Generalized maximum-likelihood sequence detection for photon-counting Free Space Optical systems

Nestor D. Chatzidiamantis, George K. Karagiannidis, Murat Uysal

Research output: Contribution to journalConference articlepeer-review


We investigate detection methods for on-off keying (OOK) photon-counting Free Space Optical (FSO) systems in the presence of turbulence-induced fading, assuming no channel state information at the receiver. To recover the performance loss which is associated with symbol-by-symbol detection in such a scenario, we consider sequence detection techniques, exploiting the temporal correlation of the FSO channel. Due to its high complexity in the calculation of its metric, optimal maximum likelihood sequence detection (MLSD) is infeasible for most practical purposes. Hence, we propose a suboptimal low-complexity detection rule, which is based on the generalized maximum-likelihood sequence estimation. The proposed scheme allows the detection of sequence lengths that are prohibitive for conventional MLSD, without using any kind of channel knowledge. Monte Carlo simulation results show its performance to be very close to the optimum for large sequence lengths and various fading models.

Original languageEnglish (US)
Article number665359
Pages (from-to)3381-3385
Number of pages5
JournalIEEE Workshop on Local and Metropolitan Area Networks
StatePublished - 1993
Event6th IEEE Workshop on Local and Metropolitan Area Networks, LANMAN 1993 - San Diego, United States
Duration: Oct 14 1996Oct 16 1996


  • Free-space optical systems
  • Generalized maximum likelihood sequence detection
  • Maximum likelihood sequence detection
  • Poisson photon counting model
  • Turbulence-induced fading

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Software
  • Electrical and Electronic Engineering
  • Communication


Dive into the research topics of 'Generalized maximum-likelihood sequence detection for photon-counting Free Space Optical systems'. Together they form a unique fingerprint.

Cite this