We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.IT

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Computer Science > Information Theory

Title: Performance Analysis of Wideband Near-Field Sensing (NISE)

Abstract: The impact of large bandwidth on near-filed sensing (NISE) is analyzed in multi-carrier systems. The fundamental Cramer-Rao bounds (CRBs) for wideband NISE are characterized. In particular, the closed-form CRBs are derived for both uniform linear arrays (ULAs) and uniform circular arrays (UCAs). Then, the asymptotic CRBs are analyzed. It is rigorously proved that: 1) as the number of antennas N increases, the maximum decay rates of asymptotic CRBs are 1/N for ULAs and 1/N^2 for UCAs; 2) as the number of subcarriers M increases, the asymptotic CRBs decay as 1/M^3 for both ULAs and UCAs; and 3) CRBs are inversely proportional to the beamforming gain. Based on the analytical results, two practical beamforming approaches are proposed for near-field wideband integrated sensing and communication (ISAC), namely independent and joint approaches. For the independent approach, the beamformer on each subcarrier is designed exclusively for either sensing or communication. For the joint approach, the beamformer on each subcarrier is jointly optimized for both functions through a low-complexity iterative algorithm. Finally, numerical results show that 1) large bandwidth sets an estimation error ceiling for NISE; 2) NISE performance converges to far-field sensing performance when the bandwidth is extremely large; 3) there is a tradeoff between array size and system bandwidth for achieving a given sensing performance; and 4) the simple independent beamforming approach achieves an ISAC performance close to the complex joint beamforming approach.
Comments: 13 pages, 11 figures
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Cite as: arXiv:2404.05076 [cs.IT]
  (or arXiv:2404.05076v1 [cs.IT] for this version)

Submission history

From: Zhaolin Wang [view email]
[v1] Sun, 7 Apr 2024 21:22:42 GMT (1079kb,D)

Link back to: arXiv, form interface, contact.