Summary

Chapter 14 Summary: Orthogonal Frequency Division Multiplexing (OFDM)

Key Points

  • 1.

    Parallel channel decomposition. OFDM converts a frequency-selective channel into NN independent flat-fading sub-channels by dividing the wideband signal into narrow orthogonal subcarriers spaced at Ξ”f=1/Tsym\Delta f = 1/T_{\text{sym}}. Each subcarrier requires only a single-tap equaliser: X^[k]=Y[k]/H[k]\hat{X}[k] = Y[k]/H[k].

  • 2.

    DFT-based implementation with cyclic prefix. The OFDM transmitter is an IDFT and the receiver is a DFT, both efficiently computed via the FFT. The cyclic prefix (of length Ncpβ‰₯Lβˆ’1N_{\text{cp}} \geq L - 1) converts linear channel convolution into circular convolution, eliminating both ISI and ICI. The CP overhead Ξ·CP=Ncp/(N+Ncp)\eta_{\text{CP}} = N_{\text{cp}}/(N + N_{\text{cp}}) is typically 5--7% in LTE/NR.

  • 3.

    Channel estimation via pilots. Known pilot symbols inserted into the time-frequency grid enable channel estimation. The LS estimator H^[k]=Yp[k]/Xp[k]\hat{H}[k] = Y_p[k]/X_p[k] is simple but noise-enhanced; the MMSE estimator exploits frequency-domain correlation for lower MSE at the cost of requiring channel statistics. Pilot spacing must satisfy the Nyquist condition in both time (Doppler) and frequency (delay spread).

  • 4.

    PAPR is the major transmitter challenge. The OFDM signal has a maximum PAPR of NN (= 10log⁑10N10\log_{10} N dB) when subcarriers add constructively. Practical PAPR at CCDF=10βˆ’3\text{CCDF} = 10^{-3} is typically 10--12 dB for N=1024N = 1024. Reduction techniques include clipping (with filtering), selected mapping (SLM), and tone reservation.

  • 5.

    Synchronisation is critical. A normalised fractional CFO Ο΅F\epsilon_F creates an ICI floor of approximately 3/(Ο€2Ο΅F2)3/(\pi^2 \epsilon_F^2), limiting achievable SINR regardless of transmit power. Both time and frequency synchronisation are essential, with preamble-based methods (Schmidl-Cox) for acquisition and pilot-based tracking for residual correction.

  • 6.

    AMC maximises throughput. OFDM's per-subcarrier flat-fading model enables adaptive modulation and coding: subcarriers with high SNR use high-order modulation (64/256-QAM) and high code rates, while weak subcarriers use QPSK or are nulled. CQI-based link adaptation is fundamental to LTE and 5G NR.

  • 7.

    Generalisations address OFDM limitations. SC-FDMA (DFT-spread OFDM) reduces PAPR for uplink via DFT precoding; FBMC improves spectral containment by replacing rectangular pulses with well-localised filters; OTFS operates in the delay-Doppler domain for high-mobility resilience. CP-OFDM remains dominant due to MIMO compatibility and implementation maturity.

Looking Ahead

Chapter 15 introduces MIMO systems, where the per-subcarrier flat-fading model Y[k]=H[k]X[k]+W[k]Y[k] = H[k] X[k] + W[k] becomes a matrix equation Y[k]=H[k]X[k]+W[k]\mathbf{Y}[k] = \mathbf{H}[k] \mathbf{X}[k] + \mathbf{W}[k]. OFDM's parallel channel decomposition is the key enabler: it converts the wideband MIMO channel into NN independent narrowband MIMO channels, each amenable to spatial multiplexing, beamforming, and space-time coding. Every modern wireless standard β€” from 802.11n to 5G NR massive MIMO β€” is built on MIMO-OFDM.