Class 3: Supervised by Enas Osama Taha

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

Ministry Of Higher Education

Northern Technical University


Technical College /Mosul
Department Of Medical Instruments

Student Name
Ahmed Mutasim Mutashar Nasir
Sdeq Mufaq Sdeq
Mohammed Younis Khalaf
Ibrahim Ahmed Khalil

Supervised By
Enas Osama Taha

Class 3rd B
2020-------------------------------------------------1441
Orthogonal Frequency Division Multiplexing (OFDM)
Orthogonal Frequency Division Multiplexing (OFDM) has been applied in many communication
fields; in copper and wireless applications it is a mature technology with wide commercial
availability. Since 2005, research on optical domain OFDM technology has been conducted and
has become a very hot field. However, the need for high speed DAC / ADC components and the
relatively high complexity has hampered its adoption in optical access networks.

Depending on the modulation, the optical domain OFDM technology is divided into Coherent
Detection OFDM (CD-OFDM) and Direct Detection OFDM (DD-OFDM). DD-OFDM has
lower cost and smaller packaging, so it’s more likely to be adopted in optical access.

The basic electrical architecture for a DD-OFDM transmitter (TX) and receiver (RX) is shown in
Fig.1.

In the TX side, incoming serial data is converted to a parallel format, mapped to symbols from
Quadrature Amplitude Modulation (QAM) constellation, and then applied to a n-point Inverse
FFT (i.e. IFFT) to generate a digital OFDM signal with n orthogonal subcarriers. A cyclic prefix
is added to minimize inter-symbol interference (ISI) and then the output data is serialized and
converted to an analog signal using high speed DAC technology. The analog signal is converted
to an optical signal using a laser for optical domain applications.

In the RX side, signal processing and data flow are opposite to the TX side. The incoming signal
is first converted to an electrical signal using an O/E converter and then applied to an ADC. The
serial ADC output is converted to parallel and the cyclic prefix is removed. An FFT is used to
decode the n OFDM subcarriers, and an equalizer may be used to compensate for chromatic
dispersion. After equalization the FFT the signal is sent to a QAM symbol detection module.
Finally, the received signal is serialized to recover the transmitted data.

Figure 1: Base electrical physical architecture of DD-OFDM

In addition, one point should be explained. In normal practice the TX side uses two DAC chips
and the RX side two ADC chips. This is because the Inverse FFT output is complex signal (i.e.
I+jQ), and it is divided into in-phase (I) component and quadrature (Q) component and each are

1
transmitted through separate DAC channels. The question then becomes how to use a single laser
to transmit the two signals. There are several methods to accomplish this. One method is to use
an electrical analog I/Q modulator to combine the two signals into one. In this case the RX side
can use a single electrical analog I/Q demodulator.

There is one way to avoid the need for two DACs and ADCs. The Fourier transform of a real
number sequence has conjugate symmetry, so when QAM symbols are applied to an Inverse FFT,
they are mapped to half of the entire subcarriers set, and the other half is assigned as the
conjugate of this set. The process is illustrated in Fig.2. Using conjugate symmetry method, a
single DAC in the TX and single ADC in the RX are needed.

Figure 2: Two mapping method

In OFDM systems multiplexing can be accomplished in either the time domain or the frequency
domain. In strict time domain multiplexing the entire OFDM frequency spectrum would be
assigned to a single user for some number of symbols, In frequency domain multiplexing a user
is assigned one or more subcarriers as illustrated in Figure3. By assigning groups of users to
different subcarrier sets the advantages of both time and frequency domain multiplexing can be
realized.

Figure 3: OFDM multiplexing

2
Figure 4: Architecture of DD-OFDM PON

One possible DD-OFDM PON architecture is shown in Fig.4, detailed component requirements
are shown in Table 1.

Table 1: DD-OFDM PON Parameters

Modulation 10 Gb/s 25 Gb/s 40 Gb/s 100 Gb/s

Bandwidth 25 GHz

(OFDM+16QA 2.5GHz 6.25 GHz 10 GHz


M)

2.5G 10G 10G 25G


LASER
DML/EML DML/EML DML/EML DML/EML

PD 2.5G APD 10G APD 10G APD 25G APD

3
DAC 5GS/s 12.5GS/s 20GS/s 50GS/s

ADC 5GS/s 12.5GS/s 20GS/s 50GS/s

OFDM has higher spectral efficiency than NRZ, PAM4 and duobinary. For 40Gbps OFDM
modulation, 10G optical components are sufficient, but component cost savings are transferred to
electrical high speed ADC/DAC components. This has become one of the major limitations of
OFDM applications in access.

Another limiting factor is the optical power budget. OFDM has good dispersion resistance
performance, better than PAM and duobinary, but has higher linearity requirements than either
PAM or duobinary. Both the nonlinearity of the optical components and the electrical
components affect the system performance and reduce the overall power budget.

Power budget analysis[NG-PON2 WP]

For 40Gbps downstream transmission, the OLT minimum launch power can be set to +10dBm
after optical amplification. When the direct detection of optical APD receiver is employed at
ONU, the achievable sensitivity at the BER (bit error rate) level of 1×10-3 (pre-FEC) is -21dBm.

For 10Gbps upstream transmission, the pre-amplifier at OLT can be added to increase the
received sensitivity and APD is employed for direct detection, the achievable sensitivity at the
pre-FEC BER level of 1×10-3 is -26dBm. FEC and equalizer can be used to reduce BER.

You might also like