Title: Cellular Communications Operational Details
1Cellular Communications Operational Details
2The Use of Frequency Modulation
- The Concept of 'Capture Ratio'
- If two FM signals of the same frequency but
drastically different power levels (like 50x)
exist in the same area, the receiver can pick up
the stronger signal without interference from the
weaker. - This allows a frequency to be 'reused' in nearby
cells. - Strength of signal varies as the 4th power of the
distance and even more due to blockage - Capture Ratio
- The ratio of the 'desired' or 'local' signal to
the interference signal that is necessary in
order to extract only the desired signal is known
as the 'capture ratio'. - C D/U
- Where C Capture Ratio, D Desired Signal, U
Undesired Signal - AM does not have this capability, therefore all
cellular communications use some form of FM -
typically phase modulation.
3Electromagnetic Spectrum
LIGHT
HARMFUL RADIATION
RADIO
SOUND
VHF VERY HIGH FREQUENCY UHF ULTRA HIGH
FREQUENCY SHF SUPER HIGH FREQUENCY EHF EXTRA
HIGH FREQUENCY
4North American 800 MHz Cellular Spectrum
- Total Range of 50 MHz (824-849 869-894)
- 824-849 is for Uplink, 869-894 is for Downlink.
25 MHz each - A and B regions assigned in 1981
- Additional carriers (A and B) assigned in 1987
- Control Carriers
- Area between A and B used for AMPS Control
Carriers - IS-54 and -136 allows any frequency to be used
for TDMA setup carrier - IS-95 (CDMA) uses 10 chunks each 1.25 MHz
bandwidth
5800 MHz Channel Allotment Breakdown
- Channel Breakdown
- Uplink 824 to 849 Downlink 869 to 894 Both
25 MHz range. - Each channel occupies 30 kHz
- Total channels per Range 25MHz/30kHz 832
- Each area has two companies (by FCC regulation),
each gets half 416 - Of those, 21 are for signaling, leaving 395
- Reuse Factor
- Frequency reuse is limited due to adjacent cell
interference - determined by the reuse factor
(rF). - The Reuse factor can vary from 1 to 21. The
lower the value the better. Typical values - AMPS 7 395/7 32
- GSM and other TDM typically are 4 395/4 98
- 1 is possible only with CDMA 395/1 395
- Note CDMA has a typical reuse factor of 4.
61900 MHz PCS Spectrum
- FCC PCS Spectrum Allocation - June 9, 1994
- Blocks A B are for use in Metropolitan Trading
Areas (MTAs) - Blocks C, D, E F for use in Basic Trading
Areas (BTAs) (suburban or rural) - Cellular operators are eligible for only one 10
MHz block in their existing services areas
7Sprint PCS
- The Arthur Andersen partner was on his cell phone
when he said... - Ship the Enron documents to the feds.
- But, his secretary heard,
- Rip the Enron documents to shreds.
- It turns out that it was all just a case of bad
cellular.
Sprint PCS The clear alternative to cellular
8Frequency Reuse
- Cellular/PCS is all about 'frequency reuse'.
- Handover is nothing new, and not that technical
by today's standards. - If you have a mobile user moving from one base
station to another, it is no problem. Whichever
antenna is receiving the mobile signal the
strongest will 'take over' the communications
with that unit. - Limited Frequencies gives developers two choices
- 1. Each user has his own frequency
- 2. User frequency changes depending upon
available frequencies in the area at that time. - You have to be able to change which frequency he
is using. I.e. it is necessary to re-program his
mobile transmitter to fit the instantaneous needs
of the overall system. - Only been made possible with recent
technological advances - particularly
miniaturization, microprocessors, and DSPs. - The base stations can 're-configure' the mobile
station to use a new frequency and a different
power level.
9Techniques to Add Capacity
The wireless business is all about spectrum
availability!! Install more RF carriers in cell
Sectorize (if originally omni) and reduce
cluster group n from e.g. from 7 to 4 then
install more carriers Overlay center part of
cell with lower power carrier(s) also called
tiered cell coverage Only adds capacity in
central cell portion not applicable to single
frequency CDMA Split cell, and install TC/n
carriers in new small cells Use half-rate
speech coder (if acceptable quality) Use
smart directional adaptive antennas
(when available and economical)
10Sectoring
- Created by installing multiple directional
antennas in a cell, each with shielding on the
'back'. - Only receive signals from the 'front', thereby
reducing the level of interference by a ration of
3/1 (6/1 for 6 sectors). - This improves signal quality, which means that
the cluster can be redesigned for a smaller rF.
For example, from 9 to 7, or 7 to 4. - Back and Side lobes are problems - co-channel
interference, which is intermittent and difficult
to identify and debug. - Smart antennas' (adaptive phase arrays) which
address this problem, but at a very high cost.
11Cell Splitting
- Increase of capacity by 7 in center cell (for
rF7) - There is a minimum cell size (due to approx.
- Minimum 5 mW handset Tx power) so you cant
split again and again without limitation - High real estate cost of new antennas/cells is a
deterrent - Cell splitting is the last choice economically,
after using methods which add capacity to an
existing cell site
12Voice CODECs
13Access Technologies
36 refers to 3 conversations at present,
planned 6 in the future with half rate speech
coder. 816 refers to 8 conversations at
present, planned 16 in the future with half-rate
speech coder.
14AMPS and N-AMPS
AMPS and N-AMPS have given way to D-AMPS (Digital
AMPS IS-54) in all but a few isolated locations
in North America. AMPS and its derivatives were
never used in any other major installations - a
few small countries. N-AMPS provided better
filtering techniques and therefore allowed the
channel width to be reduced to 12 kHz rather than
30kHz - allowing three times the capacity of
AMPS, however, the sound quality was not
satisfactory and did not receive good customer
response.
15Reasons for going to Digital Voice
- Keep in mind - the key issue is spectrum
efficiency - getting the most out of the
available spectrum. - TDM can be used. With SDM we must reserve an
entire channel (frequency) for our 'control' or
'signaling'. With TDM we can break the channels
into smaller chunks (timeslots) and allocate
timeslots rather than entire frequencies. - TDM also reduces the number of necessary
'guardbands'. Although, in IS-54, the same
frequency ranges were used (30kHz) in order to
insure backward compatibility with AMPS
equipment. - Digitized speech allows the application of newer
speech encoders, which keep reducing the bit rate
needed for transmitting speech. - Digitized information allows us to use the
advanced techniques of modulation, such as phase
shift keying, which are much more efficient than
straight analog FM.
16IS-54 (Interim Standard 54)
- Digital
- Replaces AMPS throughout North America - known
as D-AMPS (Digital AMPS) - TDM
- Six time slots per frame
- 1,944 bits/frame
- 40 mSec/frame 25 frames/sec
- Up to Six times the capacity of AMPS (using
half-rate). Typically 3 x AMPS. - 48.6kBps data rate over 30 kHz bandwidth
- Uses D-QPSK Modulation (Differential Quadrature
Phase Shift Keying) - Uses 8kbps CODEC 5kbps FEC (Error Correction
Bits) 13kpbs voice transfer rate.
17IS-54 Backward Compatibility with AMPS
- A key benefit that can be recognized by using TDM
is the capability of slicing up the available
bandwidth into smaller chunks - the concept of
Time Slots. - The channel bandwidth can now be enlarged, since
it is not dictated by the amount of data that
needs to be transferred - this reduces the number
channels and number of guardbands. - Then these timeslots can be used in a variety
of ways. Small amounts of data do not need to
occupy an entire channel bandwidth. - In order to retain backward compatibility with
AMPS - There still had to be control channels -
therefore IS-54 did not totally realize the full
benefits of going to TDM (that of integrating the
data and control information). - The 30kHz channel bandwidth had to be retained
- This illustrates the common inefficiency of
maintaining backward compatibility
18IS-54 Frame Format - Half Rate
- Low-rate traffic (data)
- Six separate users, each with their own timeslot
- Each user gets one timeslot per frame
- One frame is 1944 bits (324b/Slot x 6 Slots)
- Frame lasts 40 mSec
19IS-54 Frame Format - Full Rate
- High-rate traffic (voice)
- Three separate users, each with two timeslots
per frame - One frame is still 1944 bits (324b/Slot x 6
Slots) - Frame still lasts 40 mSec
20IS-54 One Time Slot (Downlink)
Frame format Bits 1-28 (28 bits)
Synchronization Bits 29-40 (12 bits) SACCH
(Slow Associated Control Channel) -
housekeeping Bits 41-170 (130 bits) Data Bits
171-182 (12 bits) CDVCC Bits 183-312 (130
bits) Data Bits 313-324 (12 bits) Future
Use Total Data per Time Slot 260 bits One Time
slot/frame 1 Timeslot/40 mSec 6.5kbps Two
Time slots/frame for voice 13kbps Total
transfer rate 324 bits/6.67 mSec 48.6kbps
21IS-54 One Time Slot (Uplink)
Frame format Bits 1-6 (6 bits) Guard time.
Bits 7-12 (6 bits) Ramp time Bits 13-28 and
57-178 (138 bits) Information Bits 29-56 (28
bits) Synchronization. Bits 179-190 (12 bits)
SACCH Bits 191-202 (12 bits) CDVCC Bits 203-324
(122 bits) Information Total Data per Time Slot
224 bits One Time slot/frame 1 Timeslot/40 mSec
5.6kbps Two Time slots/frame for voice 11.2kbps
22IS-54 Modulation
- IS-54 uses Differential Quadrature Phase Shift
Keying (D-QPSK) Modulation - 8-states
- Differential means that the transition from
state to state can be more 'gentle' resulting in
less harmonics generated.
23IS-136
IS-136 was the next generation after IS-54. It
extended the use of TDMA to the control channel.
This technology is now recognized by ANSI, and
published as the TIA/EIA-136 series of
standards. IS-54 and IS-136 are completely
compatible, and often considered the same
technology.
24GSM
- Groupe Speciale Mobile is original acronym.
- Has come to mean Global System for Mobile.
- Most popular Cell system in the world today
- Developed in Europe
- Designed to be compatible with ISDN
- TDMA and FDMA system
- TDM
- 8 timeslots per frame
- FDM
- Uses Frequency Hopping at 217 hops/second
- Frequency is hopped after each frame
- Compensates for signal fading
25The GSM Frame
- Each frame is allotted a 200kHz bandwidth.
- Frequency bands are 890 to 915 MHz uplink and 935
to 960 MHz for downlink (note difference from NA
allotments) - A frame will be transmitted over one of the given
channels (200kHz bands), and then the next frame
switches to a new channel. (Frequency Hopping) - Therefore, all data is is 'distributed' over
several different frequencies. This is forced by
the timeslots. - There are forward error-checking bits, which
can be described as duplication of data (thus
Sveums claim that data is duplicated. It is not
actually).
26GSM Frame Format
27GSM User Bit-Rate Calculations
- User has...
- 1 Time slot/frame
- 114 bits/timeslot
- 24 User frames/multiframe
- 1 multiframe/120 mSec
- Data rate 114 x 24/120mSec 22.8kbps
- FEC and Encryption can rob some more - usually
leaving between 13 and 21.4kbps
28GSM Time Slot Breakdown
Header and Footer are empty space at the
beginning and end of the slot. They serve to
separate a slot from its neighbors, negating
propagation delay at any distance up to 35 km
from the base station. Training Sequence A
fixed pattern in the middle of the slot, used to
help a receiver lock on to the slot - i.e. frame
synchronization. It also acts as a header and
footer if the slot is divided between two users,
as it would be when the half-rate CODEC is
used. Stealing Bits Identify the time slot as
being 'stolen' by the GSM system to carry control
information.
29GSM Modulation
Like other digital cellular technologies, GSM
encodes data using a form of Phase Modulation.
The precise type used in GSM is known as GMSK
(Gaussian Minimum Shift Keying), which achieves a
symbol rate of 270.8kbps in a 200kHz channel.
This is only in one direction, so GSM uses
separate paired channels to send and receive,
just as IS-54.
30Spread Spectrum
Spread spectrum is a type of modulation that
spreads data transmission across the available
frequency band, in excess of the minimum
bandwidth required to send the information.
Spreading the data across the frequency
spectrum makes the signal resistant to noise,
interference and eavesdropping. Two Types FHSS
- Frequency Hopping SS DSSS - Direct Sequence SS
31SS - Advantages Disadvantage
- Advantages
- Has the ability to eliminate or alleviate the
effects of multi-path interference - Can share the same frequency band with other
users - Provides privacy due to unknown random codes
- Involves low power spectral density since signal
is spread over a large frequency band - Disadvantages
- Can be Bandwidth inefficient
- Implementation is somewhat complex
32Frequency-Hopping Spread Spectrum
- Frequency Hopping is one type of a general
category of techniques known as spread spectrum
(FHSS) - GSM uses FHSS
- provides security
- provides resistance to interference (Fading)
- transmitted frequency is changed according to a
pseudo random code - the amount of time that one frequency is used is
typically less than the travel time - preventing
jamming and/or eavesdropping.
33Direct Sequence Spread Spectrum
Whereas FHSS uses several different frequency
bands, and transmissions 'hop' from band to band,
DSSS uses a single large frequency band, and
spreads the users' signals out over the entire
band. Since there are not multiple bands, there
is no 'frequency hopping'. Essentially, each user
is using the entire bandwidth. DSSS is probably
the most widely recognized form of spread
spectrum.
34DSSS Operation
A DSSS transmitter converts the original data
stream into a 'symbol stream' where each symbol
represents a group of one or more bits. The DSS
transmitter modulates (usually called
'multiplying') each symbol with a noise-like code
called a pseudo-random noise (PN) sequence -
known as a 'chip' sequence. The chip sequence
is at the frequency of the 'wide bandwidth' At
the receiving end, these unique (known as
'orthogonal') chipping patterns are added back to
the received waveform, producing the original
transmitted pattern.
35CDMA - Code Division Multiple Access
Each user in a DSSS system can be assigned a
unique chipping code that will not 'interfere'
with any other users chipping codes. The
signals from all of the different users are
'summed' in the air, so that the receiver
receives the total transmission. The receiver
multiplies the received signal with the
appropriate chipping code to recover the users
original signal
36Advantages of CDMA
- Relatively easy to configure different data rates
for different users - Easier to coordinate use of 'silent' periods.
- The use of VAD (voice activity detection)
improves capacity by about 2 to 1 compared to
continuous transmitting, since speakers are
normally silent 40 - 60 of the time. - Other systems have demonstrated VAD but all
require added coordinating signals. - In a CDMA system it is not necessary to send
additional signals between the base and mobile to
coordinate channel assignment - this is the most
significant feature of CDMA.
37CDMA Problems
- The 'Near-Far' problem
- Evaluation of the summed signal depends on the
relative strength of each piece - The transmitter's distance from the receiver will
effect the signal strength - All of the transmitters must control their power
so that the received signal is normalized. - The 'Synchronization' Problem
- The signals are 'summed' in the air. They must
exactly coincide or the summing action will not
be proper. - Since different transmitters are at different
distances from the receiver, even if they are all
'synchronized' to the same clock, their signals
will still arrive at different times. - Therefore, the base station must calculate how
much to delay each of the user's transmissions in
order for them to arrive at the receiver all at
the same time.
38IS-95 (QualComm CDMAone)
- Uses CDMA (DSSS)
- A 1.25 MHz channel is used.
- The 1.25 MHz spectrum can (theoretically) be
shared by 400 users about 3kHz/user - Chipping code is known as a Walsch Codes
- The number of users that can use the system at
the same time depends on a number of factors - Precise controlling of mobile units transmission
power - Exact synchronization
- Still, the signals from different distances
arrive at different times. These must be
'time-correlated'. A special type of receiver,
known as a RAKE receiver accomplishes this
through internal delaying - The end result is that CDMA, though theoretically
capable of more users, does not have more
capacity than TDMA - about 6 to 8x the AMPS.
39IS-95 (CDMA) Debates
- Sveum says that the number of users can be
greatly increased over other cellular systems.
However, most current sources dispute this. - When the number of users is increased, the BER
increases. Therefore, users can be increased with
reduction in dependability. This is often
marginally acceptable for voice, but will be
intolerable for data. - Also, remember to compare 'apples to apples'.
Sveum says that the increase is 12 to 30 times
that of AMPS (an analog system). To be fair, it
needs to be compared with TDMA, the other current
technology. - Adjacent cells can use the same frequencies, and
vary the Walsh codes to prevent interference. - The use of Wide BW is definitely effective in
fighting Fade.
40Multiple Access Analogy
Code Division
Time Division
Frequency Division
SOURCE WASHINGTON UNIV.
41Multiple Access Comparison
Access technology debates between FDMA, TDMA
(and later CDMA) are also called the religious
wars 1. The inherent traffic capacity of
all 3 technologies is about the same Proper
Measure conversations/kHz/km2 2.
Significant differences in implemented system
capacity arise from secondary non-access
features, for example Speech coder Voice
activity control (DSI combined with Frequency
Hopping, TASI) 3. The decisions regarding
competing technologies depend on overall
performance and economic criteria conversations/k
Hz/km2 /