment of a method for pre-selecting symbol data in accordance with the present invention.

FIG. 7 is a flow diagram showing another embodiment of a method for pre-selecting symbol data in accordance with the present invention.

FIG. 8 illustrates an 8-PSK coding constellation.

FIG. 9 is a simplified block diagram of a phase tracking system incorporating a symbol reliability device made in accordance with the present invention.

Like reference numbers and designations in the various drawings indicate like elements.

DETAILED DESCRIPTION

Throughout this description, embodiments and variations are described for the purpose of illustrating uses and implementations of the invention. The illustrative description should be understood as presenting examples of the invention, rather than as limiting the scope of the invention.

FIG. 1 shows a simple block diagram of an inventive receiver 100, including a symbol reliability device 104. Modulated channel symbol data 108 are received over a communications channel and demodulated by a demodulator 102. Demodulated data 110 comprise received symbols r_k. A symbol reliability device 104 performs symbol reliability operations as described hereinbelow to output data 112. Data 112 may comprise any combination of the following: estimated symbols ŝ_k reliability weights R_k, received symbols r_k, residual values z_k. Data 112 are provided as input to a receiver module or device 106 as shown in FIG. 1. For example, data 112 can be provided to modules for tracking carrier signal phase as taught in the above-incorporated U.S. patent applications Ser. No.: 09/715,877, filed on Nov. 17, 2000, entitled “SYSTEM FOR CARRIER PHASE TRACKING OF CODED SYMBOLS USING RELIABILITY METRICS FOR SYMBOL ESTIMATES”; Ser. No.: 09/729,652, filed Dec. 4, 2000, entitled “ITERATIVE CARRIER PHASE TRACKING SYSTEM”; and Ser. No.: 09/738,896, filed Dec. 15, 2000, and entitled “SYSTEM FOR CARRIER PHASE TRACKING OF MULTI-DIMENSIONAL CODED SYMBOLS.” Additional information and examples relating to the use of symbol reliability data is provided hereinbelow. The tracking module 106 output signal data 114 for subsequent use.

In one embodiment of the present invention, the system reliability device 104 includes a “maximum a posteriori” (MAP) decoder or log-MAP decoder. MAP decoding and log-MAP decoding relate to an optimal method for decoding linear block and convolutional codes to minimize symbol error probability. One reference that teaches principles of MAP decoding is entitled “Optimal Decoding of Linear Codes for Minimizing Symbol Error Rate,” by L. R. Bahl et al., IEEE Transactions on Information Theory, March 1974, pp. 27–30 (hereinafter referred to as “the Bahl reference”). Additional information on MAP and log-MAP decoders can be found in the following references: “Near Shannon Limit Error-Correcting Coding and Decoding: Turbo Codes,” C. Berrou et al., Proc. ICC '93 Geneva, Switzerland, May 1993, pp. 1064–1070 (hereinafter referred to as “the Berrou reference”); “An Intuitive Justification and a Simplified Implementation of the MAP Decoder for Convolutional Codes,” A. Viterbi, IEEE Journal On Selected Areas In Telecommunications, Vol. 16, No. 2, February 1998, pp. 260–264 (hereinafter referred to as “the Viterbi reference”); and J. Hagenauer and P. Hoeher, “A Viterbi Algorithm with Soft-decision Outputs and its Applications,” in Proceedings of IEEE Globecom '89, Dallas, Tex., November 1989, pp. 47.1.1–47.1.7. (“the Hagenauer reference”). Each of the Bahl, Berrou, Viterbi, and Hagenauer references is hereby fully incorporated by reference herein as though set forth in full for teachings on log-MAP decoding. Exemplary use of log-MAP decoding in accordance with the present invention is specified hereinbelow.

FIG. 2 shows an inventive method of using log-MAP decoding for estimating symbols and symbol reliability in accordance with the present invention. In the flowchart shown in FIG. 2, the notation “α_k(m)” refers to the natural log domain (hereinafter referred to as the “log domain”) forward probability of being in state “m” at a time “k”; γ_kⁱ refers to the log domain edge probability for edge i at time k; and β_k(m) refers to the log domain reverse probability of being in state m at a time k. It is assumed for purposes of the flow chart shown in FIG. 2 that a block of N channel symbols has been or will be received.

In step 200, the boundary values α₀(m) and β_N(m) are initialized for all values of m. One example of suitable boundary values are given as follows: α₀(0)=0 and α₀(m)=−∞ for m≠0; β_N(0)=0 and β_N(m)=−∞ for m≠0. Other suitable initialization values and methods are possible, as those of ordinary skill in the symbol reliability determination arts shall recognize.

In step 202, for an observation r_k, α_k(m) and γ_kⁱ are computed for all values of m and i. Note that the observation r_k is a channel symbol as perturbed by noise through passage through the channel. In one embodiment, the forward probabilities α_k(m) are computed recursively as a function of α_k−1(m) in accordance with the example described below with reference to FIG. 3.

An example of the computation of forward probabilities is illustrated in FIG. 3. In this example, there are two permissible branches into state s₃, one from state s₁ and one from state s₂. The log domain probability of s₁, α(1), and the log domain probability of s₂, α(2), have been previously determined. Similarly, the log domain edge probabilities γ¹ and γ² have been previously determined. The objective is to compute the log domain probability of the state s₃, α(3), from the foregoing parameters.

In this example, the calculation of the probability α(3) can be expressed using the following equation:α(3)=MAX*(α(1)+γ¹,α(2)+γ²)where the function MAX*(A,B) is defined to be:MAX*(A,B)≡MAX(A,B)+ln(1+exp(−|A−B|))Referring again to FIG. 2, in step 204, the reverse probabilities β_k(m) are computed for all values of m. The probabilities β_k(m) are computed recursively as a function of β_k+1(m) in accordance with the following example described below with reference to FIG. 4. An example of the computation of reverse probabilities is illustrated in FIG. 4. In this example, there are two permissible branches into state s₆, one from state s₄ and one from state s₅. The log domain probability of s₄, β(4), and the log domain probability of s₅, β(5), have been previously determined. Similarly, the log domain edge probabilities γ⁴ and γ⁵ have been previously determined. The objective is to compute the log domain probability of the state s₆, β(6), from the foregoing parameters.

In this example, the calculation of the natural log domain probability β(6) can be expressed using the following equation:β(6)=MAX*(β(4)+γ⁴,β(5)+γ⁵)where the function MAX*(A, B) is defined as before.

Referring again to FIG. 2, in step 206, at the point where the computation of the forward and reverse probabilities begins to overlap, i.e., a point of adjacency, the joint log domain probabilities λ_kⁱ=α_k(m)+γ_kⁱ+β_k+1(m′) are computed for all edges at the point of overlap. An example of this computation is described with reference to FIG. 5. Referring to the leftmost state as state m at time k, and the rightmost state as state m′ at time k+1, it is assumed that the forward state log domain probability α_k(m), the reverse state log domain probability β_k+1(m′), and the edge log domain probability γ_kⁱ have all been computed. This step involves adding these probabilities, i.e., performing the computation λ_kⁱ=α_k(m)+γ_kⁱ+β_k+1(m′) in order to compute the joint log domain probability of transitioning between the two states along the prescribed edge.

Referring again to FIG. 2, in step 208, a log-likelihood, LL_k, is determined for the symbols at the point of adjacency k. The log-likelihood for a symbol s at time k, LL_k(s), may be expressed by the following equation:LL_k(s)=MAX*_{∀i that imply release of symbol s}(λⁱ_k)−MAX*_∀i(λⁱ_k)

In step 210, an estimate ŝ_k of the symbol at the point of adjacency k is taken to be ŝ_k that has the largest LL value. This condition may be expressed as follows:LL_k(ŝ_k)=MAX_∀s(LL_k(s))

In step 212, a linear probability for the symbol estimate ŝ_k is derived from LL_k(ŝ_k) as follows:p_k=e^LLk(ŝk)where p_k is the linear probability for the symbol estimate ŝ_k at time k.

FIG. 6 shows one embodiment of a method for pre-selecting symbols. In step 601, data 110, representing received symbols r_k are received by the symbol reliability device 104 (FIG. 1). In step 602, received symbol data r_k are decoded by any device capable of producing soft symbol estimates, including, without limitation, a MAP decoder, a log-MAP decoder, a Soft-Input Soft-Output decoder, a Soft-Output Viterbi Algorithm decoder, a max-log MAP decoder, or similar devices. Decoding may also be performed by a Viterbi decoder where a reliability metric is a function of the best path metric minus the second-best path metric. Step 602 provides an estimated symbol ŝ_k and a reliability metric RM_k.

Referring still to FIG. 6, at step 604 the reliability metric RM_k is converted into a probability metric p_k. As one example, RM_k may comprise the log-likelihood metric LL_k(ŝ_k) calculated at step 210 of the method illustrated in FIG. 2. In this case, a probability metric p_k is the linear probability resulting from the step 212 of the method shown in FIG. 2. Other reliability and probability metrics may also be used in accordance with the teachings herein. In some examples the reliability and probability metrics may be identical.

Referring again to FIG. 6, at the step 606, the probability metric is compared to a suitably chosen threshold T. In one embodiment, the threshold T comprises the minimum acceptable value below which data is considered unusable. For example, when p_k is a linear probability metric ranging from 0 to 1, inclusive, then the threshold T=½ is the level at which all other data symbols are as probable as the estimated symbol. If the probability metric is not greater than the threshold T, then the method proceeds to a step 610 whereat the reliability weight factor R_k is set equal to 0. If p_k is greater than T, then the method proceeds to a step 608 whereat R_k is set equal to a value or expression other than zero. Examples include, without limitation, the following values and expressions: R_k=1; R_k=p_k; and R_k=(p_k−T)(1−T).

Referring still to FIG. 6, at a step 612, a residual z_k, relating to the difference the received symbol r_k and a corresponding estimated symbol ŝ_k, may optionally be computed. Data output by the method at step 612 may comprise z_k, r_k, neither, or both.

At step 614, data processed by the preceding steps are provided as output data 112 comprising any combination of the following: ŝ_k, R_k, z_k, r_k.

FIG. 7 shows a flow diagram of another embodiment of the present inventive symbol pre-selection method. Many of the steps of the method shown in FIG. 7 are identical to those described above with reference to the method shown in FIG. 6, and therefore these steps are not described further herein. In step 701, data 110 representing received symbols r_k are received. At a step 702, data 110 are used to derive an “uncoded” symbol estimate û_k by selecting the symbol in the coding constellation that is closest to the symbol data r_k.

Relating to an example of a method for determining uncoded symbol estimate û_k, FIG. 8 illustrates an 8-PSK coding constellation with eight possible symbols comprising a 3-tuple of bits (y₂ⁱ, y₁ⁱ, y₀ⁱ), i=0, 1. Each 8-PSK symbol is a quadrature symbol having I and Q components. For PSK modulation, such symbols may also be represented solely in terms of phase. The table below, corresponding to the constellation of FIG. 8, shows one method for mapping phase into a 3-tuple of bits:

3-tuple (y2i, y1i, y0i), i = 0,1 8-PSK symbol
(0, 0, 0)                        π/16
(0, 0, 1)                        3π/16
(0, 1, 1)                        5π/16
(0, 1, 0)                        7π/16
(1, 1, 0)                        9π/16
(1, 1, 1)                        11π/16
(1, 0, 1)                        13π/16
(1, 0, 0)                        15π/16

The phase of a received datum r_k ideally would correspond exactly to an 8-PSK symbol. However, due to noise, the phase of a received symbol r_k will generally be inaccurate. The estimated symbol û_k is selected by choosing the symbol having a phase that most closely approximates the received symbol r_k without regard to the coding information utilized during the step 704 of FIG. 7. One of ordinary skill in the symbol reliability design arts shall recognize that this is only one example of many for which the teachings of the present disclosure are applicable.

Referring again to FIG. 7, at the step 704, data 110, comprising symbols r_k, are decoded by any device including, without limitation, a MAP decoder, a log-MAP decoder, a Soft-Input Soft-Output decoder, a Soft-Output Viterbi Algorithm decoder, a max-log MAP decoder, a Viterbi decoder, or similar decoding devices. STEP 704 outputs “coded” symbol estimate ŝ_k.

The method proceeds from the step 704 to a step 706 as shown in FIG. 7. At step 706, the uncoded symbol estimate û_k is compared to a coded symbol estimate ŝ_k. If ŝ_k and û_k are the same symbol, then the method proceeds to a step 708 whereat a reliability weight R_k is set equal to 1. If ŝ_k and û_k are not the same symbol, the method proceeds to a step 710 whereat the reliability weight R_k is set equal to 0.

Still referring to FIG. 7, at a step 712 a residual value z_k, relating to the difference between a received symbol r_k and a corresponding estimated symbol ŝ_k may optionally be computed. Data output by the step 712 may comprise z_k, r_k, neither, or both.

At the step 714 of FIG. 7, data produced by the previous steps are provided as output data 112 comprising any combination of the following: ŝ_k, R_k, z_k, r_k. An exemplary application for all embodiments of the present teachings may include, without limitation, the use of estimated symbols, reliability weight factors and residuals for carrier phase tracking systems as taught in the patent applications incorporated above, especially the two applications entitled “SYSTEM FOR CARRIER PHASE TRACKING OF CODED SYMBOLS USING RELIABILITY METRICS FOR SYMBOL ESTIMATES,” (Ser. No.: 09/715,877) “ITERATIVE CARRIER PHASE TRACKING SYSTEM,” (Ser. No.: 09/738,896) and “SYSTEM FOR CARRIER PHASE TRACKING OF MULTI-DIMENSIONAL CODED SYMBOLS” (Ser. No.: 09/738,896).

Exemplary Application to a System for Carrier Phase Tracking

FIG. 9 shows a simplified block diagram of one embodiment of a carrier phase tracking system incorporating the symbol reliability determination and symbol pre-selection methods of the present invention. This embodiment employs a feed-forward structure to determine the de-rotation phase θ. As shown in FIG. 9, data 110, comprising received symbol data r_k, are received by the symbol reliability device 104. The symbol reliability device 104 determines an estimated symbol ŝ_k and a corresponding reliability weight R_k associated with the data r_k in accordance with the inventive symbol reliability methods described above.

The estimated symbol ŝ_k and associated reliability weight R_k are conveyed to the tracking loop module 106. The tracking loop module 106 determines a residual z_k between the received symbol r_k and its estimate ŝ_k. Alternatively, the residual z_k may be computed within the symbol reliability device 104 and provided as input to the tracking loop module 106. In one embodiment, the residual z_k may comprise a phase residual. The reliability weight R_k for the estimate is used to weight the residual z_k. A de-rotation phase θ is determined responsive to the weighted residual z_k. Optionally, one or more previous weighted residuals and one or more previous de-rotation phases may be used in this process. One example of how a de-rotation phase may be computed by a phase tracking module is described below.

Referring again to FIG. 9, the received symbol r_k is provided to the delay module 902 by the system reliability device 104, or optionally by an alternate data route (not shown) that conveys the received symbol data r_k directly from received data 110 to the delay module 902. The delay module 902 compensates for any processing delays caused by the system reliability device 104 and the tracking loop module 106. Although the delay module 902 is shown in FIG. 9 as a separate processing block, it can be incorporated into either the system reliability device 104 or the phase tracking module 106. The delay module 902 provides delayed received symbol r_k data to the symbol de-rotator 904.

The symbol de-rotator 904 de-rotates the received symbol r_k by the de-rotation phase θ in accordance with data received from the tracking loop module 106. In one implementation, the symbol de-rotator 904 comprises a modulator. In this embodiment, the modulator de-rotates a symbol by multiplying a quadrature baseband signal that includes the received symbol r_k by a complex exponential e^−jθ. The delay module 902 ensures proper synchronization of this de-rotation process. In particular, the delay module 902 compensates for delays introduced by the symbol reliability device 104 and the tracking loop module 106. The delay module 902 thereby ensures a proper match at the de-rotator 904 between the received symbol r_k and the de-rotation phase that corresponds to the received symbol. The de-rotated received symbol r_k output by the de-rotator 904 is provided as output data 114 for further processing. The process may be repeated for subsequent symbols.

In one example, the tracking loop 106 operates in accordance with the following mathematical expression:

${\theta }_k=\sum _{i=1}^N{a}_i·{\theta }_{k-i}+\sum _{i=0}^{M-1}{b}_i·R_{k-i}·z_{k-i}$ In this expression, θ_k comprises the de-rotation phase at time k; θ_k−i represents a previous value of the de-rotation phase at time k−I; a_i is a coefficient applied to θ_k−I; z_k−i is a residual derived from r_k−i, the (k−i)th received symbol and ŝ_k−i, the estimate of that symbol; R_k−i is the reliability weight for the estimate of the (k−i)th symbol; b_i is a coefficient applied to R_k−i●z_k−I; and N and M are non-negative integers. In one embodiment, the residual z_k−i is e_k−i, the phase residual between r_k−i and ŝ_k−i. In another embodiment, the residual z_k−i is the component of r_k−i orthogonal to ŝ_k−i.Other Uses and Applications

The example above illustrates the use of symbol estimation for phase tracking, which relates to the more general concepts of demodulation tracking. Demodulation tracking is performed to adjust and optimize the demodulation process, and it includes operations such as equalization, phase noise tracking, Automatic Frequency Control (AFC), and Automatic Gain Control (AGC). In general, the present symbol reliability determination and symbol pre-selection methods described herein can be applied to all operations relating to demodulation tracking, adjustment, and optimization.

For example, and more specifically, the present inventive methods can be employed in operations that compensate for carrier frequency changes. Frequency is the derivative of phase, therefore, for systems that incorporate second-order (i.e., error-integrating) phase tracking loops, frequency errors can be more optimally compensated using the symbol estimation methods of the present invention.

Further, the present inventive symbol reliability determination methods can be employed to compensate for channel multipath changes via the equalization process.

The present invention can also be used to compensate for signal strength changes via AGC.

The present invention can also be used in the optimization of symbol timing changes.

All of the above applications may employ methods that compare symbol estimates to received data to determine offsets, and that implement compensation loops to drive the observed offsets to zero.

For example, in equalizer devices, exemplary applications of the present invention comprise the following:

• (1) Implementation of symbol reliability methods may be used within an LMS or RLS algorithm used to update the equalizer taps (i.e., within the channel estimate innovation process). When a symbol reliability is low, as determined in accordance with the present inventive techniques, response to the error signal may be reduced.
• (2) Implementation of the present inventive symbol reliability methods can be used within the decision feedback portion of a decision feedback equalizer (i.e., within the compensation process).

For timing loop processes, the present inventive symbol reliability methods may be implemented following the cross-correlation process used to determine optimal timing. In accordance with this approach, a segment of the received data is cross-correlated with a segment of the estimated data, and the optimal timing occurs where the cross-correlation is strongest. Because the correlation function can be interpolated between points using an interpolation filter, the timing offset can be estimated with very good precision, even with symbol-spaced samples. When a symbol has a low level of reliability, response to the error signal (i.e., the estimated timing offset derived from the correlation) is scaled by an attenuation factor, in accordance with the inventive methods described above. Effectively, this instantaneously reduces the response speed of the timing tracking loop. However, the response speed is reduced because the data that the timing track loop may be using to estimate offsets is suspect.

For Automatic Gain Control (AGC), the present inventive methods can be used to track and compensate magnitude differences between received and estimated data. When a symbol datum reliability is low, the response relating to that datum may be reduced, and the tracking loop speed is reduced.

In light of the examples above, those of ordinary skill in the data processing and communication arts shall recognize that the present invention can be broadly and generally applied to data signals comprising symbol data processed via demodulation tracking.

A computer readable medium which tangibly embodies the method steps of any of the embodiments herein may be used in accordance with the present invention. For example, the method steps may be embodied as a series of computer executable instructions stored on a the computer readable medium. Such a medium may include, without limitation, RAM, ROM, EPROM, EEPROM, floppy disk, hard disk, CD-ROM, etc. The disclosure also includes the method steps of any of the foregoing embodiments synthesized as digital logic in an integrated circuit, such as a Field Programmable Gate Array, or Programmable Logic Array, or other integrated circuits that can be fabricated or modified to embody computer instructions.

A number of embodiments of the present invention have been described. Nevertheless, it will be understood that various modifications may be made without departing from the scope of the invention. For example, the methods of the present invention can be executed in software or hardware, or a combination of hardware and software embodiments. As another example, it should be understood that the functions described as being part of one module may in general be performed equivalently in another module. As yet another example, steps or acts shown or described in a particular sequence may generally be performed in a different order, except for those embodiments described in a claim that include a specified order for the steps.

Accordingly, it is to be understood that the invention is not limited by the specific illustrated embodiments, but only by the scope of the appended claims. The description may provide examples of similar features as are recited in the claims, but it should not be assumed that such similar features are identical to those in the claims unless such identity is essential to comprehend the scope of the claim. In some instances the intended distinction between claim features and description features is underscored by using slightly different terminology.

Claims

1. A method of processing received symbol data rk within a communications system, comprising: (a) decoding the received symbol data rk to produce a soft symbol estimate ŝk of the received symbol data rk;(b) computing a reliability metric RMk associated with the soft symbol estimate ŝk computed in step (a);(c) computing a probability metric pk based upon the reliability metric RMk computed in step (b);(d) setting a reliability weight Rk associated with and corresponding to the soft symbol estimate ŝk to zero if the probability metric computed in the step (c) is less than or equal to a predetermined threshold value T;(e) setting the reliability weight Rk to a non-zero value if the probability metric computed in the step (c) is greater than the predetermined threshold value T;(f) calculating a residual value zk relating to a difference between the soft symbol estimate ŝk and the received symbol data rk; and(g) generating output data comprising any of the soft symbol estimate ŝk, the reliability weight, the residual value and the received symbol data rk.

2. The method of claim 1, wherein the soft symbol estimate ŝk produced in the step (a) is generated by log maximum a posteriori (log-MAP) decoding the received symbol data.

3. The method of claim 2, wherein the log-MAP decoding the received symbol data comprises computing log-likelihood (LL) values for all symbols corresponding to the received symbol data rk.

4. The method of claim 3, wherein a log-likelihood value LLk(s) for a symbol s at a time k is computed in accordance with the following equation: LLk(s)=MAX*∀i that imply release of symbol s(λki)−MAX*∀i(λki).

5. The method of claim 4, wherein the soft symbol estimate ŝk of the received symbol data at the time k comprises the soft symbol estimate ŝk having the largest LL value.

6. The method of claim 5, wherein the soft symbol estimate ŝk having the largest LL value is computed in accordance with the equation: LLk(ŝk)=MAX ∀s(LLk(s)).

7. The method of claim 6, wherein the reliability metric RMk of step (b) comprises the largest log-likelihood value LLk(ŝk)=MAX∀s(LLk(s)).

8. The method of claim 7, wherein the probability metric pk computed in step (c) is derived from the reliability metric RMk of claim 7, and wherein the probability metric pk comprises the linear probability for the symbol estimate ŝk at time k.

9. The method of claim 8, wherein the linear probability for the symbol estimate ŝk at time k is derived from the reliability metric RMk in accordance with the following equation: Pk=eLLkk(ŝk).

10. The method of claim 1, wherein the reliability metric is identical to the probability metric.

11. The method of claim 1, wherein the soft symbol estimate ŝk produced in the step (a) is generated by a maximum a posteriori (MAP) decoder.

12. The method of claim 1, wherein the soft symbol estimate ŝk produced in the step (a) is generated by a Soft-Input Soft-Output decoder.

13. The method of claim 1, wherein the soft symbol estimate ŝk produced in the step (a) is generated by a Soft-Output Viterbi decoder.

14. The method of claim 1, wherein the soft symbol estimate ŝk produced in the step (a) is generated by a Viterbi decoder and wherein the reliability metric is a function of a first best-path metric subtracted by a second best-path metric.

15. The method of claim 1, wherein the predetermined threshold value T comprises a minimum acceptable value below which the received symbol data is unusable.

16. The method of claim 15, wherein the probability metric pk comprises a linear probability metric ranging between 0 and 1, inclusive, and wherein the threshold T=½ comprises a level at which all data symbols are as probable as the soft symbol estimate produced at step (a).

17. The method of claim 1, wherein the reliability weight Rk is set equal to 1 during step (e) when the probability metric is greater than the threshold value T.

18. The method of claim 1, wherein the reliability weight Rk is set equal to the probability metric pk during step (e) when the probability metric is greater than the threshold value T.

19. The method of claim 1, wherein the reliability weight Rk is set equal to (pk−T)/(1−T) when the probability metric is greater than the threshold value T.

20. A method of processing received symbol data rk within a communications system, comprising: (a) decoding the received symbol data rk to produce a soft symbol estimate ŝk of the received symbol data rk;(b) computing an uncoded symbol estimate ûk by selecting a symbol in a coding constellation closest to the received symbol data rk;(c) comparing the uncoded symbol estimate ûk computed in step (b) to the soft symbol estimate ŝk produced in step (a);(d) setting a reliability weight Rk, associated with and corresponding to the soft symbol estimate ŝk, to zero if the uncoded symbol estimate ûk is not identical to the soft symbol estimate ŝk;(e) setting the reliability weight Rk to 1 if the uncoded symbol estimate ûk is identical to the soft symbol estimate ŝk;(f) calculating a residual value zk relating to a difference between the soft symbol estimate ŝk and the received symbol data rk; and(g) generating output data comprising any of the soft symbol estimate ŝk, the reliability weight, the residual value and the received symbol data rk.

21. The method of claim 20, wherein the step (b) of computing an uncoded symbol estimate ûk comprises selecting a symbol, in an PSK coding constellation, wherein the symbol has a phase most closely approximating the received symbol data rk.

22. The method of claim 21, wherein the PSK coding constellation comprises an 8-PSK coding constellation, and wherein the 8-PSK coding constellation has eight possible symbols comprising a 3-tuple of bits (y2i, y1i, y0i), wherein i=0, 1.

23. A carrier phase tracking system, comprising: (a) a symbol reliability block adapted to receive symbol data rk, wherein the symbol reliability block includes: (1) a decoder producing a soft symbol estimate ŝk of the received symbol data rk;(2) means for computing a reliability metric RMk associated with the soft symbol estimate ŝk;(3) means for computing a probability metric pk based upon the reliability metric RMk; and(4) means for determining a reliability weight Rk associated with and corresponding to the soft symbol estimate ŝk;(b) a tracking loop block, coupled to the symbol reliability block, wherein the tracking loop block computes a residual value zk between the received symbol rk and the soft symbol estimate ŝk;(c) a delay block, coupled to receive the received symbol data rk from the symbol reliability block, wherein the delay block compensates for processing delays created by the symbol reliability and tracking loop blocks, and wherein the delay block outputs delayed received symbol data; and(d) a symbol de-rotator, coupled to the delay and the tracking loop blocks, wherein the symbol de-rotator de-rotates the received symbol data rk by a de-rotation phase θ in accordance with input received from the tracking loop block.

24. The carrier phase tracking system of claim 23, wherein the residual value zk computed by the tracking loop block comprises a phase residual.

25. The carrier phase tracking system of claim 24, wherein reliability weight Rk is used to weight the residual value zk.

26. The carrier phase tracking system of claim 25, wherein the de-rotation phase θ is determined responsive to the weighted residual value zk.

27. The carrier phase tracking system of claim 23, wherein the symbol de-rotator comprises a modulator.

28. The carrier phase tracking system of claim 27, wherein the modulator de-rotates a symbol by multiplying a quadrature baseband signal by a complex exponential e−jθ, and wherein the quadrature baseband signal includes the received symbol data rk.

29. The carrier phase tracking system of claim 24, wherein the tracking loop block generates the de-rotation phase θk at a time k in accordance with the equation:

30. The carrier phase tracking system of claim 29, wherein the residual value zk−i comprises a phase residual between rk−i and ŝk−i, and wherein the residual value zk−i comprises phase residual ek−i.

31. The carrier phase tracking system of claim 30, wherein the residual value zk−i comprises a component of rk−i orthogonal to ŝk−i.

32. An apparatus processing received symbol data rk in a communications system, comprising: (a) means for decoding the received symbol data rk to produce a soft symbol estimate ŝk of the received symbol data rk;(b) means for computing a reliability metric RMk associated with the soft symbol estimate ŝk;(c) means for computing a probability metric pk based upon the reliability metric RMk;(d) means for determining a reliability weight Rk associated with and corresponding to the soft symbol estimate ŝk, wherein the reliability weight determining means sets the reliability weight Rk to zero if the probability metric is less than or equal to a predetermined threshold value T, and wherein the reliability weight determining means sets the reliability weight Rk to a non-zero value if the probability metric is greater than the predetermined threshold value T; and(e) means for computing a residual value zk relating to a difference between the soft symbol estimate ŝk and the received symbol data rk.

33. The apparatus of claim 32, wherein the apparatus is used within a receiver to adjust and optimize demodulation of received signals.

34. The apparatus of claim 33, wherein the demodulation of received signals includes demodulation tracking, adjustment and optimization processes.

35. An apparatus processing received symbol data rk in a communications system, comprising: (a) a decoder, wherein the decoder outputs a soft symbol estimate ŝk of the received symbol data rk; and(b) a symbol reliability processing block having an input coupled to the decoder output, wherein the processing block produces a reliability metric RMk associated with the soft symbol estimate ŝk, and wherein the processing block computes a probability metric pk based upon the reliability metric RMk, and wherein the processing block determines a reliability weight Rk associated with and corresponding to the soft symbol estimate ŝk, wherein the reliability weight Rk is set to zero if the probability metric is less than or equal to a predetermined threshold value T, and wherein the reliability weight Rk is set to a non-zero value if the probability metric is greater than the predetermined threshold value T, and wherein the processing block computes a residual value zk relating to a difference between the soft symbol estimate ŝk and the received symbol data rk.

36. A receiver in a data communications system, comprising: (a) a demodulator, wherein the demodulator receives modulated channel symbol data transmitted over a communications channel and outputs demodulated data comprising received symbol data rk;(b) a decoder, coupled to the demodulator, wherein the decoder outputs a soft symbol estimate ŝk of the received symbol data rk;(c) a symbol reliability device having an input coupled to the decoder output, wherein the device produces a reliability metric RMk associated with the soft symbol estimate ŝk, and wherein the device computes a probability metric pk based upon the reliability metric RMk, and wherein the device determines a reliability weight Rk associated with and corresponding to the soft symbol estimate ŝk, wherein the reliability weight Rk is set to zero if the probability metric is less than or equal to a predetermined threshold value T, and wherein the reliability weight Rk is set to a non-zero value if the probability metric is greater than the predetermined threshold value T, and wherein the device computes a residual value zk relating to a difference between the soft symbol estimate ŝk and the received symbol data rk; and(d) a tracking loop block, coupled to the symbol reliability device, wherein the tracking loop block compensates for variations in carrier signal phase responsive to the symbol reliability device.

37. A computer readable medium encoded with a computer program executable by a general purpose computer, wherein the program is capable of processing data signals in a receiver circuit, the computer program comprising: (a) a first set of instructions for decoding received symbol data rk to produce a soft symbol estimate ŝk of the received symbol data rk;(b) a second set of instructions for computing a reliability metric RMk associated with the soft symbol estimate ŝk;(c) a third set of instructions for computing a probability metric pk based upon the reliability metric RMk;(d) a fourth set of instructions for setting a reliability weight Rk associated with and corresponding to the soft symbol estimate ŝk to zero if the probability metric is less than or equal to a predetermined threshold value T, and for setting the reliability weight Rk to a non-zero value if the probability metric is greater than the predetermined threshold value T;(e) a fifth set of instructions for computing a residual value zk relating to a difference between the soft symbol estimate ŝk and the received symbol data rk; and(f) a sixth set of instructions for generating output data comprising any of the soft symbol estimate ŝk, the reliability weight, the residual value and the received symbol data rk.

38. A communication system comprising: (a) at least one transmitter; and(b) at least one receiver comprising the receiver as set forth in claim 36.

39. The communication system as set forth in claim 38, wherein the communication system further comprises a television operatively coupled to the at least one receiver.

40. The communication system as set forth in claim 39, wherein the communication system further comprises a television operatively coupled to the at least one receiver and a personal recorder.

41. The communication system as set forth in claim 40, wherein the communication system further comprises a set-top box operatively coupled to the at least one receiver.

42. The communication system as set forth in claim 41, wherein the communication system further comprises a personal digital assistant (PDA) operatively coupled to the at least one receiver.

43. The communication system as set forth in claim 41, wherein the communication system further comprises computer operatively coupled to the at least one receiver.