Summary: ThroughputMaximizing FIR Transmit Filters for Linear
Naofal AlDhahir \Lambda
OptimumFIR transmit filters for symbol--by--symboltransmission on linear dispersive additive--Gaussian--
noise channels are derived by maximizing the channel throughput, subject to a fixed average input energy
constraint. This maximized throughput is compared with that achievable with water--pour and flat transmit
filters. The effect of transmit filter optimization on the receiver performance is investigated by considering
the popular MMSE--DFE receiver structure.
Maximizing the achievable throughput of noisy intersymbol interference (ISI) channels requires
optimization of both the transmitter and receiver ends of the communication system. In multicarrier
transceivers, an orthogonal transformation (such as FFT) is used to convert the wideband ISI channel
into a large number of parallel ISI--free narrow--band subchannels that can be individually decoded.
On the other hand, single carrier transceivers commonly employ finite--impulse--response (FIR) filters
at both the transmitter and receiver ends to mitigate the ISI and noise.
The FIR minimum--mean--square--error decision feedback equalizer (MMSE--DFE) is a widely--used
receiver structure on severe--ISI channels. Optimizing the MMSE--DFE receiver filter settings for the
infinite--length and finite--length cases was treated in [8, 5] and , respectively. While there have
been several studies on the transmitter optimization problem, they either assume an infinite--length