Hybrid Time–Frequency Domain Range Estimation for ISAC: Balancing Accuracy and Complexity


Yazgan M., Karahan B., Raeisi M., ARSLAN H.

IEEE Transactions on Communications, 2026 (SCI-Expanded, Scopus)

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2026
  • Doi Numarası: 10.1109/tcomm.2026.3747975
  • Dergi Adı: IEEE Transactions on Communications
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Compendex, INSPEC, zbMATH, Materials Science & Engineering Collection (ProQuest), Technology Collection (ProQuest)
  • Anahtar Kelimeler: Cramér-Rao lower bound (CRLB), deterministic complexity, frequency-domain phase estimation (FDPE), high accuracy, integrated sensing and communications (ISAC), monostatic OFDM radar
  • İstanbul Medipol Üniversitesi Adresli: Evet

Özet

In fifth-generation (5G)-Advanced and sixth-generation (6G) systems, integrated sensing and communications (ISAC) is expected to gain significant role in enabling intelligent applications requiring millimeter-level ranging accuracy and low latency. Achieving high range estimation accuracy is constrained by a trade-off with computational complexity in monostatic orthogonal frequency-division multiplexing (OFDM)-based ISAC systems. Traditional time-domain approaches are restricted by the resolution, whereas grid-search-based methods improve accuracy with increased complexity due to dense delay sampling. To handle these limitations, this work introduces a hybrid approach where the first stage exploits time-domain cross-correlation to obtain a rough estimate of the delay avoiding exhaustive dense global search while performing global coarse delay acquisition. This is followed by a refinement stage based on frequency-domain phase estimation (FDPE), which allows sub-grid accuracy employing vectorized processing. Compared to maximum likelihood estimation (MLE), which involves iterative global search over the entire unambiguous delay interval resulting in latency, the proposed method offers reduced computational performance. While preserving millimeter-level accuracy with exceptionally lower complexity, simulation results exploiting realistic 5G numerology demonstrate that the proposed method achieves near Cramér–Rao lower bound (CRLB) performance over a wide range of signal-to-noise ratios (SNRs).