Difference between revisions of "Aufgaben:Exercise 5.9: Selection of OFDM Parameters"

From LNTwww
m (Text replacement - "„" to """)
 
(5 intermediate revisions by 3 users not shown)
Line 1: Line 1:
  
{{quiz-Header|Buchseite=Modulationsverfahren/OFDM für 4G–Netze
+
{{quiz-Header|Buchseite=Modulation_Methods/OFDM_for_4G_Networks
 
}}
 
}}
  
[[File:P_ID1668__Mod_A_5_9.png|right|frame|Zeitabhängiger Dämfungsverlauf <br>zweier Mobilfunkkanäle]]
+
[[File:P_ID1668__Mod_A_5_9.png|right|frame|Time-dependent attenuation curve
In dieser Aufgabe sollen einige OFDM–Parameter eines Mobilfunksystems bestimmt werden.&nbsp;  
+
of two mobile radio channels]]
 +
In this exercise,&nbsp; some OFDM parameters of a mobile radio system are to be determined. &nbsp;  
  
Dabei wird von folgenden Voraussetzungen ausgegangen:
+
The following assumptions are made:
* Die Kohärenzzeit des Kanals ist &nbsp;$T_{\rm coh} = 0.4 \ \rm ms$.
+
* The coherence time of the channel is &nbsp;$T_{\rm coh} = 0.4 \ \rm ms$.
* Die maximale Pfadverzögerung sei &nbsp;$τ_{\rm max} = 25 \ \rm  &micro; s$.
+
* The maximum path delay is &nbsp;$τ_{\rm max} = 25 \ \rm  &micro; s$.
* Die Datenrate (Bitrate) beträgt &nbsp;$R_{\rm B} = 1 \ \rm  Mbit/s$.
+
* The data rate&nbsp; (bit rate)&nbsp; is &nbsp;$R_{\rm B} = 1 \ \rm  Mbit/s$.
* Alle Unterträger werden&nbsp; $\rm 4–QAM$–moduliert.
+
* All subcarriers are&nbsp; $\rm 4–QAM$ modulated.
  
  
Um eine gewisse Robustheit des Systems gegenüber zeit– und frequenzselektivem Fading zu gewährleisten, muss die folgende Ungleichung erfüllt werden:
+
To ensure some robustness of the system to time and frequency selective fading,&nbsp; the following inequality must be satisfied:
 
:$$T_{\rm{G}} \ll T \ll T_{{\rm{coh}}} - T_{\rm{G}}.$$
 
:$$T_{\rm{G}} \ll T \ll T_{{\rm{coh}}} - T_{\rm{G}}.$$
Insgesamt soll folgendermaßen vorgegangen werden:
+
Overall,&nbsp; the following procedure should be followed:
* Vorläufige Festlegung des Guard–Intervalls &nbsp;$(T_{\rm G}')$,
+
* Preliminary determination of the guard interval duration &nbsp;$(T_{\rm G}')$,
* Bestimmung der optimalen Kernsymboldauer &nbsp;$(T)$,
+
* Determination of the optimal core symbol duration &nbsp;$(T)$,
* entsprechende Festlegung der Stützstellenzahl der&nbsp; $\rm FFT$.
+
* corresponding determination of the number of sampling points of the&nbsp; $\rm FFT$.
  
  
Danach ist eventuell eine erneute Bestimmung einiger Systemgrößen aufgrund der bei den Berechnungen vorgenommenen Rundungen erforderlich.
+
After that,&nbsp; it may be necessary to redetermine some system quantities due to the rounding performed during the calculations.
  
Die Grafik zeigt zwei beispielhafte Dämpfungsverläufe von Mobilfunksystemen in logarithmischer Darstellung.  
+
The diagram shows two exemplary attenuation curves of mobile radio systems in logarithmic representation.
*Bei der blauen Kurve geschehen die zeitlichen Veränderungen relativ langsam, bei der roten Kurve viermal so schnell.  
+
*In the case of the blue curve,&nbsp; the changes over time occur relatively slowly,&nbsp; while in the case of the red curve they occur four times as fast.
*Demzufolge weist der blaue Kanal eine viermal größere Kohärenzzeit auf als der rote Kanal.
+
*Consequently,&nbsp; the blue channel has a four times larger coherence time than the red channel.
  
  
Line 33: Line 34:
  
  
''Hinweise:''
+
Notes:  
*Die Aufgabe gehört zum  Kapitel&nbsp; [[Modulation_Methods/OFDM_für_4G–Netze|OFDM für 4G–Netze]].
+
*The exercise belongs to the chapter&nbsp; [[Modulation_Methods/OFDM_für_4G–Netze|OFDM for 4G Networks]].
*Bezug genommen wird insbesondere auf die Seite&nbsp;  [[Modulation_Methods/OFDM_für_4G–Netze#Bestimmung_einiger_OFDM.E2.80.93Parameter|Bestimmung einiger OFDM-Parameter]].
+
*Reference is made in particular to the section&nbsp;  [[Modulation_Methods/OFDM_for_4G_Networks#Determination_of_some_OFDM_parameters|Determination of some OFDM parameters]].
* Weitere Informationen zum Thema finden Sie im LNTwww&ndash;Buch &nbsp;[[Mobile Kommunikation]].  
+
* Further information on the topic can be found in the LNTwww book &nbsp;[[Mobile Communications]].  
 
   
 
   
  
Line 42: Line 43:
  
  
===Fragebogen===
+
===Questions===
  
 
<quiz display=simple>
 
<quiz display=simple>
{Bestimmen Sie die minimal sinnvolle Dauer &nbsp;$T_{\rm G}'$&nbsp; des "vorläufigen Guard–Intervalls&rdquo;.
+
{Determine the minimum reasonable duration &nbsp;$T_{\rm G}'$&nbsp; of the&nbsp; "preliminary guard interval".
 
|type="{}"}
 
|type="{}"}
 
$T_{\rm G}' \ = \ $  { 25 3% } $\ \rm &micro; s$
 
$T_{\rm G}' \ = \ $  { 25 3% } $\ \rm &micro; s$
  
{Bestimmen Sie die optimale Kernsymboldauer &nbsp;$T_{\rm opt}$&nbsp; als geometrisches Mittel.
+
{Determine the optimal core symbol duration &nbsp;$T_{\rm opt}$&nbsp; as a geometric mean.
 
|type="{}"}
 
|type="{}"}
 
$T_{\rm opt} \ = \ $ { 97 3% } $\ \rm &micro; s$
 
$T_{\rm opt} \ = \ $ { 97 3% } $\ \rm &micro; s$
  
  
{Bestimmen Sie die benötigte Anzahl an Nutzträgern.
+
{Determine the required number of useful carriers.
 
|type="{}"}
 
|type="{}"}
$N_{\rm Nutz} \ = \ $ { 61 1% }  
+
$N_{\rm user} \ = \ $ { 61 1% }  
  
{Geben Sie die daraus resultierende Stützstellenzahl der FFT an.
+
{Specify the resulting number of interpolation points of the FFT.
 
|type="{}"}
 
|type="{}"}
 
$N_{\rm FFT} \ = \ $ { 64 1% }
 
$N_{\rm FFT} \ = \ $ { 64 1% }
  
{Berechnen Sie die Anzahl &nbsp;$N_{\rm G}$&nbsp; der Zeitabtastwerte des Guard–Intervalls und daraus die neue resultierende Schutzzeit &nbsp;$T_{\rm G}$.
+
{Calculate the number &nbsp;$N_{\rm G}$&nbsp; of time samples of the guard interval and from this the new resulting guard duration &nbsp;$T_{\rm G}$.
 
|type="{}"}
 
|type="{}"}
 
$N_{\rm G} \ = \ $ { 17 1% }  
 
$N_{\rm G} \ = \ $ { 17 1% }  
 
$T_{\rm G} \ = \ $ { 26 1% } $\ \rm &micro; s$
 
$T_{\rm G} \ = \ $ { 26 1% } $\ \rm &micro; s$
  
{Geben Sie nun anhand Ihrer Berechnungen die Dauer  &nbsp;$T_{\rm R}$&nbsp; eines Rahmens an.
+
{Now use your calculations to specify the duration &nbsp;$T_{\rm R}$&nbsp; of a frame.
 
|type="{}"}
 
|type="{}"}
 
$T_{\rm R} \ = \ $ { 123 1% } $\ \rm &micro; s$
 
$T_{\rm R} \ = \ $ { 123 1% } $\ \rm &micro; s$
  
{Wie groß ist die Anzahl der insgesamt in einem Rahmen enthaltenen Abtastwerte?
+
{What is the total number of samples contained in a frame?
 
|type="{}"}
 
|type="{}"}
$N_{\rm gesamt} \ = \ $ { 81 1% }  
+
$N_{\rm total} \ = \ $ { 81 1% }  
  
{Ermitteln Sie mit den bestimmten Parametern die Nutzträgeranzahl &nbsp;$N_{\rm Nutz}'$&nbsp; erneut.
+
{Using the determined parameters,&nbsp; determine the number of useful carriers &nbsp;$N_{\rm use}'$&nbsp; again.
 
|type="{}"}
 
|type="{}"}
$N_{\rm Nutz}' \ = \ $ { 62 1% }  
+
$N_{\rm user}' \ = \ $ { 62 1% }  
 
</quiz>
 
</quiz>
  
===Musterlösung===
+
===Solution===
 
{{ML-Kopf}}
 
{{ML-Kopf}}
'''(1)'''&nbsp; Es gilt &nbsp;$T_{\rm G}' = \tau_{\rm max}  \hspace{0.15cm}\underline { = 25\ \rm &micro; s}$.
+
'''(1)'''&nbsp; It holds that &nbsp;$T_{\rm G}' = \tau_{\rm max}  \hspace{0.15cm}\underline { = 25\ \rm &micro; s}$.
*Damit ist die untere Grenze der Ungleichung&nbsp; $T_{\rm{G}}' \ll T \ll T_{{\rm{coh}}} - T_{\rm{G}}'$&nbsp; festgelegt.
+
*This establishes the lower limit of the inequality&nbsp; $T_{\rm{G}}' \ll T \ll T_{{\rm{coh}}} - T_{\rm{G}}'$.&nbsp;
*Aber auch die obere Grenze lässt sich nun berechnen, da die Kohärenzzeit&nbsp; $T_{\rm coh} = 400\ \rm &micro; s$&nbsp; bekannt ist.
+
*However, the upper limit can now also be calculated since the coherence time&nbsp; $T_{\rm coh} = 400\ \rm &micro; s$&nbsp; is known.
  
  
  
'''(2)'''&nbsp; Zur sinnvollen Lösung der Ungleichung aus&nbsp; '''(1)'''&nbsp; wird das geometrische Mittel verwendet:
+
'''(2)'''&nbsp; The geometric mean is used to reasonably solve the inequality from&nbsp; '''(1)''':&nbsp;  
 
:$$T_{{\rm{opt}}} = \sqrt {T_{\rm{G}} ' \cdot (T_{{\rm{coh}}} - T_{\rm{G}} ')} = \sqrt {{25\,\,{\rm &micro; s}} \cdot ({400\,\,{\rm &micro; s}} - {25\,\,{\rm &micro; s}})} \hspace{0.15cm}\underline { \approx {97\,\,{\rm &micro; s}}}.$$
 
:$$T_{{\rm{opt}}} = \sqrt {T_{\rm{G}} ' \cdot (T_{{\rm{coh}}} - T_{\rm{G}} ')} = \sqrt {{25\,\,{\rm &micro; s}} \cdot ({400\,\,{\rm &micro; s}} - {25\,\,{\rm &micro; s}})} \hspace{0.15cm}\underline { \approx {97\,\,{\rm &micro; s}}}.$$
  
  
'''(3)'''&nbsp; Die benötigte Anzahl der Nutzträger ergibt sich aus folgender Gleichung:
+
'''(3)'''&nbsp; The required number of useful carriers is given by the following equation:
:$$N_{{\rm{Nutz}}} = \left\lceil {\frac{{R_{{\rm{B}}} \cdot (T + T_{\rm{G}} ')}} {{{\rm{log}_2}(M)}}}\right\rceil
+
:$$N_{{\rm{user}}} = \left\lceil {\frac{{R_{{\rm{B}}} \cdot (T + T_{\rm{G}} ')}} {{{\rm{log}_2}(M)}}}\right\rceil
 
  = \left\lceil {\frac{10^6\,\,{\rm bit/s} \cdot ({97\,\,{\rm &micro; s}} + {25\,\,{\rm &micro; s}} )}
 
  = \left\lceil {\frac{10^6\,\,{\rm bit/s} \cdot ({97\,\,{\rm &micro; s}} + {25\,\,{\rm &micro; s}} )}
 
  {{{\rm{log}_2}(4)}}}\right\rceil\hspace{0.15cm}\underline {= 61}.$$
 
  {{{\rm{log}_2}(4)}}}\right\rceil\hspace{0.15cm}\underline {= 61}.$$
  
  
'''(4)'''&nbsp; Die Stützstellenzahl der&nbsp; $\rm FFT$&nbsp; muss stets eine Zweier–Potenz sein.&nbsp; Daraus folgt:
+
'''(4)'''&nbsp; The number of interpolation points of the&nbsp; $\rm FFT$&nbsp; must always be a power of two.&nbsp; It follows that:
 
:$$ N_{{\rm{FFT}}} = 2^{\left\lceil {{\rm{log_2}} \hspace{0.05cm}(61 )} \right\rceil } = 2^6\hspace{0.15cm}\underline {= 64}.$$
 
:$$ N_{{\rm{FFT}}} = 2^{\left\lceil {{\rm{log_2}} \hspace{0.05cm}(61 )} \right\rceil } = 2^6\hspace{0.15cm}\underline {= 64}.$$
*Ungenutzte Träger können an den Rändern des Spektrums als Guard–Band verwendet werden.
+
*Unused carriers can be used as guard bands at the edges of the spectrum.
  
  
'''(5)'''&nbsp; Wir bezeichnen die gerundete Anzahl der Stützstellen des Guardintervalls mit&nbsp; $N_{\rm{G}}$.&nbsp; Dann gilt:
+
'''(5)'''&nbsp; We denote the rounded number of grid points of the guard interval by&nbsp; $N_{\rm{G}}$.&nbsp; Then holds:
 
:$$N_{\rm{G}} = \left\lceil {\frac{{T_{\rm{G}} '}} {{T_{{\rm{opt}}} }} \cdot N_{{\rm{FFT}}} } \right\rceil = \left\lceil {\frac{25\,\,{\rm &micro; s}} {97\,\,{\rm &micro; s}} \cdot 64 } \right\rceil \hspace{0.15cm}\underline {= 17},$$
 
:$$N_{\rm{G}} = \left\lceil {\frac{{T_{\rm{G}} '}} {{T_{{\rm{opt}}} }} \cdot N_{{\rm{FFT}}} } \right\rceil = \left\lceil {\frac{25\,\,{\rm &micro; s}} {97\,\,{\rm &micro; s}} \cdot 64 } \right\rceil \hspace{0.15cm}\underline {= 17},$$
 
:$$ T_{\rm{G}} = N_{\rm{G}} \cdot \frac{{T_{{\rm{opt}}} }} {{N_{{\rm{FFT}}} }}= 17 \cdot \frac{{97\,\,{\rm &micro; s}}} {64}\hspace{0.15cm}\underline { \approx {26\,\,{\rm &micro; s}}}.$$
 
:$$ T_{\rm{G}} = N_{\rm{G}} \cdot \frac{{T_{{\rm{opt}}} }} {{N_{{\rm{FFT}}} }}= 17 \cdot \frac{{97\,\,{\rm &micro; s}}} {64}\hspace{0.15cm}\underline { \approx {26\,\,{\rm &micro; s}}}.$$
  
  
'''(6)'''&nbsp; Die Rahmendauer ergibt sich zu
+
'''(6)'''&nbsp; The frame duration results to
 
:$$T_{\rm{R}} = T + T_{\rm{G}} = {97\,\,{\rm &micro; s}} + {26\,\,{\rm &micro; s}}\hspace{0.15cm}\underline {= {123\,\,{\rm &micro; s}}}.$$
 
:$$T_{\rm{R}} = T + T_{\rm{G}} = {97\,\,{\rm &micro; s}} + {26\,\,{\rm &micro; s}}\hspace{0.15cm}\underline {= {123\,\,{\rm &micro; s}}}.$$
  
  
'''(7)'''&nbsp; Mit den Ergebnissen der Teilaufgaben&nbsp; '''(4)'''&nbsp; und&nbsp; '''(5)'''&nbsp; erhält man:
+
'''(7)'''&nbsp; Using the results of subtasks&nbsp; '''(4)'''&nbsp; and&nbsp; '''(5)''',&nbsp; we obtain:
:$$ N_{\rm{gesamt}} = N_{\rm{FFT}} + N_{\rm{G}} = 64 + 17 \hspace{0.15cm}\underline {= 81}.$$
+
:$$ N_{\rm{total}} = N_{\rm{FFT}} + N_{\rm{G}} = 64 + 17 \hspace{0.15cm}\underline {= 81}.$$
  
  
'''(8)'''&nbsp; Die Neuberechnung ist nötig, da sich die Dauer des Guard–Intervalls geändert haben kann.&nbsp;  
+
'''(8)'''&nbsp; The recalculation is necessary because the duration of the guard interval may have changed.&nbsp;  
*Gegenüber der Teilaufgab&nbsp;e '''(3)'''&nbsp; wird die vorläufige Länge&nbsp; $T_{\rm{G}} '$&nbsp; durch&nbsp; $T_{\rm{G}} $&nbsp; ersetzt und man erhält ein geringfügig anderes Ergebnis:
+
*Compared to subtask&nbsp;'''(3)''',&nbsp; the provisional length&nbsp; $T_{\rm{G}} '$&nbsp; is replaced by&nbsp; $T_{\rm{G}} $&nbsp; and a slightly different result is obtained:
:$$N_{\rm Nutz}' = \left\lceil {\frac{10^6\,\,{\rm bit/s} \cdot ({97\,\,{\rm &micro; s}} + {26\,\,{\rm &micro; s}} )} {{{\rm{log_2}}(4)}}}\right\rceil = \left\lceil 61.5\right\rceil\hspace{0.15cm}\underline {= 62}.$$
+
:$$N_{\rm user}' = \left\lceil {\frac{10^6\,\,{\rm bit/s} \cdot ({97\,\,{\rm &micro; s}} + {26\,\,{\rm &micro; s}} )} {{{\rm{log_2}}(4)}}}\right\rceil = \left\lceil 61.5\right\rceil\hspace{0.15cm}\underline {= 62}.$$
*Damit ergibt sich aber weiterhin&nbsp; $N_{\rm FFT} = 64$.  
+
*However,&nbsp; this still gives&nbsp; $N_{\rm FFT} = 64$.  
  
  
Line 127: Line 128:
  
  
[[Category:Modulation Methods: Exercises|^5.7 OFDM für 4G–Netze^]]
+
[[Category:Modulation Methods: Exercises|^5.7 OFDM for 4G Networks^]]

Latest revision as of 12:13, 25 January 2022

Time-dependent attenuation curve of two mobile radio channels

In this exercise,  some OFDM parameters of a mobile radio system are to be determined.  

The following assumptions are made:

  • The coherence time of the channel is  $T_{\rm coh} = 0.4 \ \rm ms$.
  • The maximum path delay is  $τ_{\rm max} = 25 \ \rm µ s$.
  • The data rate  (bit rate)  is  $R_{\rm B} = 1 \ \rm Mbit/s$.
  • All subcarriers are  $\rm 4–QAM$ modulated.


To ensure some robustness of the system to time and frequency selective fading,  the following inequality must be satisfied:

$$T_{\rm{G}} \ll T \ll T_{{\rm{coh}}} - T_{\rm{G}}.$$

Overall,  the following procedure should be followed:

  • Preliminary determination of the guard interval duration  $(T_{\rm G}')$,
  • Determination of the optimal core symbol duration  $(T)$,
  • corresponding determination of the number of sampling points of the  $\rm FFT$.


After that,  it may be necessary to redetermine some system quantities due to the rounding performed during the calculations.

The diagram shows two exemplary attenuation curves of mobile radio systems in logarithmic representation.

  • In the case of the blue curve,  the changes over time occur relatively slowly,  while in the case of the red curve they occur four times as fast.
  • Consequently,  the blue channel has a four times larger coherence time than the red channel.




Notes:



Questions

1

Determine the minimum reasonable duration  $T_{\rm G}'$  of the  "preliminary guard interval".

$T_{\rm G}' \ = \ $

$\ \rm µ s$

2

Determine the optimal core symbol duration  $T_{\rm opt}$  as a geometric mean.

$T_{\rm opt} \ = \ $

$\ \rm µ s$

3

Determine the required number of useful carriers.

$N_{\rm user} \ = \ $

4

Specify the resulting number of interpolation points of the FFT.

$N_{\rm FFT} \ = \ $

5

Calculate the number  $N_{\rm G}$  of time samples of the guard interval and from this the new resulting guard duration  $T_{\rm G}$.

$N_{\rm G} \ = \ $

$T_{\rm G} \ = \ $

$\ \rm µ s$

6

Now use your calculations to specify the duration  $T_{\rm R}$  of a frame.

$T_{\rm R} \ = \ $

$\ \rm µ s$

7

What is the total number of samples contained in a frame?

$N_{\rm total} \ = \ $

8

Using the determined parameters,  determine the number of useful carriers  $N_{\rm use}'$  again.

$N_{\rm user}' \ = \ $


Solution

(1)  It holds that  $T_{\rm G}' = \tau_{\rm max} \hspace{0.15cm}\underline { = 25\ \rm µ s}$.

  • This establishes the lower limit of the inequality  $T_{\rm{G}}' \ll T \ll T_{{\rm{coh}}} - T_{\rm{G}}'$. 
  • However, the upper limit can now also be calculated since the coherence time  $T_{\rm coh} = 400\ \rm µ s$  is known.


(2)  The geometric mean is used to reasonably solve the inequality from  (1)

$$T_{{\rm{opt}}} = \sqrt {T_{\rm{G}} ' \cdot (T_{{\rm{coh}}} - T_{\rm{G}} ')} = \sqrt {{25\,\,{\rm µ s}} \cdot ({400\,\,{\rm µ s}} - {25\,\,{\rm µ s}})} \hspace{0.15cm}\underline { \approx {97\,\,{\rm µ s}}}.$$


(3)  The required number of useful carriers is given by the following equation:

$$N_{{\rm{user}}} = \left\lceil {\frac{{R_{{\rm{B}}} \cdot (T + T_{\rm{G}} ')}} {{{\rm{log}_2}(M)}}}\right\rceil = \left\lceil {\frac{10^6\,\,{\rm bit/s} \cdot ({97\,\,{\rm µ s}} + {25\,\,{\rm µ s}} )} {{{\rm{log}_2}(4)}}}\right\rceil\hspace{0.15cm}\underline {= 61}.$$


(4)  The number of interpolation points of the  $\rm FFT$  must always be a power of two.  It follows that:

$$ N_{{\rm{FFT}}} = 2^{\left\lceil {{\rm{log_2}} \hspace{0.05cm}(61 )} \right\rceil } = 2^6\hspace{0.15cm}\underline {= 64}.$$
  • Unused carriers can be used as guard bands at the edges of the spectrum.


(5)  We denote the rounded number of grid points of the guard interval by  $N_{\rm{G}}$.  Then holds:

$$N_{\rm{G}} = \left\lceil {\frac{{T_{\rm{G}} '}} {{T_{{\rm{opt}}} }} \cdot N_{{\rm{FFT}}} } \right\rceil = \left\lceil {\frac{25\,\,{\rm µ s}} {97\,\,{\rm µ s}} \cdot 64 } \right\rceil \hspace{0.15cm}\underline {= 17},$$
$$ T_{\rm{G}} = N_{\rm{G}} \cdot \frac{{T_{{\rm{opt}}} }} {{N_{{\rm{FFT}}} }}= 17 \cdot \frac{{97\,\,{\rm µ s}}} {64}\hspace{0.15cm}\underline { \approx {26\,\,{\rm µ s}}}.$$


(6)  The frame duration results to

$$T_{\rm{R}} = T + T_{\rm{G}} = {97\,\,{\rm µ s}} + {26\,\,{\rm µ s}}\hspace{0.15cm}\underline {= {123\,\,{\rm µ s}}}.$$


(7)  Using the results of subtasks  (4)  and  (5),  we obtain:

$$ N_{\rm{total}} = N_{\rm{FFT}} + N_{\rm{G}} = 64 + 17 \hspace{0.15cm}\underline {= 81}.$$


(8)  The recalculation is necessary because the duration of the guard interval may have changed. 

  • Compared to subtask (3),  the provisional length  $T_{\rm{G}} '$  is replaced by  $T_{\rm{G}} $  and a slightly different result is obtained:
$$N_{\rm user}' = \left\lceil {\frac{10^6\,\,{\rm bit/s} \cdot ({97\,\,{\rm µ s}} + {26\,\,{\rm µ s}} )} {{{\rm{log_2}}(4)}}}\right\rceil = \left\lceil 61.5\right\rceil\hspace{0.15cm}\underline {= 62}.$$
  • However,  this still gives  $N_{\rm FFT} = 64$.