1G PCS

1
(No Transcript)
2
Overview of Wireless NetworksIntroduction
3
Overview of Wireless NetworksExisting Network
Infrastructure

• Public Switched Telephone Network (PSTN) Voice
• Internet Data
• Hybrid Fiber Coax (HFC) Cable TV

4
Overview of Wireless NetworksMarket Sectors for
Applications

• Four segments divided into two classes
  voice-oriented and data-oriented, further divided
  into local and wide-area markets
• Voice
• Local low-power, low-mobility devices with
  higher QoS cordless phones, Personal
  Communication Services (PCS)
• Wide area high-power, comprehensive coverage,
  low QoS - cellular mobile telephone service
• Data
• Broadband Local and ad hoc WLANs and WPANs
  (WPAN-Wireless Personal Area Network)
• Wide area Internet access for mobile users

5
Overview of Wireless NetworksEvolution of
Voice-Oriented Services
Year Event
Early 1970s First generation of mobile radio at Bell Labs
Late 1970s First generation of cordless phones
1982 First generation Nordic analog NMT
1983 Deployment of US AMPS
1988 Initiation of GSM development (new digital TDMA)
1991 Deployment of GSM
1993 Initiation of IS-95 standard for CDMA
1995 PCS band auction by FCC
1998 3G standardization started
FDMA Frequency Division Multiple AccessNMT
Nordic Mobile TelephonyAMPS Advanced Mobile
Phone SystemGSM Global System for Mobile
CommunicationTDMA Time Division Multiple Access
IS-95 Interim Standard 95CDMA Code Division
Multiple AccessPCS Personal Communication
SystemFCC Federal Communication Commission
6
Overview of Wireless NetworksEvolution of
Data-Oriented Services
Year Event
1979 Diffused infrared (IBM Rueschlikon Lab - Switzerland
Early 1980s Wireless modem (Data Radio)
1990 IEEE 802.11 for Wireless LANs standards
1990 Announcement of Wireless LAN products
1992 HIPERLAN in Europe
1993 CDPD (IBM and 9 operating companies)
1996 Wireless ATM Forum started
1997 U-NII bands released, IEEE 802.11 completed, GPRS started
1998 IEEE 802.11b and Bluetooth announcement
1999 IEEE 802.11a/HIPERLAN-2 started
HIPERLAN High Performance Radio LANCDPD
Cellular Digital Packet DataU-NII Unlicensed
National Information InfrastructureGPRS
General Packet Radio Service
7
Overview of Wireless Networks Different
Generations

• 1G Wireless Systems Analog systems
• Use two separate frequency bands for forward
  (base station to mobile) and reverse (mobile to
  base station) links Frequency Division Duplex
  (FDD)
• AMPS United States (also Australia, southeast
  Asia, Africa)
• TACS EU (later, bands were allocated to GSM)
• NMT-900 EU (also in Africa and southeast Asia)
• Typical allocated overall band was 25 MHz in each
  direction dominant spectra of operation was 800
  and 900 MHz bands.

AMPS Advanced Mobile Phone SystemTACS Total
Access Communication SystemNMT Nordic Mobile
Telephony
8
Overview of Wireless Networks Different
Generations

• 2G Wireless Systems Four sectors
• Digital cellular
• GSM (EU/Asia) TDMA
• IS-54 (US) TDMA
• IS-95 (US/Asia) CDMA
• PCS residential applications
• CT-2 (EU,CA) TDMA/TDD
• DECT(EU)TDMA/TDD
• PACS (US) TDMA/FDD

CT-2 Cordless Telephone 2DECT Digital
Enhanced Cordless TelephonePACS Personal
Access Communication System
9
Overview of Wireless Networks Different
Generations

• 2G Wireless Systems Four sectors (contd)
• Mobile data
• CDPD shares AMPS bands and site infrastructure
• GPRS shares GSMs radio system - data rates
  suitable for Internet
• WLAN Unlicensed bands, free of charge and
  rigorous regulations very attractive!
• IEEE 802.11 and IEEE 802.11b use DSSS physical
  layer
• HIPERLAN/1 uses GMSK
• IEEE 802.11a and HIPERLAN/2 use OFDM next
  generation

CDPD Cellular Digital Packet DataGPRS
General Packet Radio Service DSSS Direct
Sequence Spread SpectrumGMSK Gaussian Minimum
Shift KeyingOFDM Orthogonal Frequency Division
Multiplexing
10
Overview of Wireless Networks Different
Generations

• 3G and Beyond
• Purpose develop an international standard that
  combines and gradually replaces 2G digital
  cellular, PCS, and mobile data services, at the
  same time increasing the quality of voice,
  capacity of the network, and data rate of the
  mobile data services.
• Radio transmission technology of choice W-CDMA
• 3G was envisioned to provide multimedia services
  to users everywhere

11
Overview of Wireless Networks Different
Generations

• 3G and Beyond
• WLANs provide broadband services in hot spots
• WPANs connect personal devices together laptop,
  cellular phone, headset,speakers, printers
• WLAN and WPAN are the future of broadband and ad
  hoc wireless networks
• WPANs first standard bluetooth lower rates
  than WLAN but uses a voice-oriented wireless
  access method for integration of voice and data
  services

12
Overview of Wireless Networks Different
Generations
Relative coverage, mobility, and data rates of
generations of cellular systemsand local
broadband and ad hoc networks.
13
Operation of Wireless Networks

• Getting familiar with terms
• MS/MT Mobile Station/Mobile Terminal
• BS Base Station
• MSC Mobile Switching Center
• HLR Home Location Register (database)
• VLR Visitor Location Register (database)
• Cellular Network Architecture
• Mobility Management

14
Cellular Network Architecture
LocationRegister (Database)
Radio Network
Mobile Switching Center
Base Station Controller
MSC
Backbone Wireline Network
Mobile Terminal
Base Station
Cell
15
BASIC ARCHITECTURE
Home Location Register
(HLR)
BACKBONE TELEPHONE NETWORK
Visitor Location Register
(VLR)
Mobile Switching Center
MSC
(MSC)
VLR
Mobile Terminal (MT)
Local Signaling
Long Distance Signaling
16
Cellular Concept

• A CELL is the radio coverage area by a Base
  Station (BS).
• The most important factor is the SIZE and the
  SHAPE of a CELL.
• Ideally, the area covered a by a cell could be
  represented by a circular cell with a radius R
  from the center of a BS.
• Many factors may cause reflections and
  refractions of the signals, e.g., elevation of
  the terrain, presence of a hill or a valley or a
  tall building and presence in the surrounding
  area.
• The actual shape of the cell is determined by the
  received signal strength.
• Thus, the coverage area may be a little
  distorted.
• We need an appropriate model of a cell for the
  analysis and evaluation.
• Many posible models HEXAGON, SQUARE, EQUILATERAL
  TRIANGLE.

17
Cell Shape
R
R
R
Cell
R
R
(c) Different Cell Models
(a) Ideal Cell
(b) Actual Cell
18
Size and Capacity of a Cell per Unit of Area and
Impact of the Cell Shape on System
Characteristics
19
Cellular Concept - Example

• Consider a high-power transmitter that can
  support 35 voice channels over an area of 100 km2
  with the available spectrum
• If 7 lower power transmitters are used so that
  they support 30 of the channels over an area of
  14.3 km2 each.
• Then a total 730 35 80 channels are
  available instead of 35.

2
3
1
7
4
6
5
20
Cellular Concept

• If two cells are far away from enough that the
  same set of frequencies can be used in both
  cells, it is called frequency reuse.
• With frequency reuse, a large area can be divided
  into small areas, each uses a subset of
  frequencies and covers a small area.
• With frequency reuse, the system capacity can be
  expanded without employing high power
  transmitters.

21
Capacity Expansion by Frequency Reuse

• Same frequency band or channel used in a cell can
  be REUSED in another cell as long as the cells
  are far apart and the signal strength do not
  interfere with each other.
• This enhances the available bandwidth of each
  cell.
• A group of cells that use a different set of
  frequencies in each cell is called a cell
  cluster.

22
NUMBER OF CELLS IN A CLUSTER
23
CELL CLUSTER
24
FREQUENCY REUSEExample A typical cluster of 7
such cells and 4 such clusters with no
overlapping area
------? D
----------?
FREQUENCY REUSE DISTANCE D
25
RULE to Determine the Nearest Co-Channel
NeighborsDetermining the Cluster Size
j

• To find nearest co-channel neighbors of a
  particular cell
• Step 1 Move i cells along any chain of hexagons
• Step 2 Turn 60 degrees counterclockwise and move
  j cells.
• i and j measure the number of nearest neighbors
  between co-channel cells
• The cluster size, N,
• N i2ijj2

i
3
2
1
4
If i 2 and j 0, then N 4 If i 2 and j 1,
then N 7
1
2
3
2
1
26
RULE to Determine the Nearest Co-Channel
Neighbors Determining the Cluster Size
27
Frequency Reuse

• The distance between 2 cells using the same
  channel is known as the REUSE DISTANCE D.
• There is a close relationship between D, R
  (radius of each cell) and N (the number of cells
  in a cluster)
• D (sqrt 3N) . R
• The REUSE FACTOR is then
• D/R sqrt (3N)

28
Frequency Reuse

• Let N be the cluster size in terms of number of
  cells within it and K be the total number of
  available channels without frequency reuse.
• N cells in the cluster would then utilize all K
  available channels.
• Each cell in the cluster then uses 1/N-th of the
  total available channels.
• N is also referred as the frequency reuse factor
  of the cellular system.

29
Capacity Expansion by Frequency Reuse

• Assume each cell is allocated J channels (JltK).
  If the K channels are divided among the N cells
  into unique and disjoint channel groups, each
  with J channels, then K J N
• The N cells in a cluster use the complete set of
  available frequencies.
• The cluster can be replicated many times.
• Let M be the number of replicated clusters and C
  be the total number of channels in the entire
  system with frequency reuse, then C is the system
  capacity and computed by
• C M J N

30
Cellular System Capacity - Example

• Suppose there are 1001 radio channels, and each
  cell is Acell 6 km2 and the entire system
  covers an area of Asys 2100km2.
• Calculate the system capacity if the cluster size
  is 7.
• How many times would the cluster of size 4 have
  to be replicated in order to approximately cover
  the entire cellular area?
• Calculate the system capacity if the cluster size
  is 4.
• Does decreasing the cluster size increase the
  system capacity?
• Solution

1. JK/N143, AclusterN642km2, M2100/4250,
CMJN50,050 chs.
2. N4, Ac4624km2, M2100/2487.
3. N4, J 1001/4 250 chs/cell. C 87 250
4 87,000 chs.

• Decrease in N from 7 to 4? increase in C from
  50,050 to 87,000.
• ? Decreasing the cluster size increases system
  capacity. So the answer is YES!

31
Geometry of Hexagonal Cells (1)How to determine
the DISTANCEbetween the nearest co-channel cells
?

• Planning for Co-channel cells
• D is the distance to the center of the nearest
  co-channel cell
• R is the radius of a cell

j
D
i
30o
R
0
32
Geometry of Hexagonal Cells (2)

• Let D be the actual distance between two centers
  of adjacent co-channel cells where D
• Let Dnorm be the distance from the center of a
  candidate cell to the center of a nearest
  co-channel cell, normalized with respect to the
  distance between the centers of two adjacent
  cells, .
• Note that the normalized distance between two
  adjacent cells either with (i1,j0) or (i0,j1)
  is unity.

33
Geometry of Hexagonal Cells (3)

• Let D be the actual distance between the centers
  of two adjacent co-channel cells. D is a function
  of Dnorm and R.
• From the geometry we have

• From N and Dnorm equations

34
Geometry of Hexagonal Cells (4)

• With the actual distance between the centers of
  two
• adjacent hexagonal cells, the actual distance
  between the center
• of the candidate cell and the center of a nearest
  co-channel is
• then

• For hexagonal cells there are 6 nearest
  co-channel neighbors to each cell.
• Co-channel cells are located in tiers.
• In general, a candidate cell is surrounded by 6k
  cells in tier k.
• For cells with the same size the co-channel cells
  in each tier lie on the boundary of the hexagon
  that chains all the co-channel cells in that
  tier.

35
Geometry of Hexagonal Cells (5)

• As D is the radius between two nearest
• co-channel cells, the radius of the hexagon
  chaining
• the co-channel cells in the k-th tier is given
  by k.D.
• For the frequency reuse pattern with i2 and j1
  so
• that N7, the first two tiers of co-channel
  cells are
• given in Figure.
• It can be readily observed from Figure that the
  radius of
• the first tier is D and the radius of the
  second tier is 2.D.

36
(No Transcript)
37
Number of Cells in A Cluster

• A candidate cell has 6 nearest co-channel cells.
  Each of them in turn has 6 neighboring co-channel
  cells. So we can have a large hexagon.
• This large hexagon has radius equal to D which is
  also the co-channel cell separation.
• The area of a hexagon is proportional to the
  square of its radius, (let ?2.598),

R
D
38
Number of Cells in A Cluster

• The number of cells in the large hexagon is then

• In general the large hexagon encloses the center
  cluster of N cells
• plus 1/3 the number of the cells associated
  with 6 other
• peripheral large hexagons.
• Hence, the total number of cells enclosed by the
  large hexagon is

39
Geometry of Hexagonal Cells (6)

• We assume the size of all the cells is roughly
  the same, as long as the cell size is fixed
  co-channel interference will be independent of
  transmitted power of each cell.
• The co-channel interference will become a
  function of q where q D/R sqrt (3N).
• q is the CO-CHANNEL REUSE RATIO and is related to
  the cluster size.
• A small value of q provides larger capacity since
  N is small.
• For large q, the transmission quality is better,
  smaller level of co-channel interference.
• By increasing the ratio of D/R spatial separation
  between co-channel cells relative to the coverage
  distance of a cell is increased.Thus,
  interference is reduced from improved isolation
  of RF energy from the nmber of cells per cluster
  N co-channel cells.

40
Geometry of Hexagonal Cells (7)

• Furthermore, D (distance to the center of the
  nearest cochannel cell) is a function of NI and
  S/I in which NI is the number of co-channel
  interfering
• cells in the first tier and S/I received signal
  to interference ratio at the desired mobile
  receiver.

41
Frequency Reuse Ratio

• The frequency reuse ratio, q, is defined as
• q D/R
• which is also referred to as the co-channel
  reuse ratio.
• Also ? q sqrt(3N)
• Tradeoff
• q increases with N.
• A smaller value of N has the effect of
  increasing the capacity of the cellular system
  and increasing co-channel interference
• Tradeoff between q and N

42
Interference

• MAJOR LIMITING FACTOR for Cellular System
  performance is the INTERFERENCE
• Implications
• ? CROSS TALK
• ? Missed and Blocked Calls.
• SOURCES OF INTERFERENCE?
• Another mobile in the same cell
• A call in progress in neighboring cell.
• Other base stations operating in the same
  frequency band
• Non-cellular systems leaking energy into cellular
  frequency band

43
Interference

• 1. CO-CHANNEL INTERFERENCE
• 2. ADJACENT CHANNEL INTERFERENCE

44
CO-CHANNEL INTERFERENCE

• Frequency Reuse ? Given coverage area? cells
  using the same set of frequencies ? co-channel
  cell!!!
• Interference between these cells is called
• CO-CHANNEL INTERFERENCE.
• (Thermal noise ? increase SNR and combat it).
• However, co-channel interference ? cannot be
  overcome just by increasing the carrier power of
  a transmitter.
• Because increase in carrier transmit power
  increases the
• interference.
• Reduce co-channel interference
• Co-channel cells must be physically separated
  by a minimum distance to provide sufficient
  isolation.

45
Co-Channel Interference

• Intracell Interference interferences from other
  mobile terminals in the same cell.
• Duplex systems
• Background white noise
• Intercell interference interferences from other
  cells.
• More evident in the downlink than uplink for
  reception
• Can be reduced by using different set of
  frequencies
• Design issue
• Frequency reuse
• Interference
• System capacity
• Bottomline It determines link performance which
  in turn dictates the frequency reuse plan and
  overall capacity of the system.

46
Co-Channel Interference
Cell Site-to-Mobile Interference (Downlink)
Mobile-to Cell-Site Interferences (Uplink)
47
Co-Channel Interference
Base ? Mobile ? DOWNLINK Mobile? Base ?
UPLINK UPLINK? All mobiles in 6 cells central
cell assigned to the same frequency
channel DOWNLINK? All base stations in 6 cells
and central cell have the same
frequency channel. DOTTED LINES show the
interference of all 6 mobiles (all co-channel)
received at central base station
(interference) Actual signal is from the mobile
in the center cell to its own base
station. (Uplink Signal Interference ratio)
48
Co-Channel Interference
Base ? Mobile ? DOWNLINK CASE From the base
stations (from co-channel cells) interference
received by the mobile in the center
cell. Desired signal is from the base to mobile
in the center cell. Alarge is the area of the
hexagonal cells of the large one. Asmall is the
area of each cell. Alarge/Asmall ? A number of
cells in this each repetitous pattern (3N).
49
Co-Channel Interference

• For simplicity, we consider only the average
  channel quality as a function of the distance
  dependent path loss.
• Signal-to-Co-channel interference ratio, (S/I),
  at the desired mobile receiver which monitors the
  forward channel is defined by
• S is the desired signal power from desired base
  station
• Ii interference power caused by the i-th
  interfering co-channel cell base station.
• NI is the number of co-channel interfering cells

50
Co-Channel Interference

• The desired signal power S from desired base
  station
• is proportional to r-?, where r is the
  distance between the mobile and the serving base
  station. ? is the path loss component.
• The received interference, Ii, between the ith
  interferer and the mobile is proportional to
  (Di)-?.
• The white background noise is neglected in the
  interference-dominant environment.
• Assume the transmisson powers from all base
  stations are equal, then we have

51
Co-Channel Interference

• Consider only the first tier of interfering
  cells, if all interfering base stations are
  equidistant from the desired base station and if
  this distance is equal to the distance D between
  cell centers, then the above equation can be
  simplified to
• (i.e., rR and assume DiD and use qD/R)

52
Co-Channel Interference

• Frequency reuse ratio,
• e.g., NI 6 ?

53
Co-Channel Interference

• Example In AMPS systems
• for ?4, S/I 18dB (i.e., 63.1),
• 20 log (S/I) ? dB
• are acceptable then (assume N6)
• q (6? 63.1)1/4 ? 4.41.
• Thus, the cluster size N should be
• (from eq. qsqrt(3N)? N q2/3 6.79 ? 7.
• i.e.,A 7-cell reuse pattern is needed for an S/I
  ratio of 18dB. Based on qD/R, we can select D by
  choosing the cell radius R.

54
Co-Channel Interference

• An S/I of 18 dB is the measured value for the
  accepted voice quality from the present day
  cellular mobile receivers.
• Sufficient voice quality is provided when S/I is
  greater than or equal to 18dB.

55
Example Co-Channel Interference

• If S/I 15 dB required for satisfactory
  performance for forward channel performance of a
  cellular system.
• What is the Frequency Reuse Factor?
• What Cluster Size should be used for maximum
  capacity?
• (Use path loss component of ?3 and ?4)
  .
• Assume 6 co-channels all of them (same distance
  from the mobile)

56
Example Co-Channel Interference

• N 7 and ?4
• The co-channel reuse ratio is qD/Rsqrt(3N)4.583
  

Or 18.66 dB ? greater than the minimum required
level ? ACCEPT IT!!! b) N 7 and ?3

• Or 12.05 dB ? less than the minimum required
  level ? REJECT IT!!!

57
Example Co-Channel Interference

• We need a larger N. Use eq. N i2ijj2
• for ij2 ? next possible value is N12.
• qD/Rsqrt(3.N) 6 and ?3

Or 15.56 dB ? N12 can be used for minimum
requirement, but it decreases the capacity since
12 cell reuse offers a spectrum utilization of
1/12 within each cell.
58
Worst Case Co-Channel Interferencei.e., mobile
terminal is located at the cell boundary where it
receives the weakest signal from its own cell but
is subjected to strong interference from all all
the interfering cells.

• We need to modify our assumption, i.e., assume
  DiD.
• The S/I ratio can be expressed as

Used D/Rq and ?4. Where q4.6 for normal
seven cell reuse pattern.
59
Example Worst Case Cochannel Interference (2)

• A cellular system that requires an S/I ratio of
  18dB.
• (a) If cluster size is 7, what is the
  worst-case S/I?
• (b) Is a frequency reuse factor of 7
  acceptable
• in terms of co-channel interference?
• c) If not, what would be a better choice of
• frequency reuse ratio?

60
Example Worst Case Cochannel Interference (2)

• Solution
• (a) N7 ? q .
• If a path loss component of ?4, the
  worst-case signal- to-interference ratio is
  S/I 54.3 or 17.3 dB.
• (b) The value of S/I is below the acceptable
  level of 18dB. To ncrease S/I ? we need to
  decrease I,
• I.e., Increase the frequency reuse factor,
  q5.20 by using N 9.
• The S/I becomes then 95.66 or 19.8dB.
  Acceptable

61
ADJACENT CHANNEL INTERFERENCE

• Interference resulting from signals which are
  adjacent
• in frequency to the desired signal is called
• ADJACENT CHANNEL INTERFERENCE.
• WHY?
• From imperfect receiver filters (which allow
  nearby frequencies) to leak into the pass-band.
• NEAR FAR EFFECT
• Adjacent channel user is transmitting in very
  close range to a subscribers receiver, while the
  receiver attempts to receive a base station on
  the desired channel.
• NEAR FAR EFFECT also occurs, when a mobile close
  to a base station transmits on a channel close to
  one being used by a weak mobile.
• Base station may have difficulty in
  discriminating the desired mobile user from the
  bleedover caused by the close adjacent channel
  mobile.

62
ADJACENT CHANNEL INTERFERENCE

• How to reduce?
• Careful filtering
• Channel assignment? no channel assignment which
  are all adjacent in frequency.
• Keeping frequency separation between each channel
  in a given cell as large as possible.
• e.g., in AMPS System there are 395 voice channels
  which
• are divided into 21 subsets each with 19
  channels.
• In each subset, the closest adjacent channel is
  21 channels away.
• 7-cell reuse -gt each cell uses 3 subsets of
  channels.
• 3 subsets are assigned such that every channel in
  the cell is assured of being separated from every
  other channel by at least 7 channel spacings.

63
Cell Splitting

• A method to increase the capacity of a cellular
  system by dividing one cell into more smaller
  cells.
• Each smaller cell has its own base station and
  accordingly antenna height and transmission power
  can be reduced.
• Cell splitting reduces the call blocking
  probability because the number of channels is
  increased.
• But it increases the handoff rate, i.e., more
  frequent crossing of borders between the cells.
• We have the formula in calculating path loss
  Pr(dBW) P0(dBW) - 10 ?log10(d/d0)
• where d0 is the distance from the reference
  point to the transmitter, and P0 is the power
  received at the reference point.

64
Cell Splitting (2)

• Let Pt1 and Pt2 be the transmit power of the
  large cell BS and medium cell BS, respectively.
• The received power at the edge of large cell is
• Pr1 P0 - 10 ? log10(R/d0)
• The received power at the edge of large cell, Pr1
  is proportional to
• Pt1 (R)- ?.
• The received power at the edge of R/2 cell, Pr2
  is proportional to
• Pt2 (R/2)- ?.
• With the equal received power, we have Pt1 (R)-?
  Pt2 (R/2)- ?, i.e., Pt1/Pt2 2 ?

R
R/2
65
Example Cell Splitting

• Suppose each BS is allocated 60 channels
  regardless of the cell size. Find the number of
  channels contained in a 3x3 km2 area without cell
  splitting, i.e., R 1km and with cell splitting,
  R/2 0.5km.

• The number of cells for R1km.
• Each large cell can cover 3.14km2, for 9 km2
  approximately need 9/3.14 gt 3 cells. However, 3
  hexagon cannot cover a square of 3x3. A better
  approximation is 4 cells. So the number of
  channels is 4x60240.
• With small cells, the number of cells is
  approximately (1/0.5)2x4 16. Then the number of
  channels is 16x60960.

66
Cell Sectoring(Directional Antennas)

• Omni-directional antennas allow transmission of
  radio signals with equal power strength in all
  directions.
• Reality is an antenna covers an area of 60
  degrees or 120 degrees ? DIRECTIONAL ANTENNAS!!!!
• Cells served by these antennas are called
  SECTORED CELLS!!!
• Many sectored antennas are mounted a BS tower
  located at the center of the cell and an adequate
  number of antennas is placed to cover the entire
  360 degrees of the cell.

.
67
CELL SECTORINGDirectional Antennas (Sectoring)
1
1
2
2
4
3
3
90 DEGREE SECTOR
120 DEGREE SECTOR
OMNIDIRECTIONAL
6
1
5
4
2
3
60 DEGREE SECTOR
68
Cell Sectoring(Directional Antennas)

• Advantages of Cell Sectoring
• Borrowing of channels
• Coverage of smaller area by each antenna and
  hence lower power is required in transmitting
  radio signals.
• Helps to decrease interference between
  co-channels.
• Also the spectrum efficiency of the overall
  system is enhanced.

.
69
Co-Channel Interference Reduction with the Use of
Directional Antennas (Cell Sectoring)

• The basic form of antennas are omnidirectional.
  Directional antennas can increase the system
  capacity.
• If we sectorize the cell with
• 120o in each sector, the S/I
• becomes
• (The number of interferers is reduced from 6
  to 2.)
• The capacity increase in (S/I)120 vs (S/I)omni
  is 3.

1
2
3
1
2
3
70
Worst-Case Scenario in 120o Sectoring

• Let D be the distance between the adjacent
  co-channel cells.
• With the distance approximation and use path loss
  component ?, the signal-to-interference ratio is

D0.7R
1
2
3
D
71
Fixed Channel Assignment (FCA)

• Each cell is allocated a predetermined set of
  voice channels.
• The BS is the entity that allocates channels to
  the requests. If all channels are used in one
  cell, it may borrow a channel from its neighbors
  through MSC.
• Fast allocation, but may result high call
  blocking probabilities.

72
Dynamic Channel Assignment (DCA)

• Voice channels are not allocated to each cell
  permanently.
• When a request is received at the BS, this BS
  request a channel from MSC.
• DCA can reduce the call blocking probability, but
  it needs real-time data collection and signaling
  transmission between BS and MSC.

73
Call Admission Control

• CAC is used to avoid congestions over the radio
  links and to ensure the QoS requirements of
  ongoing services.
• Quality of service (QoS)
• Packet-level factors
• Packet loss rate, packet delay, packet delay
  variation, and throughput rate.
• Grade of service (GoS)
• Call-level factors
• New call blocking probability, handoff call
  dropping probability, connection forced
  termination probability.

74
CAC Procedure

• Determine the amount of available channels, i.e.,
  the number of channels for accepting new and
  handoff requests.
• When the N-th request arrives, i.e., there are
  (N-1) ongoing services.
• If there are enough resources to admit the N-th
  request, then the new request is admitted.
• Otherwise, it will be denied.
• In order to maintain the continuity of a handoff
  call, handoff calls are given higher priority
  than the new call requests.
• The prioritized call admission is implemented by
  reserving channels for handoff calls. This method
  is referred to as guard channels.
• Fixed reservation and dynamic reservation.

75
Cell Capacity

• The average number of mobiles requesting service
  (average call arrival rate) ?
• The average length of time a mobile requires
  service (the average holding time) T
• The offered traffic load A ?T
• e.g., in a cell with 100 mobiles, on an average,
  if 30 requests are generated during an hour, with
  average holding time T360 seconds, then the
  arrival rate ?30/3600 requests/sec.
• A servicing channel that is kept busy for an hour
  is quantitatively defined as one Erlang.
• Hence, the offered traffic load (A) by Erlang is
  then

76
Call Blocking

• How likely a new user can get a connection
  established successfully? Admission control of
  new calls.
• It is measured by the probability of call
  blocking, which is a quality of service (QoS)
  factor, a.k.a., (GoS) factor.
• Assume we have a total number of C channels in a
  radio cell.
• If the number of active users during any period
  of time is C, then the call blocking probability
  is 1.
• If and only if the number of ongoing calls is
  less than C, the probability of call blocking
  will be less than 1.

77
Erlang B

• Probability of an arriving call being blocked
  Prob that (Blocked Calls Cleared Lost)) is
• where C is the number of channels in a cell.
• Prob(A,C) is also called blocking probability,
  probability of loss, or probability of rejection.

78
Erlang C

• Probability of an arriving call being delayed,
  i.e.,
• probability that no trunk (serverchannel) is
  available for an arriving call in a system with C
  channels and the call is delayed, is

where Prob(A, C) is the probability of an
arriving call being delayed with load A and C
channels.
79
SEMANTICS

• Prob that calls are lost..
• GOS for telephone calls (realstic values 10-2
  10-3
• Physical Interpretation-gt Ratio of calls rejected
  to the total number of calls.
• What can we do with these Erlang formulas?
• Given a fixed offered traffic A in Erlangs and a
  fixed device capacity C, then find the prob of
  blocking
• Determine the offered traffic in Erlangs that
  produces a given blocking probability for a fixed
  device capacity C.
• Determine the required device capacity given the
  blocking probability and the offered traffic in
  Erlangs.

80
EXAMPLE

• An urban PC area has a population of 2 Million
  residents. A cellular company serves this area.
  System has 394 cells with 19 channels each.
• Find the number of users that can be supported at
  2 blocking if each user averages 2 calls/per
  hour at an average call duration of 3 minutes!!

81
EXAMPLE

• Prob of Blocking? 0.02 (GOS)
• Number of Channels ? C19
• Traffic Intensity per User ? A/mu 2 3/60 0.1
  Erlangs
• From Erlang B chart, total carried traffic
  obtained as 12 ERLANGS.
• The number of users which can be supported per
  cell is
• 12/0.1 120.
• There are 39 cells ? total number of subscribers
  supported is 120 394 47280.

82
Efficiency (Utilization)

• Example for previous example, if C2, A 3
• then
• B(2, 3) 0.6, ------ Blocking probability,
• i.e., 60 calls are blocked.
• Total number of rerouted calls 30 x 0.6 18
• Efficiency 3(1-0.6)/2 0.6

83
Summary

• The advantage of cellular communications
• Capacity extension by frequency reuse
• Cell cluster and cochannel cells
• Number of cells in a cluster
• Frequency reuse ratio
• Co-Channel interference
• Impact of cluster size
• Worst-case cochannel interference
• Traffic load and call blocking probability
• Average delay
• Probability of queuing delay
• Cell splitting and sectoring
• Fixed channel allocation and dynamic channel
  allocation