Introduction to 3rd Generation Cellular Networks

Wednesday, November 20, 2002

Prelim

姓名: 學號:

  1. Close book.
  2. You have 150 minutes.
  3. Unjustified answers earn no points.
  4. Good luck.

Problem 1: Explanation [38].

  1. [2] Is it possible to increase the system capacity (number of users per unit area) by using sectoring scheme only in GSM system? If it is possible, explain how. If it is not, explain why.
  2. [1] Is 62 a valid reuse factor? Justify your answer.
  3. [2] Describe the main purpose of open loop power control in WCDMA FDD mode. Why open loop power control is far too inaccurate?
  4. [2] Describe the main purpose of outer loop power control in WCDMA. Explain the reasons why outer loop power control is implemented in RNC.
  5. [3] Describe the principles of relative threshold handover scheme in WCDMA by drawing a diagram using Window_Add, Window_Drop, Monitored set, and Active set.
  6. [1] Describe the main difference between soft-handover and softer-handover.
  7. [1] Describe the main aspects of a half-duplex communications system.
  8. [2] Describe the main differences between FDD and TDD. Which on is more sensitive to timing synchronization?
  9. [2] What is the meaning of the blocking probability? What is the meaning of the dropping probability?
  10. [2] Describe at least three major differences between high tier and low tier cellular systems.
  11. [2] In modeling the mobile radio propagation, what are the three major factors that describe loss/fading effects?
  12. [2] For a received signal , what is its RMS (root mean square) delay spread?
  13. [1] What are the two major factors that lead to small scale fading?
  14. [2] How do we differentiate between a wideband system and a narrowband system?
  15. [2] What are the two types of fading that can be determined based on Doppler spread? How do we differentiate between them?
  16. [1] In applying Rayleigh fading model, which kind of fading channel do we assume?
  17. [1] Describe the characteristics of a Ricean fading channel.
  18. [2] In computer simulation, how do we generate a frequency selective fading using Rayleigh fading? (You may draw a diagram to explain your idea.)
  19. [2] Describe at least four diversity schemes that can be adopted for small scale fading counteraction.
  20. [1] Describe the main differences between open-loop transmit diversity and closed-loop transmit diversity in WCDMA.
  21. [1] Describe the cause of co-channel interference.
  22. [1] Describe the cause of adjacent channel interference.
  23. [1] Define . Explain the effects when Δ is too large and too small, respectively.
  24. [1] In practical handover, we suffer from problem of simultaneous traffic of high speed and low speed mobiles. Describe one of the possible solutions.

Problem 2:[10] Signal Constellation

An M-ary signalingscheme uses the following signals:

Draw the signaling constellation and determine the optimum decision regions.

Problem 3. [6] Small Scale Fading

Consider two propagation environments: (A) Outdoor with delay spread =20 us and (B) Indoor with delay spread =5 ns

(a)[2]Find the coherence BW of channel A and channel B with above 50% correlation (), respectively.

(b)[2]Consider a wireless cellular system with bandwidth of 200 kHz. Determine if Channel A is frequencyselective fading or flat fading? Are equalizers necessary or not?

(c)[2]Repeat question (b) for the case of Channel B.

Problem 4.[12] Linear Block Codes

The extended (n+1,k) Hamming code is obtained form the original (n,k) Hamming code by appending an overall parity bit. As a result, all the code-words have even number of “1” bits. Let H denote the parity-check matrix of the original Hamming code. Then, the parity-check matrix of the extended Hamming matrix is defined by:

Consider the (7,4) Hamming code, whose parity-check matrix is given by:

(a)[6]What are the minimum distances of the Hamming code and extended Hamming code, respectively?

(b)[3]How many erroneous bits that can be corrected by the Hamming code and extended Hamming code, respectively?

(c)[3]How many erroneous bits that can be detected by the Hamming code and extended Hamming code, respectively?

Problem 5.[8] Rake Receiver

Assume a 3-path multipath channel model as shown below.

(a) [5] Design a maximum ratio combining (MRC) RAKE receiver, where c(t) is the PN waveform and t0=0. What are the values for t1, t2, a0, a1 and a2.

------/ Path delay [ns] / Amplitude / Phase
0 / 0 / 1 / /3
1 / 260 / 0.5 / /4
2 / 521 / 0.2 / /6

(b) [3] Describe the reasons why a RAKE receiver, instead of an equalizer, is required in a WCDMA system.

Problem 6. [5] Co-channel Cells

For a reuse factor of N=19, draw all the co-channel cells in the following figure.

Problem 7.[6] Huffman Coding

A discrete memoryless source has an alphabet of five symbols with their probabilities for its output as given below:

Symbol / S0 / S1 / S2 / S3 / S4
Probability / 0.55 / 0.15 / 0.15 / 0.10 / 0.05

Among several different Huffman codes of this source, construct the one with the smallest code-word length variance.

Problem 8.[15] Convolutional Code

Consider a rate 1/3 convolutional encoder as shown in the following figure:

(a)[4]Complete the following table:

PresentState / Input Bit / NextState / Output Bits
00 / 0
1
01 / 0
1
10 / 0
1
11 / 0
1

(b)[3]If the input message is [0 1 1 1] (0 is the first input bit) with two tail bits of 0, determine the output bits.

(c)[8]If the received bits (after hard decision) are [000 111 001 001 010 011] (leftmost bit corresponds to first-received bit), apply the Viterbi decoding algorithm to determineall the possible outputs of the decoder.