130
M. El Ghzaoui and S. Das
7 THz 16 QAM/QPSK Transmission
High data rate over wireless channels at THz frequencies is a challenging task because
of the high bit error rate in these channels. THz communication with high efficiency
has been demonstrated with advanced modulation schemes such as QPSK and 16QAM which are crucial to optimize spectral efficiency over wireless networks [24].
To increase capacity, optimization of the channel bandwidth is required. One of the
promising candidates is to use modulation format with a high modulation index; that
is, the modulation format would be enhanced from QPSK to 16-QAM, 64-QAM,
and so on. However, the required resolution of a digital-to-analog converter set at the
transmitter side and an analog-to-digital converter, which is located at the receiver,
would become higher. Thus, increasing the degree of the multi-levels is difficult
because of its electrical device issue. In addition, on the receiver side, it is necessary to
use high resolution analog-to-digital converter (ADC), which generally translates into
high power consumption per conversion. Therefore, the ADC resolution bits must be
kept as low as possible to decrease the cost and power consumption. Consequently,
increasing the bits resolution of an ADC will lead to high precision DSP elements,
which are unwanted for wireless transceivers.
Another possible solution is a space-division-multiplexing technique, so-called a
multi-input-multi-output (MIMO) technique. The spectral efficiency in the space is
augmented easily as the number of the antenna at the transmitter and receiver side
increases, and thus, the total capacity of the MIMO link will increase. Moreover, the
diversity effect under multi-antenna configuration can enhance an effective SNR.
7.1 Power Spectral Density
As it was mentioned before, GMSK is a variant of MSK. We can also note that with
the introduction of Gaussian filtering, the GMSK signal can no longer be viewed as
Offset-QPSK.
Usually, the receiver of GMSK is a linear estimation of the MSK, and this technique might be treated as a sum of pulse PAM signals. Based on this approximation,
we can use the Viterbi detector to receive the signal. Notice that lower complexity
demodulator for GMSK is typically used for low power devices, such as BTLE. In
Figs. 3 and 4, we compare the PSD for MSK and GMSK with the PSD for QPSK
with oversampling factor of 16, respectively.
Figures 3 and 4 show that the spectral efficiency of GMSK modulation is significantly better than the spectral efficiency of QPSK and MSK modulations. One can
be seen from this figure that the width of the spectrum of GMSK is mainly narrow
for the low level of the signal (−60 dBm). Consequently, the GMSK modulation
does not disturb adjacent radio channels.
In addition, the spectrum of the MSK modulation occupies the same main lobe
spectrum as the QPSK modulation.
M. El Ghzaoui and S. Das
7 THz 16 QAM/QPSK Transmission
High data rate over wireless channels at THz frequencies is a challenging task because
of the high bit error rate in these channels. THz communication with high efficiency
has been demonstrated with advanced modulation schemes such as QPSK and 16QAM which are crucial to optimize spectral efficiency over wireless networks [24].
To increase capacity, optimization of the channel bandwidth is required. One of the
promising candidates is to use modulation format with a high modulation index; that
is, the modulation format would be enhanced from QPSK to 16-QAM, 64-QAM,
and so on. However, the required resolution of a digital-to-analog converter set at the
transmitter side and an analog-to-digital converter, which is located at the receiver,
would become higher. Thus, increasing the degree of the multi-levels is difficult
because of its electrical device issue. In addition, on the receiver side, it is necessary to
use high resolution analog-to-digital converter (ADC), which generally translates into
high power consumption per conversion. Therefore, the ADC resolution bits must be
kept as low as possible to decrease the cost and power consumption. Consequently,
increasing the bits resolution of an ADC will lead to high precision DSP elements,
which are unwanted for wireless transceivers.
Another possible solution is a space-division-multiplexing technique, so-called a
multi-input-multi-output (MIMO) technique. The spectral efficiency in the space is
augmented easily as the number of the antenna at the transmitter and receiver side
increases, and thus, the total capacity of the MIMO link will increase. Moreover, the
diversity effect under multi-antenna configuration can enhance an effective SNR.
7.1 Power Spectral Density
As it was mentioned before, GMSK is a variant of MSK. We can also note that with
the introduction of Gaussian filtering, the GMSK signal can no longer be viewed as
Offset-QPSK.
Usually, the receiver of GMSK is a linear estimation of the MSK, and this technique might be treated as a sum of pulse PAM signals. Based on this approximation,
we can use the Viterbi detector to receive the signal. Notice that lower complexity
demodulator for GMSK is typically used for low power devices, such as BTLE. In
Figs. 3 and 4, we compare the PSD for MSK and GMSK with the PSD for QPSK
with oversampling factor of 16, respectively.
Figures 3 and 4 show that the spectral efficiency of GMSK modulation is significantly better than the spectral efficiency of QPSK and MSK modulations. One can
be seen from this figure that the width of the spectrum of GMSK is mainly narrow
for the low level of the signal (−60 dBm). Consequently, the GMSK modulation
does not disturb adjacent radio channels.
In addition, the spectrum of the MSK modulation occupies the same main lobe
spectrum as the QPSK modulation.