US 7,397,869 B2Granted

Process for providing a pilot aided phase recovery of a carrier

AssigneeAgence Spatiale Europeenne

InventorsAlberto Ginesi, Domenico Fittipaldi, Alan Bigi, Riccardo De Gaudenzi

Life of this patent

Estimating the term…

Loading patent…

Abstract

Providing a pilot aided phase recovery of a carrier of an input digital signal Zk having signal fields of Ls symbol signals, LP pilot symbol signals ZP(k) and (Ls−LP) data symbol signals Zd(k), by for each signal field: calculating an unwrapped pilot phase estimate {circumflex over (θ)}f(P)(l); initiating with {circumflex over (θ)}f(P)(l) a first digital phase locked loop implementing a phase estimate algorithm and calculating a forward phase trajectory {circumflex over (θ)}F(k) from Zd(k), ks varying between 1 and Ls−LP over the data field (l), {circumflex over (θ)}F(P)(l) having Ls−LP forward phase estimates {circumflex over (θ)}F(ks); and initiating with {circumflex over (θ)}(P)f(l+1) a second digital phase locked loop, implementing a phase estimate algorithm and calculating a backward phase trajectory {circumflex over (θ)}B(ks) from Zd(k), ks varying between Ls−LP and 1 over said data field (l), {circumflex over (θ)}B(ks) having Ls−LP backward phase estimates {circumflex over (θ)}B(ks); and from said phase trajectories calculating a phase correction (e−j{circumflex over (θ)}(ks)).

Claims15

Dependent claims

A process for providing pilot aided phase recovery of a carrier of an input digital signal Zk, said signal Zk comprising signal fields comprised of Ls symbols, namely a block of LP pilot symbols ZP(k) and a data field of Ls−LP data symbols Zd(k), and characterized in that it comprises for each signal field:

calculating an unwrapped pilot phase estimate {circumflex over (θ)}f(P)(l);
initiating with {circumflex over (θ)}f(P)(l) a first digital phase locked loop (DPLLF) implementing a phase estimate algorithm and calculating a forward phase trajectory {circumflex over (θ)}F(k) from the data symbols Zd(k) of the data field, ks being an integer varying between 1 and Ls−LP over said data field, said forward phase trajectory {circumflex over (θ)}F(P)(l) being comprised of Ls−LP forward phase estimates {circumflex over (θ)}F (ks);
and initiating with {circumflex over (θ)}(P)f(l+1) a second digital phase locked loop (DPLLB), implementing a phase estimate algorithm and calculating a backward phase trajectory {circumflex over (θ)}B (ks) from the data symbol signals Zd(k), ks varying between Ls−LP and 1 over said data field (l), said backward phase trajectory {circumflex over (θ)}B (ks) being comprised of Ls−LP backward phase estimates {circumflex over (θ)}B (ks);
from said phase trajectories calculating a phase correction (e−j{circumflex over (θ)}(ks));
and multiplying said phase correction e−j{circumflex over (θ)}(ks) with said input signal Zk to generate a phase corrected input signal for the data symbols, and wherein:
{circumflex over (θ)}(p)f(l)={circumflex over (θ)}(p)f(l−1)+αSAW└{circumflex over (θ)}(p)(l)−{circumflex over (θ)}(p)f(l−1)┘
where SAW [Φ]≡[Φ]−π+π is a sawtooth nonlinearity that reduces Φ to the interval [−π, π) and α is a parameter in the range 0<α≦1, and {circumflex over (θ)}(P)(l) is a phase estimate in the block of Lp pilot symbol of said pilot field (l).

4 more elements

calculating an unwrapped pilot phase estimate {circumflex over (θ)}f(P)(l);
initiating with {circumflex over (θ)}f(P)(l) a first digital phase locked loop (DPLLF) implementin a hase estimate algorithm and calculating a forward phase trajectory {circumflex over (θ)}F(k) from the data symbols Zd(k) of the data field, ks being an integer varying between 1 and Ls−LP over said data field, said forward phase trajectory {circumflex over (θ)}F(P)(l) being comprised of Ls−LP forward phase estimates {circumflex over (θ)}F (ks);
and initiating with {circumflex over (θ)}(P)f(l+1) a second digital phase locked loop (DPLLB). implementing a phase estimate algorithm and calculating a backward phase trajectory {circumflex over (θ)}B (ks) from the data symbol signals Zd(k), ks varying between Ls−LP and 1 over said data field (l), said backward phase trajectory {circumflex over (θ)}B (ks) being comprised of Ls−LPbackward phase estimates {circumflex over (θ)}B (ks);
from said phase trajectories calculating a phase correction (e−j{circumflex over (θ)}(ks));
and multiplying said phase correction e−j{circumflex over (θ)}(ks) with said input signal Zk to generate a phase corrected input signal for the data symbols, wherein:
{circumflex over (θ)}F(kS+1)={circumflex over (θ)}(kS)+μ(k)
μ(k+1)=μ(k)+γ(1+ρ)e(k+1)−γe(k)
e and γ designating loop parameters
e(k)=Im {y(k)ĉ*(k)}
ĉ*(k) being an estimate of the kth transmit data symbol;
ĉ*(k) being the conjugate of ĉ(k).

8 more elements

Dependencies are inferred from the claim text.