Iterative mitigation of nonlinear co-channel interference in high-efficiency multibeam satellite systems
A communications apparatus to resolve a composite signal including an induced nonlinear distortion, a desired signal and interferer signals, wherein the desired signal includes desired symbols and the interferer signals include interferer symbols using N frameworks, each framework including a detector to partition the desired symbols and the interferer symbols based on an interference severity into a dominant group and a non-dominant group, and to generate A Posteriori Probabilities (APP) of the desired symbols and the interferer symbols, wherein the detector of each of the N frameworks generates the APP based on a feedback of the APP from each of the N frameworks, and the detector reduces the induced nonlinear distortion of the desired signal using a nonlinear mathematical model such as Volterra series.
1 . A communications apparatus comprising:
a composite signal comprising an induced nonlinear distortion, a desired signal and interferer signals, wherein the desired signal comprises desired symbols and the interferer signals comprise interferer symbols; and
N frameworks, each framework comprising a detector to partition the desired symbols and the interferer symbols based on an interference severity into a dominant group and a non-dominant group, and to generate A Posteriori Probabilities (APP) of the desired symbols and the interferer symbols,
wherein the detector of each of the N frameworks generates the APP based on a feedback of the APP from each of the N frameworks, and
the detector reduces the induced nonlinear distortion of the desired signal using a nonlinear mathematical model, wherein the nonlinear mathematical model comprises a Volterra series represented as
s
NL
,
n
(
t
)
=
∑
l
λ
~
n
,
l
·
(
s
n
(
t
)
·
❘
"\[LeftBracketingBar]"
s
n
(
t
)
❘
"\[RightBracketingBar]"
l
-
1
)
,
where {tilde over (λ)} n,l are complex-valued coefficients to account for the induced nonlinear distortion of a respective High-Power Amplifier (HPA) of N HPAs and s n (t) is a baseband representation of the desired signal after pulse shaping.
2 . The communications apparatus of claim 1 , wherein the detector of each of the N frameworks transforms the APP to a first Log-Likelihood Ratios (LLRs) using a bit-to-symbol mapping rule, and
each of the N frameworks further comprises
a deinterleaver to deinterleave the first LLRs into a decoder input,
a Forward Error Correcting (FEC) Decoder to decode the decoder input and to generate a second LLRs from a decoded decoder input, and
an interleaver to interleave the second LLRs,
wherein the second LLRs are fed back as the APP for a respective N framework after the interleaver of the respective N framework, and
the second LLRs of each of the N frameworks represent either the desired symbols or the interferer symbols.
3 . The communications apparatus of claim 2 , wherein the desired signal is more robust than each of the interferer signals, and the second LLRs from the FEC decoder of a first framework of the N frameworks represent the desired symbols.
4 . The communications apparatus of claim 2 , wherein at least one of the interferer signals is more robust than the desired signal, and the second LLRs from the FEC decoder of a framework other than a first framework of the N frameworks represent the desired symbols.
5 . The communications apparatus of claim 4 , wherein the desired symbols are recovered by Simultaneous Decoding (SD) or by Simultaneous Non-unique Decoding (SND).
6 . The communications apparatus of claim 1 , wherein the induced nonlinear distortion is generated by a High-Power Amplifier (HPA) operating close to its saturation.
7 . The communications apparatus of claim 1 , wherein the communications apparatus is disposed in a SISO Iterative Divide and Conquer (IDAC) receiver and the detector is a SISO DAC detector.
8 . The communications apparatus of claim 1 , wherein the dominant group comprises an Optimal-Bayesian (OB) group via a probability mass function (pmf).
9 . The communications apparatus of claim 1 , wherein the non-dominant group comprises a Noise-Floor (NF) group incorporated via a power of each member and a Subtractive-Cancellation (SC) group incorporated via first- and second-order moments derived from the APP, and the dominant group comprises an Optimal-Bayesian (OB) group incorporated via a probability mass function (pmf).
10 . The communications apparatus of claim 1 , wherein a count of the N frameworks is selected from one (1), two (2) or three (3).
11 . The communications apparatus of claim 1 , wherein the desired signal and the interferer signals comprise DVB-S2X standard compliant signals.
12 . The communications apparatus of claim 1 , wherein a computational complexity of the detector is lowered by focusing on an instantaneous version of a current time sample.
13 . A computer implemented method comprising:
providing a composite signal comprising an induced nonlinear distortion, a desired signal and interferer signals, wherein the desired signal comprises desired symbols and the interferer signals comprise interferer symbols; and
iteratively computing, N frameworks, each of the N frameworks comprising
partitioning, with a detector, the desired symbols and the interferer symbols based on an interference severity into a dominant group and a non-dominant group, and
generating A Posteriori Probabilities (APP) of the desired symbols and the interferer symbols,
wherein the detector of each of the N frameworks generates the APP based on a feedback of the APP from each of the N frameworks,
the detector reduces the induced nonlinear distortion of the desired signal using a nonlinear mathematical model, and
the nonlinear mathematical model comprises a Volterra series represented as
s
NL
,
n
(
t
)
=
∑
l
λ
~
n
,
l
·
(
s
n
(
t
)
·
❘
"\[LeftBracketingBar]"
s
n
(
t
)
❘
"\[RightBracketingBar]"
l
-
1
)
,
where {tilde over (λ)} n,l are complex-valued coefficients to account for the induced nonlinear distortion of a respective High-Power Amplifier (HPA) of N HPAs and s n (t) is a baseband representation of the desired signal after pulse shaping.
14 . The method of claim 13 , wherein the detector of each of the N frameworks transforms the APP to a first Log-Likelihood Ratios (LLRs) using a bit-to-symbol mapping rule, and
the iteratively computing further comprises:
deinterleaving the first LLRs into a decoder input,
decoding the decoder input,
generating a second LLRs from a decoded decoder input and
interleaving the second LLRs,
wherein the second LLRs are fed back as the APP for a respective N framework after the interleaver of the respective N framework, and
the second LLRs of each of the N frameworks represent either the desired symbols or the interferer symbols.
15 . The method of claim 14 , wherein the desired signal is more robust than each of the interferer signals, and the second LLRs from the decoding of a first framework of the N frameworks represent the desired symbols.
16 . The method of claim 14 , wherein at least one of the interferer signals is more robust than the desired signal, and the second LLRs from the decoding of a framework other than a first framework of the N frameworks represent the desired symbols.
17 . The method of claim 13 , wherein the non-dominant group comprises a Noise-Floor (NF) group incorporated via a power of each member and a Subtractive-Cancellation (SC) group incorporated via first- and second-order moments derived from the APP, and the dominant group comprises an Optimal-Bayesian (OB) group incorporated via a probability mass function (pmf).
18 . The method of claim 17 , wherein the induced nonlinear distortion is generated by a High-Power Amplifier (HPA) operating close to its saturation.