The sensitivity characteristics of optical receiver frontends for high-speed data communications depend on modulation format, detector type, and specific operational constraints. A general mathematical model of the receiver sensitivity that fits to analytical as well as measured data is required to compare different receiver implementations and assess the reliability of data links under varying received power as common in free-space optical communication links. In this paper, a new approach based on $Q$-factor modeling is presented, compared with analytical receiver models, and applied to a multitude of exemplary receiver implementations. A methodology is introduced to generally apply the model to ideal or practical binary optical receiver frontends.

Christos K. Datsikas, Kostas P. Peppas, Nikos C. Sagias, and George S. Tombras J. Opt. Commun. Netw. 2(8) 576-586 (2010)

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

You do not have subscription access to this journal. Article level metrics are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

The parameters are as follows: ${\sigma}_{t}={i}_{n}\xb7\sqrt{B}$ is the thermal noise current, with thermal noise current density ${i}_{n}$; ${\sigma}_{s,1}^{2}=2e{M}^{2}{F}_{A}RB\xb7(2{\overline{P}}_{Rx})$ is the signal shot noise current variance in the APD, during the reception of a binary one with instantaneous power $2{\overline{P}}_{Rx}$; $M$ is the multiplication gain factor; ${F}_{A}$ is the excess noise factor of the APD; $R$ is the detector responsivity (in A/W); and $e$ is the elementary charge.

Table 2.

Value Ranges of the Receiver Model Parameters

Parameter

Typical Range

Explanation

${\overline{E}}_{Q=2}$

0.2–700 aJ

sensitivity per bit in ${10}^{-18}\text{\hspace{0.17em}}\text{\hspace{0.17em}}\mathrm{J}$

$s$

typical $3<s<9$

shot-noise-limited: $s=9$, $n=0.5$

$n$

typical $0.5<n<1$

thermal-limited: $s=3$, $n=1$

Table 3.

Measured Receiver Model Parameters Summary^{a,}^{b}

Sorted by mean energy per bit ${\overline{E}}_{Q=2}$.
All measured RFEs except {1} and {2} are tested with 1550 nm wavelength, $R=1\text{\hspace{0.17em}}\text{\hspace{0.17em}}\mathrm{A}/\mathrm{W}$ is assumed, and PRBS is ${2}^{7}-1$ unless otherwise noted.
M-fix indicates that the multiplication factor $M$ of the APD is kept constant for all values of ${\overline{P}}_{Rx}$.
M-opt indicates that $M$ is optimized for the minimum BER at each measured ${\overline{P}}_{Rx}$.
RMSRE refers to the root mean square of the relative error between the measured and fitted data points of $Q$ versus ${\overline{P}}_{Rx}$.

The parameters are as follows: ${\sigma}_{t}={i}_{n}\xb7\sqrt{B}$ is the thermal noise current, with thermal noise current density ${i}_{n}$; ${\sigma}_{s,1}^{2}=2e{M}^{2}{F}_{A}RB\xb7(2{\overline{P}}_{Rx})$ is the signal shot noise current variance in the APD, during the reception of a binary one with instantaneous power $2{\overline{P}}_{Rx}$; $M$ is the multiplication gain factor; ${F}_{A}$ is the excess noise factor of the APD; $R$ is the detector responsivity (in A/W); and $e$ is the elementary charge.

Table 2.

Value Ranges of the Receiver Model Parameters

Parameter

Typical Range

Explanation

${\overline{E}}_{Q=2}$

0.2–700 aJ

sensitivity per bit in ${10}^{-18}\text{\hspace{0.17em}}\text{\hspace{0.17em}}\mathrm{J}$

$s$

typical $3<s<9$

shot-noise-limited: $s=9$, $n=0.5$

$n$

typical $0.5<n<1$

thermal-limited: $s=3$, $n=1$

Table 3.

Measured Receiver Model Parameters Summary^{a,}^{b}

Sorted by mean energy per bit ${\overline{E}}_{Q=2}$.
All measured RFEs except {1} and {2} are tested with 1550 nm wavelength, $R=1\text{\hspace{0.17em}}\text{\hspace{0.17em}}\mathrm{A}/\mathrm{W}$ is assumed, and PRBS is ${2}^{7}-1$ unless otherwise noted.
M-fix indicates that the multiplication factor $M$ of the APD is kept constant for all values of ${\overline{P}}_{Rx}$.
M-opt indicates that $M$ is optimized for the minimum BER at each measured ${\overline{P}}_{Rx}$.
RMSRE refers to the root mean square of the relative error between the measured and fitted data points of $Q$ versus ${\overline{P}}_{Rx}$.