2.9 Problems 21 22 23 24 25 Prove that for a hexagonal geometry, the co-channel reuse ratio is given by Q = J3N , where N = i” +ij+ j’ . (Hint: use the cosine law and the hexago- nal cell geometry) Show that the frequency reuse factor for a cellular system is given by k/S, where k is the average number of channels per cell and S is the total number of channels available to the cellular service provider. A cellular service provider decides to use a digital TDMA scheme which can tolerate a signal-to-interference ratio of 15 dB in the worst cage, Find the opti- mal value of N for (a) omni-directional antennas, (bj 120° sectoring, and (c) 60° sectoring. Should sectoring be used? If so, which case (60° or 120°) should be used? (Assume a path loss exponent of 2 = 4 and consider trunking effi- ciency). If an intensive propagation measurement campaign showed that the mobile radio channel provided a propagation path loss exponent of n = 3 instead of 4, how would your design decisions in Problem 2.3 change? What is the optimal value of N for the case of n = 3? For a N = 7 system with a Pr[Blocking| = | % and average call length of 2 minutes, find the traffic capacity loss due to trunking for 57 channeis when going from omni-directional antennas to 60° sectored antennas. (Assume that blocked calls are cleared and the average per user call rate is A = | per hour.Defining D,D, RR for « macros architecture with N = 7. The smalierheaagans form zones and {ee anon otieed bl ogee frm cl Si eae costal oleae show. 26 Ausare that a cll named “Radio Knob ht 57 chance each with an effective dated power of 82 watia and a cll radia of 10 ke The grade of wre tstabished tobe a protublity of Uocking of 8% (aturing backed salle are Sleared), Assume the average call length i 2 minutes and each wr averegee 2 calls por hour Further, abume the ell hes ust reached te maximem eae igy and mnt be split inte 4 new microceis a provie 4 times the expacty 19 {he some arc) What i the current capacity ofthe “Radio Knob” cell? (b ‘What isthe radius and transmit power of Une new cul?! How many chan: nels are ded in enc ofthe new cls te maintain fequency ease stabity Inthe syeer? (traffic uniforly istibuied, what edhe new tafe car Fed by each new cell? Will the probaly of blocking in these new ells be below 0.4 after the opi? 27, Breese in trusking (queueing) theory (a) What i the maimum system eapaciy (ota and per channel in Erlang vehen providing a 2% blocking probability with «channel, with 20 char ‘ela with 40 channel? (le) Using the trafic intensity per channel calelated in par a), find the (grade of service noe eal delayed sytem for the ate of dls being [ester than 20 seconds. Assume tha H= 1055 and determine the GOS = ‘Oh 2 The Cour Cont — Sten Dean Frau {or channels, for 20 channel, for 40 channels, (a) Comparing pars (a) and pare does alee eall delayed system with #20 second queve perform better than aaystem thet cleus blocked al? 28 A receiver in an urban cellolar radio sysors detects a 1 mW signal at {f =idy = meter rm the treamiver In order to ign cechanel inter Terence effects, i it required that the gal received a any base station ‘wots from Cnither bom Sato toemmriier which epertes th the ano ‘nae must be beirw 100 mA measurement team has deermied that ‘the average path Ise exponent inthe systom tn = B. Determine the major ‘adi af each cll if T-cell reve pattern is sed. Whats Ue major radios fa ‘tool reuse pattern i uted? 20 Acalllareystem using » cluster sizeof 7 is described ip Problem 2.8 It ‘porated with 650 channels, 30 0f which are designated x typ (conta) ‘channels otha there are about 90 vice channls avilable per cll Ifthe i ‘pore! user denny 69000 user in the eyetem, and each ner makes tan swerage af ne call per bour and each call lata 1 mine during peak hors, determine the probaly thats wor wil experience delay greater than 20 Second fall alls are quod, 2:10 Show that fn = 4 cll an be splints our smaller cella each with alf he ‘radius and U6 ofthe transmiter power ofthe original cel. If extensive mew: ‘urements show thst the path lass exponents 3, how should the traneraiter ‘ower be changed inorder to split all ito fur smaller sla? What impact ‘rill his have on the celular geometry? Explain yeur aaswer and prorde Sranings hat shoe how the new ele wold ft within the arginal macrcels Forimpicty se ome! deco! antonnaa 2.11 Using the frequency sasgnment chart in Table 2.2, design a channelization scheme fers Bede carrier that aes cll ease and 3 weer pr cal och exUhecomanh Gebdaia berth wLaeuaa cmaceane2.12 Repeat Prublem 2.11 for the case of 4-cell rouse and 6 sectors per cell 2.13 In practical cellular radio systems, the MSC is programmed to allocate radio channels differently for the closest co-channel cells. This technique, called a hunting sequence, ensures that co-channel cells first use different channel from within the co-channel set, before the same channels are assigned to calls in nearby cells. This minimizes co-channel interference when the cellular sys- ‘tem is not fully londed. Consider 3 adjoining clusters, and design an algorithm that may be used by the MSC to hunt for appropriate channels when requested from co-channel cells. Assume a 7-cell reuse pattern with 3 sectors per cell, and use the US. cellular channel allocation scheme for the A-side carrier. 2.14 Determine the noise floor (in dBm) for mabile receivers which implement the following standards; (a) AMPS, (b) GSM, (c) USDC, (4) DECT, (e) 18-98, and if) (CT2. Assume all receivers have ¢ noise figure of 10 4B. 2.18 If base station provides « signal level of 90 dBm at the cell fringe, find the SNR for each of the mobile receivers deseribed in Problem 2.14. 2.16 From first principles, derive the expression for Erlang B given in this chapter 2.17 Carefully enalyze the trade-off between sectoring and trunking efficiency for a 4-cell cluster size. While sectoring improves capacity by improving SNR, there ia loss due to decreased trunking efficieney, since each sector must be trunked separately. Consider a wide range of total available channels per cell and con: sider the impact of using 3 sectors and 6 sectors per cell. Your analysis may involve computer simulation, and should indieate the “break even* point when wovsoftsnet ldeoad ~ Ch, 2+ The Cetus Concopt — System Design Fundamentals sectoring is not practical.ac re ee Tea tee er ge te ee aE re Pe ee calls per hour, each call lasting an average of 5 minutes. (a) What is the traffic intensity for each user? (b) Find the number of users that could use the eystem with 1% blocking if only one channel is available. (e) Find the number of users that could use the system with 1% blocking if five trunked channels are available (d) If the number of users you found in (c) is suddenly doubled, what is the new blocking probability of the five channel trunked mobile radio system? Would this be acceptable performance? Justify why or why not. 2.19 The US. AMPS system is allocated 50 MHz of spectrum in the 800 MHz range, and provides 832 channels. Forty-two of those channels ere control channels. The forward channel frequency is exactly 45 MHz greater than the reverse channel frequency. (a) Is the AMPS system simplex, half-duplex, or duplex? What is the bandwidth for each channel and how is it distributed between the base station and the eubscriber? (b) Asoume a base station transmits control information on channel 352, operating at 880.560 MHz. What is the transmission frequency of a sub- scriber unit transmitting on channel 352? (©) The A-side and B-side cellular carriers evenly split the AMPS channels. Find the number of voice channels and number of control channels for each carrier. (d) Let's suppose you are chief engineer of a cellular carrier using 7-cell reuse. Propose a channel assignment strategy for a uniform distribution of users throughout your cellular system. Specifically, assume that each cell has 3 control channels (120° sectoring is employed) and specify the number of voice channels you would assign to each control channel in your system. {e) For an ideal hexagonal cellular layout which has identical cell sites, what in the distance between the centers of two nearest co-channel cells for 7- cell reuse? for 4-cell reuse?2.20 Pretend your company won a license to build a U.S. cellular system (the appli cation cost for the license was only $600!) Your license is to cover 140 square km, Assume a base station costs $500,000 and a MTSO costs $1,500,000. An ‘extra $500,000 is needed to advertise and start the business. You have con- vineed the bank to loan you 86 million, with the idea that in four years you will have earnad $10 million in gross billing revenues, and will have paid off the loan. (@) How many base stations (ie, cell sites) will you be able to install for 86 million? (b) Assuming the earth is flat and subscribers are uniformly distributed on the ground, what assumption can you make about the coverage area of each of your cell sites? What is the major radius of each of your celis, ‘assuming # hexagonal mosai (©) Assume that the average customer will pay $50 per month over & 4 year period. Assume that on the first day you turn your system on, you have a certain number of customers which remains fixed throughout the year. On ‘the first day of each new year the number of customers using your system pm. vioftenet deroad 7 Ch. 2+ The Gottular Concept — System Design Fundamentals doubles and then remains fixed for the rest of that year. What is the mini- mum number of customers you must have on the first day of service in order to have earned $10 million in gross billing revenues by the end of the 4th year of operation? (2) For your answer in (c), how many users per square km are needed on the first day of service in order to reach the $10 million mark after the 4th year?(Ch. 10 + Wireless Syatoms and Standards 38t 10.13 Problems 1oL 102 103 104 107 Los. 109 10.10 10. 10.12 10.13 1014 10.18 10.16 Which of the following is NOT true of GSM? Check all that apply (a)The uplink and downlink channels are separated by #5 MHz. (6) There are eight half-rate users in one timeslot. (c) The peak frequency deviation of the GSM modulator is an integer malti- pleof the GSM data rate. (4) GSM uses a constant envelope modulation. List all the cellular systems that do not support Mobile Assisted Handoff ‘What is the cut-off frequency of the baseband, Gaussian, pulse-shaping filter used in the GSM system? The FCC allocated an additional 10 MHz spectrum for cellular services in 1989. How many additional channels can be accommodated in this band- width for each ofthe cellular standards? The IS-95 system uses a rate 1/2 convolutional eneading in the forward chan- nel and « rate 1/8 convolutional coding in the reverse channel. What were the reasons for doing so? ‘Which of the following is NOT true of the IS-95 system? (a) No hard limits on capacity (b) Soft handof possible {) Uses slow frequency hopping (d) The number of channels that can be accommodated in the forward and reverse links are different How are different channels within the forward link in an 18-95 system iden- tified? ‘What is the spectral efficiency of the GMSK modulation scheme used in Gsm? Which of the fallowing systems is solely based on Frequency Division Multi- ple Access (FDMA(? (@)DECT \b)CT2 USDC (aS Which cellular system uses the most bandwidth efficient modulation? ‘Which cellular systems do not use # constant envelope modulation ssheme? What were the reasons for choosing x4 D@PSK modulation scheme for USDC against DQPSK? How many full-rate physical channels per cell can s GSM system accommo- dave? Which of the following is @ blank-and-burst channel in the AMPS system? (a) Paging Channel (PC) () Reverse Voice Channel (RVC) (¢) Slow Associated Control Channel (SACCH) (d) Fast Associated Control Channel (FACCH) ‘Compute the modulation index of the FM modulator used in the AMPS eys- tem. Digital cellular syetems use convolutional cading for error protection as ‘opposed to block coding. The primary reason for doing so is (a) Convolutional coding performs better under Rayleigh fading conditions (b) Convolutional coders are more robust to burst errors (c) Convolutional coding is leas complex when compared to block coding (d) Real time soft-decision decoding algorithms which improve performance are available for decoding convolutional codes32 1017 1018 10.19 1020 1021 1022 1023 2024 1025 1026 1027 1028 10.29 1030 (Gh 10+ Wieless Systems and Standaras Which ofthe following systems is based on a microcell architecture? (iGSM(sDECT. (USDC (ai 15.95 What is the maximum data rate at which s user can transmit data on a DECT system? How many ful AMPS chanrel?” ‘What isthe maximum fade slope iin dBis) which can be compensated for by the reverse power control subchannel in the 18-95 CDMIA system? In the normal GSM time slot of Figure 10, why would the 26 equalizer training bits be placed in the middle of the frame rather than atthe begin ning? What isthe reason for having an 8:25 bit guard period after the data bursts? Compare the number of omni-directional cells required to cover a 1000 sa km area using GSM at 800 MHz and DCS-1900 at 1900 MHz. Assume the sensitivity of both the GSM and the DCS-1800 receivers equal -104 dim Assume equ transmitter power and antenna gains for the two systems Compare the number of omni-directional cells required to cxver 1000 59 sem ates using DCS-1900 and 18-95, Assume equa’ transmitter powers and antenna gains How would the cell coverage radius be affected by the system loading for each of the following technologies: AMPS, IS-95 CDMA, and IS-54 TDMA? What mechanisms would cause breakdown in the reverse link ofan 15-95 CDMA systems asthe nomber of users ina sector approaches the theoretical limit? For what reasons would the PACS version forthe unlicensed PCS band use "TDD instead of FD? ‘What is likely to happen to a DECT system if it is deplayed outdoors in an environment where significant multipath could occ? Explain your answer, aod provide a qualitative analysis Draw the allocation of bits im @ USDC half-rate time alt. Then answer the folowing (a) What isthe channet data rate for the USDC air interface? (b) How many user bits are there in each USDC time sion? (c) What is the time duration foreach USDC frame? (@) Using the definition of frame effcieney in the text determine the frame Mfiency for USDC. Consider an AMPS or ETACS system. (a) List the possible SAT tones in AMPS. (b} Why would the ST tone be useful toa service provider? (6) List atleast five ways a cellular phone call may'be terminated. How could ‘a caricr distinguish between these? For a fal-rate USDC syetem, consider how the data is allocated between channels. (a) What is the gross RF data rate? (b) What is the gross RF data rite forthe SACCH? ()What is the gross RF data ca'e for the CDDVC? (a) What is the gross data rate for synchronization, remp-up, and guard time? {e) Verify that your answers in (b)4a) add up to your anawer in (a) (8 What is the end-user data rate provided in fullrate USDC? IDC system accommodate in a single te users can a UCh. 10 « Wireless Systems and Standards 553 10.31 Prove that the GSM system allocates gross RF data rate of 33.854 kbps/user. Show this by summing up the individual user data rates for (a) the speech coder. {b) speech error protection. (c) SACCH. (d) guard time, ramp-up, and synchronization. 10.32 Compute the longest ‘ime over which a mobile station would have to wait in order to determine the frame number being transmitted by a GSM base sta- tion.Questions ‘What seomotre shape is use in gear Systm design? ‘What is the principle of frequency reuse inthe context of acellular network? List five ways ofimereasing the capacity of acellular system Explain the paging fonction ofa cellar sym, ‘Whats fading? ‘What is the difference between diffraction ad seatertng? ‘What isthe difference between last and slow ti ‘What isthe ditteence between tat and yelectve fading? ‘What are the key dillsrences betwec fist and secon-gencration cellular systems? ‘What are the advantages of using CDMA fo a cellular network? ‘Wha are the disidvantages of using CDMA for a cali ar retwork? What are some key characteristics that distinguish third-generation cellular systems from second: goveration cellular systems? REBERERERERE 141 Consider (yur eltferentcellulit systems that share the following characteristics, The ey buts are 25 to M49 MIF for motile ait eiasminion and $70 to 39) Mit for base station transmission. A duplex Siruit consist one 30-KHz channel in tach direction. The systems are distinguished by the reuse facor, which is, 7.12, ad 1S-respectively | Suppose that in each of the systems the eluster of cells (4,7, 12,19) ts duplicated {Wtmes. Fie the oumber of slaultanevun comanun ation that can be supported by each system, |). Find the auniber of simaltaneoasscomtiunications that can be supported by asi le cell in each system, ‘¢. Whats the area coveted in cells by euch system? Suppose the cll se isthe same in al foxr systems nda fixed aren of 100 cals is covered by cach systers Find the ber f stu Laneous cemesunics Hom that tem be supperted by each stern 142 Deserie w equence of eeuts ilar to that of Figute U6 for {4 cal roma mobile unit oa fied subsesber 1b, cll from a fixed subsenibat to a mobile unit 143 Av analog cellar sistem has a total of 33 MEL of bandwidth and uses two 25-K1 ‘imple (one say) channels to prowide fill dpiex voice ac control chan. “A. What the sumer of channels available pr cll fora reuency reuse faetor of (1) 4 cells (2) 7 ells and (3) 12 cells? ‘b. Assume that | MH is dedicated to contol channels but that only one contol ‘channe! imesded per ell, Determinc a resonable distebtvs of control chanel. ‘and Voice channels in each cll Cor the thce frequency reuse factors ofpat (4). AAA celular system uses FDMA with spectrum allocation of 12.5 Mz in each dee tion, a guard bod at the edge of the allocated spectrum of 10 KHz, and a channel bandwidth of SOKH2 What isthe number of available channels? {AS rss Ar fg att = EE B= channel bandwidth A, = total bandwidth in one direction i eR er ere