TY - GEN
T1 - Characterization of the Complexity of Computing the Capacity of Colored Noise Gaussian Channels
AU - Boche, Holger
AU - Grigorescu, Andrea
AU - Schaefer, Rafael F.
AU - Poor, H. Vincent
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - This paper investigates the computational complexity involved in determining the capacity of the band-limited additive colored Gaussian noise (ACGN) channel and its capacity-achieving input power spectral density (p.s.d.). A band-limited polynomial time computable continuous and strictly positive noise p.s.d. is constructed for the ACGN channel such that the computation of its corresponding capacity is ≠P1-complete. This means that it is even more complex than problems that are ≠P1-complete. Additionally, it is shown that computing the capacity-achieving input p.s.d. is also ≠≠P1-complete. Furthermore, under the widely accepted assumption that ≠P1≠P1, there are two significant implications for the ACGN channel. First, there exists a polynomial time computable noise p.s.d. for which computing its capacity is not polynomial-time feasible, meaning the number of computational steps on a Turing Machine grows faster than any polynomial. Second, there is a polynomial time computable noise p.s.d. where determining its capacity-achieving input p.s.d. is also not achievable in polynomial time.
AB - This paper investigates the computational complexity involved in determining the capacity of the band-limited additive colored Gaussian noise (ACGN) channel and its capacity-achieving input power spectral density (p.s.d.). A band-limited polynomial time computable continuous and strictly positive noise p.s.d. is constructed for the ACGN channel such that the computation of its corresponding capacity is ≠P1-complete. This means that it is even more complex than problems that are ≠P1-complete. Additionally, it is shown that computing the capacity-achieving input p.s.d. is also ≠≠P1-complete. Furthermore, under the widely accepted assumption that ≠P1≠P1, there are two significant implications for the ACGN channel. First, there exists a polynomial time computable noise p.s.d. for which computing its capacity is not polynomial-time feasible, meaning the number of computational steps on a Turing Machine grows faster than any polynomial. Second, there is a polynomial time computable noise p.s.d. where determining its capacity-achieving input p.s.d. is also not achievable in polynomial time.
UR - http://www.scopus.com/inward/record.url?scp=85202822030&partnerID=8YFLogxK
U2 - 10.1109/ICC51166.2024.10622861
DO - 10.1109/ICC51166.2024.10622861
M3 - Conference contribution
AN - SCOPUS:85202822030
T3 - IEEE International Conference on Communications
SP - 4114
EP - 4119
BT - ICC 2024 - IEEE International Conference on Communications
A2 - Valenti, Matthew
A2 - Reed, David
A2 - Torres, Melissa
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 59th Annual IEEE International Conference on Communications, ICC 2024
Y2 - 9 June 2024 through 13 June 2024
ER -