Scribd
Upload
296 views

EE497E - Lab 3:prelab: Student Name

The prelab document provides background information and questions to help prepare for Lab 3 on symbol timing recovery. It discusses two methods for finding the peak of the matched filter output: direct maximization of output energy and an indirect maximization using an early-late gate algorithm. It asks the student to generate LabVIEW code implementing these two algorithms and to integrate them into an existing symbol timing recovery simulator VI to observe their performance in the presence of various channel impairments.

Uploaded by

cpayne10409
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
296 views

EE497E - Lab 3:prelab: Student Name

The prelab document provides background information and questions to help prepare for Lab 3 on symbol timing recovery. It discusses two methods for finding the peak of the matched filter output: direct maximization of output energy and an indirect maximization using an early-late gate algorithm. It asks the student to generate LabVIEW code implementing these two algorithms and to integrate them into an existing symbol timing recovery simulator VI to observe their performance in the presence of various channel impairments.

Uploaded by

cpayne10409
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 2

EE497E - Lab 3:Prelab

Student Name:

Reminder: On your LabVIEW programs, pay attention to use: Descriptive variable names Insight ul comments using te!t labels Easy to read" code style #i$e$ easy to ollo% data lo%& 'hese parameters %ill be evaluated as part o your lab report$ Additional Information: As the lab manual is not sufficient, I hope you find this information useful.

/30

'he undamental goal o symbol synchroni(ation is to sample the pulse at its pea) value$ *s you might recall, in the previous lab, %e did pulse shaping and matched iltering$ In ideal conditions, the output o the matched ilter is a triangle and %e need to ind the +pea), o that triangle in this lab$ 'hen, %e shi t our input array appropriately so that, a ter do%nsampling, %e do not lose that pea) #I you loo) at bura)-symbol-timing$vi, you can see that there is a Do%nsample bloc) a ter the timing recovery model case& $ *s you might thin) o , there are many %ays o inding a +pea), in a stream$ In this lab, %e are going to implement t%o di erent %ays or this purpose$ Direct .a!imi(ation and Indirect ma!imi(ation o the output energy$ In Direct Ma imi!ation #/$0, page 12&, %e are basically simulating that %e o set by )34,/,5,06 values, %e decimate and loo) at the total energy$ We loo) or the ma!imum energy and %e use that o set value to construct the chopped array$ 'his approach ensures that, %hen %e do%nsample at the ne!t step #%hich is already implemented in bura)-symbol-timing$vi&, %e )eep the sample %hich corresponds to the matched ilter pea)$ In Indirect Ma imi!ation a"#"a Earl$ Late %ate Al&orit'm #/$2, page 12&, %e ollo% an alternative approach and %e DO 7O' go directly #)3/,5,06& to loo) or the ma!imum$ 'he earlylate gate synchroni(ation techni8ue e!ploits the symmetry o the matched- iltered signal$ 'he matched- iltered signal output typically has a symmetric shape #neglecting distortion and noise&$ 'he optimal timing is let us say %hen the matched iltered output R#t& is sampled at t3'$ I %e ta)e t%o measurements R#'o9 :& and R#'o- :& and i 'o 3 ', the t%o measurements are e8ual and the optimal sample %ill be located hal %ay bet%een the t%o samples$ 'his can be illustrated as in ;igure /$

What %e are doing is actually loo)ing at the irst derivative #slope o the tangent& at di erent locations #'his is actually given as appro!imation in the irst e8uation o page 11&$ When the slope is (ero, %e )no% that it is either a ma! #or min<<<&$ 'he symmetry o matched iltering help at this point$& We can summari(e the implementation o this part as ollo%s$ ;irst %e have to generate early, late and ontime gates # as in E8uation #2&, page 11&$ 'hen %e subtract, late gate rom early gate and multiply by the ontime gate$ Remember that the lab manual suggest you to use / as the : value$ 'his e8uation is actually trying to tell you that, you are minimi(ing =late gate part energy> ? =early gate part energy>$ @o start %ith a good guess #initial o set value, this is an indirect approach&$ 'he initial o set value should be %ithin the irst symbol # or accuracy&$ 'hen i the e8uation output #energy di erence& rom irst guess is positive, you shi t the o set to the right$ I the energy output rom the irst guess is negative, you shi t to the le t$ Aou )eep doing this until you ind %hen they are e8ual #ideally&$ 'he real %orld is not ideal so I %ould suggest you to ind a better %ay to get you out o this %hile loop$ Pre Lab 3 /$ () P*S+ @ho% that in the absence o noise, and in E8$ #5& #page 15& do not have any impact on the ma!imum output energy solution$ 5$ () P*S+ What are the t%o critical assumptions used to ormulate the indirect ma!imi(ation o the output energy< Bonsider ho% the presence o the lat ading channel *WC7 can impact this method$ @peci ically, using at least one o the critical assumptions, e!plain ho% you might mitigate the impact o these impairments by suitable selection o parameters$ 0$ () P*S+ * ter do%nsampling a se8uence originally sampled at rate / /Tz by a actor M, %hat is the sample period o the resulting signal< 2$ (,0 P*S+ Cenerate LabVIEW code to estimate the direct ma!imi(ation o the output energy in discrete time and the early?late gate algorithm by using 'able / #yourname-align-.a!Energy$vi& and 'able 5 #yourname-align-ELgate$vi&$ -SE *.E D/0P102 on AN%EL *0 /E*-/N 30-/ IMPLEMEN*A*I0NS" 1$ () P*S+ * ter building align_ !gate."i and align_Ma# nergy."i, place theseVIs into your simulator #i$e$, inside symbol timing."i&$ .odi y the symbol timing recovery method and the delay introduced by the channel rom the ront panel o $imulator."i$ * ter veri ying the unctionality o your code, observe ho% these algorithms per orm in the presence o a narro%band channel %ith *WC7$ a& In the presence o these impairments, observe ho% the pac)et length a ects the per ormance o the algorithms you have implemented$ b& Dsing delay %d 3 4.02Ts and the parameters listed above, plot the timing error statistic &'(, as de ined in E8$ #1&, or oversample actors ' 3 5, 2, E, . . . , 54 #i$e$, even integers up to 54&$ Dse a logarithmic scale or timing error and a linear scale or '$

You might also like