optical data transmission (original) (raw)
Definition: the transmission of information using light beams, e.g. in fibers
Category:
lightwave communications
- optical data transmission
Related: optical fiber communicationstelecom transceiverstelecom transmitterstelecom receiversfiber-optic linksfiber to the homeradio and microwave over fiberquantum cryptographyfree-space optical communications
Page views in 12 months: 1002
DOI: 10.61835/641 Cite the article: BibTex BibLaTex plain textHTML Link to this page! LinkedIn
Content quality and neutrality are maintained according to our editorial policy.
đŚ For purchasing optical data transmission, use the RP Photonics Buyer's Guide â an expert-curated directory for finding all relevant suppliers, which also offers advanced purchasing assistance.
Contents
Introduction
Light has an enormous potential for data transmission with very high data rates. This is basically a consequence of the high optical frequencies, which also make it possible to utilize very broad optical bandwidths. For example, the wavelength range from 1.3 to 1.6 Îźm, which may be transmitted in an optical fiber, corresponds to a bandwidth as large as 43 THz, which is orders of magnitude higher than that of any electrical cable. Although the theoretical potential of this bandwidth can so far not be fully utilized, an optical link (either a free-space link or a fiber-optic link) can have a capacity far beyond that of an electrical cable, or of a radio frequency link. There have already been demonstrations of fiber-optical data transmission systems with dozens of Tbit/s, using a single fiber. With multi-core fibers, the capacity could be scaled further.
Laser-based communications is clearly the most important laser application in terms of global laser sales, and in addition still exhibits strong growth.
Guided versus Free-space Transmission
There are two fundamentally different methods of optical data transmission:
- In most cases, one uses optical fibers as the transmission medium (â optical fiber communications) because light can be guided in fibers over very long distances with very low losses, also avoiding alignment issues, atmospheric influences and the like.
- However, there are also applications for free-space optical communications, mostly based on light beams, e.g. between Earth-orbiting satellites, between a remote spacecraft and an Earth-based station, or over short distances between metropolitan buildings.
Technical Details
Wavelength Regions
Optical transmission typically works in standard wavelength regions â particularly in case optical fiber communications, where where wavelength-dependent propagation losses and chromatic dispersion need to be considered.
The typically used wavelength regions are:
- 850 nm: suitable for multimode fibers as used in short-range (LAN/data center), but also for free-space transmission
- 1310 nm: zero-dispersion region of silica fibers, used for metro and access networks
- 1550 nm: minimum attenuation, compatible with erbium-doped fiber amplifiers (EDFAs), used for long-haul and DWDM systems, also for free-space transmission with relatively eye-safe lasers
Advanced systems may operate in extended bands (e.g., the L-band around 1625 nm) or emerging O/E/C/S bands for extra capacity.
Modulation Formats and Spectral Efficiency
In optical data transmission, information always must be encoded into a light signal using some modulation format:
- In the simplest case, one modulates only the optical power (less accurate: optical intensity), which can be monitored with a photodetector. This method is called intensity modulation with direct detection (IM/DD) and typically achieves a spectral efficiency of about 1 bit/s/Hz. (For example, for 1 Gbit/s, one requires roughly 1 GHz of optical bandwidth.) Two common variants of intensity modulation are:
- Non-return-to-zero (NRZ): Each bit is encoded as a low or high optical power, in practice with continuous power transitions between the bits.
- Return-to-zero (RZ): The optical power always returns to zero (or near zero) between bits. This provides better timing synchronization but requires more optical bandwidth.
- Both are variants of onâoff keying (OOK), where only two different intensity levels (high / close to zero) are used. Another possibility is PAM4 = four-level pulse amplitude modulation, where one uses four different intensity levels for encoding two bits instead of one.
- Substantially higher spectral efficiencies can be achieved with coherent transmission, where both the amplitude and phase are modulated and detected. This approach requires coherent detection, typically using a tunable narrow-linewidth laser as a local oscillator in a heterodyne or homodyne receiver.
- A common method is quadrature amplitude modulation (QAM), where the optical field is represented by a field amplitude in the 2D complex plane (also called I/Q or quadrature space), and each symbol corresponds to a point (or practically a certain region) in this space. For example, 16-QAM uses 24 = 16 distinct constellation points, allowing the transmission of 4 bits per symbol and thus enabling a spectral efficiency of roughly 4 bit/s/Hz. Higher-order formats like 64-QAM or 256-QAM can further increase this efficiency to 6 or 8 bit/s/Hz, but require better signal-to-noise ratios.
- An additional doubling of spectral efficiency is possible through polarization-division multiplexing (PDM), where two independent data streams are transmitted on orthogonal polarization states of light. Although optical transmission fibers are generally not polarization-maintaining), polarization multiplexing is still feasible using sophisticated digital signal processing (DSP) techniques at the receiver to separate and demodulate the two polarization components.
The following table gives a brief overview of the use of digital modulation formats:
| Application | Typical format | Data rate | Example technology |
|---|---|---|---|
| Ethernet | NRZ or PAM4 | 10â400 Gb/s | VCSEL |
| Metro DWDM | OOK or DPSK | 10â40 Gb/s | DFB + EAM |
| Long-haul | QPSK / 16-QAM (coherent) | 100â800 Gb/s | DFB/DBR + MZM |
| Data center | PAM4 | 50â800 Gb/s | VCSEL or SiPh Tx |
There are also methods of analog transmission, e.g. used for 6G / radio-over-fiber, where DFB or tunable lasers are used.
Bit Error Issues
Causes of Bit Errors
Even in digital optical transmission, received data are not entirely error-free. Various types of noise influences, often increased by certain imperfections, can cause a small fraction of transmitted bits to be received incorrectly. Several key sources of noise contribute to these bit errors:
- Laser transmitters exhibit some laser noise and possibly additional non-ideal properties. For example, a power-modulated laser diode exhibits changes in instantaneous frequency which causes additional signal degradation during fiber transmission.
- When optical amplifiers are needed (in particular fiber amplifiers), some amount of amplifier noise is unavoidable (â quantum noise, amplified spontaneous emission), and noise beyond that is reflected with a non-ideal noise figure.
- Optical receivers introduce some excess noise, e.g. due to thermal fluctuations in electronics, but also quantum noise due to non-perfect quantum efficiency.
Error Correction
In modern transmission systems, virtually all of the occurring bit errors can be detected and corrected, so that the system as a whole is virtually error-free. These techniques may include:
- Forward Error Correction (FEC) means adding redundancy to the transmitted data to allow the receiver to correct certain errors without retransmission. One may then work with a substantial raw bit error rate and reduce the effective error rate to a very low level.
- Checksums or cyclic redundancy checks (CRC) allow detection of corrupted data. One can then apply retransmission protocols, particularly in packet-based systems like TCP/IP.
For such a system to work effectively and efficiently, the raw bit error rate (i.e., the fraction of incorrectly transmitted bits in the raw data) must be limited to some level. For example, an acceptable level may be 10â12 for a terrestrial fiber telecommunication system, or 10â6 for satellite control.
Influence of Signal Power
Typically, the bit error rate is strongly dependent on the transmitted optical power; therefore, the received power (affected by propagation losses in fibers plus insertion loss of additional components) must be high enough to keep the bit error rate acceptable.
If the system experiences impairments such as chromatic dispersion or background light (in free-space optical systems), additional optical power may be required to maintain the target BER. This extra power requirement is called a power penalty, or more specifically e.g. a dispersion penalty if chromatic dispersion is the considered factor.
For a given modulation format and a limited optical bandwidth, the maximum possible bit rate for data transmission depends on the signal-to-noise ratio of the transmission system, which itself depends on the received optical power among other factors. According to the ShannonâHartley theorem, the possible bit rate scales with the logarithm of 1 plus the signal-to-noise ratio. Particularly for high signal-to-noise ratios, fiber-optical transmission systems (see below) cannot fully realize that theoretical potential, since nonlinearities lead to signal distortions.
Devices Used in Optical Data Transmission
Optical data transmission systems employ specialized transmitters, receivers, and often integrated transceivers to convert electrical signals to light and back again:
- The transmitter generates the optical signal, usually with a semiconductor laser or LED, often modulated either directly or via an external modulator.
- The receiver detects the incoming light with a photodiode (pâiân or APD) or a coherent receiver front end and reconstructs the electrical data stream.
In practice, both functions are almost always combined into a transceiver module, which provides a complete bidirectional optical interface. Transceivers are typically realized as standardized pluggable modules (e.g. SFP, QSFP, OSFP, CFP) containing a transmitter optical subassembly (TOSA), a receiver optical subassembly (ROSA), driver and amplifier electronics, and digital diagnostics accessible through an I2C interface. Some modules employ direct detection (for short-reach and access links), while coherent detection with integrated DSP and FEC is used for long-haul and metro systems.
A recent trend is co-packaged optics (CPO), where optical engines are mounted directly beside the switching ASIC instead of in front-panel pluggable modules. CPO reduces electrical interconnect losses and power consumption at very high data rates, while retaining the same fundamental transceiver functions.
Further details on the operation of transmitters, receivers and transceivers are provided in the respective articles.
From Larger to Smaller Distances
Optical data transmission is increasingly used in various areas, such as telephony, Internet traffic, cable TV â mostly for larger transmission distances of at least a few kilometers, while shorter distances (e.g. from roadside cabinet to a home installation) are often still treated with electrical cables (e.g. telecom or TV cables). There is, however, a tendency for also using optical systems with smaller and smaller transmission distances. Particularly in Japan, many Internet connections are already delivered to homes with optical fibers (â fiber to the home). Current local area networks (LANs) work well with electrical cables for data rates of 1 Gbit/s, and 10 Gbit/s are also possible, but only over quite limited distances; electrical cables are expensive and lossy for use at microwave frequencies. Therefore, optical connections start to become more important even within buildings.
Supercomputers more and more extensively use optical interconnects even for quite short distances. Optical board-to-board connections, optical chip-to-chip and even intra-chip connections are being seriously considered, and are partly already developed. For example, an increasing portion of the area of current CPU chips is occupied by electrical data transmission lines. At high data rates, it is expected to be favorable in terms of space and also of energy consumption to realize such transmission lines with optical means. A challenge, however, is still to develop suitable micro- or nano-lasers, which can be operated with very low power levels, and can be fabricated on silicon wafers.
Another possible application (yet at the R & D stage) is short-distance free-space data transmission within rooms, e.g. from an LED-based transmitter at the ceiling to a PC, a notebook or a smaller mobile device. The transmitter does not send out a directed beam, but illuminates a large area. Due to the relatively small distances, one can still achieve a reasonably high signal-to-noise ratio for realizing high data rates. This approach is competing with traditional wireless communications e.g. based on Wi-Fi (WLAN). Possible advantages of optical transmission include higher data rates and avoiding interference or interception problems, as the sent-out light will usually not be detectable in other rooms. A problem, however, is the back channel, particularly for small mobile device which cannot use a powerful light emitter for reasons of power consumption. One possibility is to use radio transmission for the back channel. For many applications such as video streaming this can work well, since download data rates are far higher than upload rates.
Case Study
The following case study is available, which discusses some aspects of mode-locked fiber lasers:
Suppliers
Bibliography
| [1] | R.-J. Essiambre et al., âCapacity limits of optical fiber networksâ, J. Lightwave Technol. 28 (4), 662 (2010); doi:10.1109/JLT.2009.2039464 |
|---|---|
| [2] | R. S. Tucker, âGreen optical communications â part I: energy limitations in transportâ, IEEE J. Sel. Top. Quantum Electron. 17 (2), 245 (2011); doi:10.1109/JSTQE.2010.2051216 |
| [3] | R. S. Tucker, âGreen optical communications â part II: energy limitations in networksâ, IEEE J. Sel. Top. Quantum Electron. 17 (2), 261 (2011); doi:10.1109/JSTQE.2010.2051217 |
| [4] | D. A. B. Miller, âWaves, modes, communications, and optics: a tutorialâ, Advances in Optics and Photonics 11 (3), 679 (2019); doi:10.1364/AOP.11.000679 |
| [5] | Harry J. R. Dutton, Understanding optical communications, http://www.freeinfosociety.com/media/pdf/5475.pdf, IBM Redbooks |
| [6] | G. P. Agrawal, Fiber-Optic Communication Systems, John Wiley & Sons, New York (2002) |
(Suggest additional literature!)
Questions and Comments from Users
Here you can submit questions and comments. As far as they get accepted by the author, they will appear above this paragraph together with the authorâs answer. The author will decide on acceptance based on certain criteria. Essentially, the issue must be of sufficiently broad interest.
Please do not enter personal data here. (See also our privacy declaration.) If you wish to receive personal feedback or consultancy from the author, please contact him, e.g. via e-mail.
By submitting the information, you give your consent to the potential publication of your inputs on our website according to our rules. (If you later retract your consent, we will delete those inputs.) As your inputs are first reviewed by the author, they may be published with some delay.