US5467374A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Low complexity adaptive equalizer for U.S. digital cellular radio receivers
Assignee: Ericsson Inc
•
Filing: 1993-10-29
•
Granted: 1995-11-14
•
Expires: 2013-10-29
Abstract
Abstract A low complexity adaptive equalizer for use in U.S. digital cellular radios demodulates π/4-shifted differentially encoded quadrature phase shift keyed (DQPSK) encoding in the presence of intersymbol interference (ISI) with reduced decoding complexity by employing an estimated received constellation which takes into account channel changes over time and ISI. The decoding complexity is reduced by tracking a reduced number of estimated reference constellation points and taking advantage of the geometry to estimate the remaining symbol constellation points. A slot of symbols is decoded, employing unique filtering, in segments from the beginning in a forward direction and from the end in a reverse direction meeting at a fade point. This isolates a fade and allows the equalizer to perform more reliably.
Inventors
- 👤 Sandeep Chennakeshu
- 👤 Paul W. Dent
- 👤 Ravinder D. Koilpillai
- 👤 Raymond L. Toy
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-07 23:45
CPC Classifications
H; H04; H04L; H04L25/00; H04L25/02; H04L25/03; H04L25/03006; H04L25/03178; H04L25/03184; H04L25/03197; H04L2025/03433; H04L2025/03535; H04L2025/03547; H04L2025/0356; H04L2025/03777; H04L2025/03783; H04L2025/03796
Claims Analysis
Total Claims: 7
Claim 1
Independent Claim
279 words
A method of equalizing and demodulating a transmitted signal having synchronization ("sync") s pre [n] symbols and message symbols s sp [n] selected from even and odd π/4-shifted differential quadrature phase shift keyed (DQPSK) symbol constellation points {S}, {S}, into decoded symbols at a receive unit comprising the steps of: a) receiving the transmitted signal; b) synchronizing and sampling the transmitted signal into received samples r[n], where n is an integer, and separating the received data r[n] into sequences of sync samples r pre [n] and received message samples r sp [n]; c) selecting a portion of the received message samples r sp [n] for decoding; d) determining initial channel impulse response (CIR) coefficients from the sync samples, r pre [n]; e) determining initial sets of estimated reference constellation points {Z ref } and {Z ref } for even and odd constellations respectively, from the initial CIR coefficients and the predetermined set of sync symbols s pre [n]; f) rotating {Z ref } and {Z ref } to result in a full set of estimated symbol constellation points {Z} and {Z} which account for changes through the channel and intersymbol interference; g) determining a most likely path of state transitions over a plurality of time period stages through a trellis, corresponding to the selected portions of the received message samples r sp [n] where each possible state transition has been preassigned to one of the π/4-shifted DQPSK symbol constellation point, {S} or {S}, for even and odd stages, respectively; h) tracing back the state transitions, and the corresponding symbol constellation points of the most likely path being the decoded symbols; i) repeating steps "c" through "h" for subsequent decoding.
Claim 2
Independent Claim
279 words
A method of equalizing and demodulating a transmitted signal having synchronization ("sync") s pre [n] symbols and message symbols s sp [n] selected from even and odd π/4-shifted differential quadrature phase shift keyed (DQPSK) symbol constellation points {S}, {S}, into decoded symbols at a receive unit comprising the steps of: a) receiving the transmitted signal; b) synchronizing and sampling the transmitted signal into received samples r[n], where n is an integer, and separating the received data r[n] into sequences of sync samples r pre [n] and received message samples r sp [n]; c) selecting a portion of the received message samples r sp [n] for decoding; d) determining initial channel impulse response (CIR) coefficients from the sync samples, r pre [n]; e) determining initial sets of estimated reference constellation points {Z ref } and {Z ref } for even and odd constellations respectively, from the initial CIR coefficients and the predetermined set of sync symbols s pre [n]; f) rotating {Z ref } and {Z ref } to result in a full set of estimated symbol constellation points {Z} and {Z} which account for changes through the channel and intersymbol interference; g) determining a most likely path of state transitions over a plurality of time period stages through a trellis, corresponding to the selected portions of the received message samples r sp [n] where each possible state transition has been preassigned to one of the π/4-shifted DQPSK symbol constellation point, {S} or {S}, for even and odd stages, respectively; h) tracing back the state transitions, and the corresponding symbol constellation points of the most likely path being the decoded symbols; i) repeating steps "c" through "h" for subsequent decoding.
Claim 3
Dependent Claim
261 words
the method of equalizing and demodulating of claim 1 wherein the step of determining a most likely path comprising the steps of: a) setting a set of accumulated path metrics P i to zero; b) determining a branch metric d ji [n] for each state transition (branch), the metric being the Euclidean distance from each estimated signal constellation point {Z} to each r sp [n], for even values of n and determining the branch metric d ji [n] being a Euclidean distance from each estimated signal constellation point {Z}, to each r sp [n], for odd values of n; c) determining the lowest d ji [n] for all "i" state transitions ending in state "j" at stage n; d) storing a state transition for the lowest d ji [n] for each starting state "i" in a path history register PH i [n]; e) adding the lowest d ji [n] for starting state "j" to the accumulated path metric P j ; f) adding this state transition to the end of the path, and change the path subscript to "i" to reflect its new ending state; g) updating estimated symbol constellation points pertaining to the lowest branch metric d ji [n-δ] at time n-δ where δ represents a number of stages which passed since a last update, to compensate for changes in the channel CIR; and h) repeating steps "b" through "g" for a plurality of stages and received message samples r sp [n]; i) selecting the path having the lowest accumulated path metric, P i , as the most likely path.
Claim 4
Dependent Claim
132 words
The method of equalizing and demodulating of claim 1 wherein the initial CIR coefficients h are determined according to: h=MR where h=[h 0 [n] h 1 [n] . . . h L-1 [n]] T , the received symbols r sp [n] are R=[r 1 [n] r 2 [n] . . . r N .sbsb.P [n]], M=[S H S] -1 S H where S H is defined as S H =[S T ]* the superscripts H, T, and * stand for the Hermitian and transpose matrix operators, and complex conjugate, respectively, and the matrix S, symbols of the preamble known to a receive unit, is given by: ##EQU15## where L is an estimate of the number of rays in a hypothesized multi-path channel model and N p is the number of sync symbols.
Claim 5
Dependent Claim
112 words
The method equalizing the demodulating of claim 3 wherein the initial CIR coefficients {h 0 , h 1 } are determined by implementing the following equations in an iterative loop for n=1,2, . . . ,N p , where N p is the number of sync symbols: e(n)=r.sub.pre [n]-h.sup.T [n]S[n] where e(n) is an intermediate value used in further calculations, h.sup.T =[h.sub.0 [n] h.sub.1 [n]] and A.sup.T [n]=s.sub.pre [n] s.sub.pre [n-1]], h[n+1]=h[n]+0.5 e(n) S*[n] where S*[n] is the complex conjugate of S[n]; ##EQU16## where σ(n)=λ σ(n)+1, σ(0)=0, 1 is from 0.6 to 1.0, s is a weighting factor, and h=]h 0 [n]h 1 [n]] is a smoothed version of the CIR coefficients.
Claim 6
Dependent Claim
37 words
The method of equalizing and demodulation of claim 1 wherein the step of selecting a portion of the received message samples r sp [n] for decoding comprises the step of performing double sided slot traversing (DSST) decoding.
Claim 7
Dependent Claim
179 words
The method of equalizing and demodulating of claim 1 wherein the step of selecting a portion of the received message samples r sp [n] for decoding comprises: a) for an initial decoding: breaking the received message samples, r sp [n], into blocks with a starting block at the beginning of the slot and an ending block at the end of the slot; decoding the starting block in a forward sense; calculating a forward metric; decoding the ending block in a reverse sense; calculating a reverse metric; b) for subsequent decoding: i. if the forward metric is lower than the reverse metric then: decoding a first block not yet decoded in a forward sense, calculating a replacement forward metric, updating the forward metric with the replacement forward metric, ii. if the forward metric is not lower than the reverse metric then: decoding a last block not yet decoded in a reverse sense, calculating a replacement reverse metric, and updating the reverse metric with the replacement reverse metric; and c) repeating step "b" for blocks which have not yet been decoded.
References Cited (Backward Citations)
US5063524A
Cited
US5005189A
Cited
US5166924A
Cited
US5311545A
Cited
US5311546A
Cited
US5363408A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1998-11-13 | 1999-05-15 | 1999-11-14 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2002-11-13 | 2003-05-15 | 2003-11-14 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2006-11-13 | 2007-05-15 | 2007-11-14 | $7700.00 | Window Closed / Lapsed |