(3G) Radio Network Planning for WCDMA

Radio Network Planning for WCDMA

Introduction

With the emergence of 3G mobile communication technology, the construction of the

UMTS network will bring a profound evolution, which makes higher requirements for

network planning. At present, the public is greatly interested in this new technology.

The construction of 3G mobile communication network is in the ascendant. This new

mobile communication technology is different from the traditional GSM network

planning in terms of essence. Worldwide people are developing new planning tool and

algorithm and designing new work flow.

Comparison between UMTS network planning and GSM network planning:

1. GSM network planning

The GSM network planning is based on radio wave propagation analysis. According to

the transmitting power and antenna configuration of BTS, its coverage area is

calculated. Normally, only the downlink coverage area is calculated, so the GSM

technology does not take uplink coverage area into consideration. The next step is

performed by network planning engineers to analyze the required cell capacity.

According to the calculated cell area, the traffic of cells can be estimated with the help

of electronic map, and then the required channel numbers are calculated through traffic

models (such as Erlang-B or Erlang-C). Next, it is the frequency distribution for BTSs,

and the same frequency can only be reused among enough-distanced cells, to avoid

interference.

To expand the network in the future, the network planning engineer just needs to

distribute new channels to the corresponding cells. As long as there is appropriate

frequency in the overall frequency planning and the expansion does not exceed the

maximum capacity of the BTS, the network does not require other adjustments.

Otherwise, new BTSs or sectors should be added, and new frequency calculation and

channel distribution are required.

2. UTMS network planning

Compared with the GSM network planning, the UMTS network planning features the

following differences:

Cell breathing:-

The CDMA network is totally different from the GSM network. Since channels and

users are not separated for consideration, there is tight relation between coverage and

capacity. The cell with more traffic has less coverage area. In the CDMA network, more

traffic means more interferences. This kind of effect of dynamic change of cell area is

called as “cell breathing”. This can be illustrated through the following visual example.

In a birthday party of a friend, many guests come. More people taking will cause it is

harder to hear the voice of opposite party clearly. If in the beginning, you can talk with a

friend at another end of the room, you can not hear what he is saying at all when the

room is very noisy. It indicates that the “cell radius” of talking area is shortened. The

UMTS network planning engineer faces a network changing dynamically.

In UMTS network planning, the network expandability should be taken into

consideration first of all. The network planning engineer can not simply add frequency

to the related cell as planning GSM network. In the beginning of network planning, a

determined traffic signal redundancy should be taken into consideration, and this

redundancy will be used as “compensation” to the interference caused by increased

traffic. This shows that, from the very beginning, it is required to construct the network

with smaller cells or more NodeBs, resulting in higher investment cost. If the traffic

signal surplus is too small, there is only one way: adding NodeBs when expanding.

The network planning engineer should notice the above issues, because enhancing

transmitting power simply can not reduce the receiving signal deterioration caused by

traffic incensement. Enhancing transmitting power can only improve the receiving

signal of a cell but will add interference to the adjacent cells. As a result, the whole

network communication quality will be influenced.

To enhance transmitting power, the valid range or capacity (for the CDMA network, they

are homonymous) of the CDMA cell is limited. When the UMTS network transmitting

power is doubled, the cell capacity is increased by 10%. The enhanced transmitting

power raises the valid range of cell, but to satisfy the requirements of remote mobile

subscribers, it is necessary to enhance transmitting power by multiple times, which will

influence the talking quality of other mobile subscribers. Let’s return to the above party

example. You can enhance your voice to continue the conversation with the friend at

another end of the room, but at the same time, other guests also raise voice to talk with

others. As a result, the whole room is submerged in noise.

The corresponding relationship between transmitting power and cell capacity is gradual.

Since the UMTS cell load is subject to saturate, the UMTS network planning engineer

must reduce the full-load ratio. The detailed parameters depend on different services

and how much the network carrier would like to risk. Usually, the full-load factor is

preset as 60% in network design. Here, the “cell breathing” effect is used. The adjacent

cells can mutually compensate load, called as soft load. Due to cost, the network

capacity can not be increased on a large scale. The mathematical demonstration on

the UMTS service with large scale data transmission shows that, as data transmission

volume increases, it is more possible for a service cell to borrow capacity from the

adjacent cells. This result is satisfied.

Near-far effect:-

Another typical issue of the CDMA network is near-far effect. Since all the subscribers

in the same cell share the same frequency, it is important that each subscriber in the

whole system transmits signal at the minimum power. Let’s return to the above party

example. If someone is shouting in the room, the conversation among all the other

guests will be influenced. In the CDMA network, this problem can be solved through

power control. For example, the UMTS network uses closed loop power control at

frequency 1500 Hz. For GSM network, however, the power control works at 2 Hz and is

for uplink only.

This kind of fast power control mechanism has been implemented in the UMTS

hardware. But the network planning engineer will face the other case of this problem.

When a subscriber is far away from the NodeB, he needs a majority of transmitting

power, resulting in power shortage for other subscribers. This means that the cell

capacity is related to the actual subscriber geographical distribution. When subscriber

density is rather large, statistics average value can be used to solve this problem.

When subscriber quantity is small, it is necessary to perform dynamic analysis to the

network through simulations.

Uplink and downlink:-

The UMTS network traffic is asymmetrical, that is, the data transmission quantity in the

network uplink differs from that in downlink. The network planning engineer should

calculate values in two directions and then combine them together properly. In this way,

the network planning work will be very complicated. Uplink is a typical limit factor for the

valid coverage range of UMTS cell, or we can say that uplink is coverage-limited and

downlink is capacity-limited. The transmitting power in uplink is provided by UE and the

one in downlink is provided by NodeB.

The above problems also occur in the existing CDMA network. For the UMTS network,

these problems are more complicated. The UMTS network can satisfy different

services with different requirements in communication quality and traffic at the same

time, including simple voice service and packet service up to 2Mbps.

Integrated services:-

In fact, the UMTS network should satisfy requirements of different services at the same

time. So, the network planning engineer should take different services into

consideration. For the service with low communication quality requirement, the UMTS

cell has rather large coverage. For the service with high communication quality

requirement, the cell has very small coverage. So, in the actual work, the network

planning engineer should not just consider UMTS cell radius, because different

Basic Principles of WCDMA System services correspond to different cell radiuses. If the minimum cell radius, that is, the service with highest communication quality requirement, is used as the standard for

network planning, the network establishment cost will be very high and it is not realistic.

The UMTS network planning engineer should start from the cell radius of middle-class

service. As a result, the actual valid range of the cell can only satisfy the requirement of

high-class service partially. At present, many network planning software companies

have started to develop valid algorithms for this kind of new UMTS network integrated

service.

Other differences:-

Compared with GSM network, the UMTS network features other differences. The GSM

network solves capacity problem with sector partition method. The cell with too much

traffic is divided into multiple sectors, and antennas are added correspondingly. This

method is also used for the UMTS network, but its effect is not enough. The change of

cell coverage will cause the near-far effect mentioned above, and overlapped sectors

will interfere mutually because they use the same frequency.

The declination angle (mechanical or electronic) of antenna plays an important role in

the UMTS network. It can reduce the interference among adjacent cells and raise cell

capacity implicitly. In the actual application, large declination angle can be chosen to

solve this problem.

In the WCDMA system, multi-path propagation is not a negative factor but an ideal

result, because receiver can combine the signal with delay of at least 1 Chip (the UMTS

network chip transmission rate is 3.84 Mcps, that is, 1 Chip=0.26 microsecond, equal to

78 meters) into valid signal.

In addition, the UMTS network also uses the soft handover. In this case, a mobile

subscriber can connect to several cells. This method solves network signal fluctuation,

but raises network traffic. The traditional Erlang model is not applicable any more.

Compared with 2G traditional GSM network, the UMTS network features many

differences. Especially, the UMTS network can run asynchronously, which causes

“non-orthogonality” of transmission channel. Let’s return to the party example again.

Even if the perfect planning can be made theoretically, that is, planning the person to

talk in the certain time, it is impossible to reach that ideal goal factually, because the

watches of all the guests can not be synchronous exactly.

Through the above analysis, we can clearly see that the UMTS network planning needs

more cost, compared with the current mobile communication network planning. The

UMTS network planning is rather complicated, because many system parameters are

closely related to each other and should be calculated at the same time. In the current

mobile communication network, however, these parameters are calculated separately.

3G Network Planning Procedure

Compared with the second generation mobile communication, it is difficult to forecast

different service models in the third generation system network owing to introduction of

several kinds of high bit rate services. As for radio network planning, in any case, it is

required to calculate the link budget, capacity and number of NodeBs, estimate the

coverage of base station and design the parameters. In addition, it is required to design

the whole network, calculate the number of channel units, capacity of transmit lines,

RNCs, MSCs and other units in a base station.

In network planning, performance measurements (such as dropped call rate and GOS)

should be introduced to measure the network performance. High-bit rate services are

provided at the cell area where base stations are covered equably, while low-bit rate

services are provided to the edge of cell. The coverage can be designed as continuous

coverage or hotspot coverage. You should estimate coverage of base station carefully

according to different services and different implemented policies. Radio network

planning can be divided into the following phases:

ü  Preparation Phase

1) Confirming Coverage Object

2) Confirming Capacity Object

3) Confirming Coverage Policy

ü  Estimation Phase

1) Estimating Cell Service Amount

2) Estimating Cell Capacity

3) Estimating Coverage Area

4) Calculating Capacity and Link budget

ü  Design and Adjustment Phase

5) Wireless Coverage Optimization Adjustment

6) Control Channel Power Design

7) Pilot Design

8) Soft Handover Parameter Design

9) PN Offset Handover

After the above phases, you get to know the radio network features and confirm control

channel allocation and design handover parameters, and then you may analyze the

coverage of base station in detail. As for some cell, inter-cell interference vs. total

interference ratio is unique. During the process of planning, you can continually take an

analysis to the network and evaluate the interference ratio to estimate the coverage in

different cells. Such iterative process may be repeated until the convergence of

coverage is achieved. Design tools can be adopted to realize the process automation

and in the meantime detect leaks in the coverage. Usually, the coverage in 3G network

service is not equable, which would lower network performance. On the one hand, the

interference in the service-intensive area gets more, resulting in low quality. On the

other hand, it is not necessary to get high quality, for it will result in a waste of resources.

System effectiveness can be improved with the method of self-adapting controlling cell

radius, antenna direction and uplink received power threshold. Cell radius can be

varied with pilot power adjustment. If signal-interference ratio (SIR) is higher than the

required value, cell radius can be increased. If not, cell radius can be decreased. The

uplink/downlink cell radius can be balanced by increasing/decreasing uplink received

power threshold. In the configuration of separate sectors, the communication quality of

the base station can be balanced by changing central angle of each sector.

3G Radio Network Antenna

Introduction

In 3G system (including WCDMA and cdma2000), which is as a new generation mobile

communication system, multi-access mode has changed from TDMA/FDMA to

CDMA/FDMA. However, as far as the wireless signal is concerned, it’s still facing the

difficulties in making use of frequency resources efficiently and decreasing network

interferences and transforming electrical signals with the utmost of efficiency.

A antenna is a communication system bridge between a user terminal and a base

station control equipment, widely applying to cellular mobile communication system. As

the communication technology is developing, antennas will be on progress

consequentially. The mobile communication system in the seventies adopted

omnidirectional antennas or angle reflector antennas, for the reason that a few carriers

and base stations can meet the demands of few users in a mobile communication

system of a city. As the economy goes forward, the amount of mobile terminals, whose

demands can not be met by the old base stations, is boosting so rapidly. Especially as

the development of digital cellular technology goes, new antennas are required to be

configured to improve the multi-path fading, site assignment and multi-channel network

in metropolis.

Plate type antennas are widely applying to 2G digital cellular system thanks to the

features of low section, light framework, easy setting and good electrical specifications.

From the middle 80’s to the late 90’s, vertical polarization (VP) antennas are usually

adopted. A cell is usually divided into 3 sectors, each of which demands 3 antennas, so

9 antennas should be set in a cell. However, too many antennas will result in many

problems, such as high setting cost. In addition, the optimum diversity reception gains

are unable to be achieved with diversity reception antennas set, saying nothing of that

the antennas are unable to be set in some base stations. In that case, the technology of

dual polarized antennas emerges as the times require.

In 3G phase, as the wireless technology grows and the signal detection varies, the

cellular network should be adjusted and optimized, which demands new base station

antennas, such as self-adapting control antenna and intellectualized antenna.

3G Network Structure

3G system, a board-band CDMA system, inherits the features of narrowband CDMA

system in the aspect of network structure. Thanks to code division multiple access,

frequency reuse is not a big problem any longer. The network interference sources

from itself, related to the amount of calling users at the same time. In a metropolis,

three-sector base stations are commonly configured. In suburb, town and road,

three-sector base stations or omni-directional base stations are configured as required

and in arterial traffic, two-sector base stations are commonly configured, as shown in

Figure.

3G Radio Network Structure

3G Antenna's

For 3G (WCDMA), the channels per carrier are decided together by OVSF code and

scrambling code, so capacity of the channels per carrier is great. You should set the

number of the carriers and traffic channels in each base station based on the

requirements of practical traffic distribution in engineering design.

When implementing multi-carrier, you should pay attention to the followings in

designing radio network.

1) Optimize hard handover to minimize the possibility of call drops.

2) Avoid isolated multi-carrier base station and implement multi-carrier in central

cells to avoid hard handover.

3) Avoid heavy-traffic cells of being edge cells where hard handover occurs

3G Radio Network Typical Antennas

You have several options to choose 3G typical antenna base on the following

principles:

1) Properly choosing half-power beam width and gain of antenna based on the

number of base station sectors, traffic density and coverage requirements.

2) Adopting duplexer to save antenna locations.

3) Adopting dual polarization antennas in the dense urban areas.

The adjustment of antenna direction is the same as that in 2G engineering. The main

lobe direction and angle of tilt of the directional antennas should be properly adjusted to

the traffic distribution and communication quality requirements. When setting antennas,

you should note that the isolation between antennas should meet the requirements of

horizontal and vertical isolation to avoid interference. The setting height of antenna is

up to the coverage. Therefore, it should be properly considered according to the

coverage, interference, isolation and future development requirements.

The antennas used in 3G network are similar to those used in 2G network, whose

requirements are as follows:

ü  Sector antenna gain: 13-16 dBd

ü  Omnidirectional antenna gain: 9-10 dBd

ü  Sector antenna half-power beam width: 60-65 degrees or 90 degrees

ü  Omnidirectional antenna deviation in roundness: < +/- 1 dB

ü  Voltage standing wave ratio (VSWR): <1.5

ü  Impedance: 50 ohm (unbalanced type)

ü  Maximum input: >500W

Antenna diversity:

Space diversity or polarization diversity is taken as standard

configurations.

3G Smart Antenna (SA)

1. Principles of SA Smart Antenna

By adopting SDMA that signals differ in the direction of transmission path, SA reduces

the effects of time delay spread, Rayleigh fading, multi-path, channel interference,

distinguishes the signals from the same frequency and timeslot, and combines with

other multiple access technologies to maximize the use of frequency spectrum

resources.

The SA in base stations is a kind of array antenna comprising multiple antenna cells. By

adjusting the weight scope and phase of each signal and changing array pattern, it can

cancel interferences and increase S/N. In addition, it can measure a user’s direction so

that a beam is directed to the user.

"SA" Pattern

From the Figure, it can be shown that the array antenna is composed of N antenna

units, each of which has a corresponding weigher, totaling M groups of weighers and

forming beams in M directions. M indicates the number of users, which can be greater

than the number of antenna units. The dimension of array antenna and the number of

antenna units decide the maximum gain and the minimum beam width, which means

that the dimension of array antenna and antenna gain should keep balance with the

antenna side lobe performance. By adjusting the signal phase and amplitude received

from each antenna, SA combines them to a desired beam. This is called beam forming,

which can form all kinds of beams, such as scanning beam, multi-beam, shaped beam

and the beam with zero position controlled. According to the pattern, there are two

types of SA: self-adapting pattern and shaped pattern.

The key technology of SA is to identify the signal angle of arrival (AOA) and the

implement of digit-shaped. The algorithms to identify the signal AOA are MUSIC

algorithm, ESPRIT algorithm, maximum likelihood algorithm, etc. The implementation

of digit-shaped is to choose the optimum weight coefficient to get the optimum beam.

For self-adapting algorithm, the first step is to set rules, which commonly are maximum

likelihood, maximum S/N, minimum mean square error (MMSE), minimum square error.

You should choose one of them according to the specific conditions. Beam forming SA

pattern is shown in the Figure.

 Beam Forming SA Principle Pattern Shown Below

2. Applications of SA in 3G

The application example of SA in 2G network indicates that SA can efficiently prevent

from interference. According to the 3G criteria, the SA application is required to

improve the network capacity and performance, and take the technical factors, such as

“converged beam”, “self-adapting beam forming” and “beam handover”, into

consideration.

“Converged beam” is applied to special areas, aiming at enlarging coverage and

increasing capacity. Such beam does not associate with a user, nor does it trace mobile

users in the coverage. However, by increasing link scope and converging beams, it can

reduce transmission power of mobile users and according increase capacity. If mobile

user enters the area with great transmission attenuation, the converged beam will point

to the mobile user and rest on him. If a mobile user enters the area with good coverage

which converged beams become unnecessary, the mobile user will be in the charge of

common pilot channel.

“Self-adapting beam forming”, applying to downlink, is in favor of link budget for

individual user and a group of users to improve the system performance. In poor

transmission conditions, such as cell edge (basement), the coverage is required to be

spread to users with an aim to improve link scope.

“Beam handover system” can switch users between narrow beams to form narrow

sectors without handover loss. Because the capacity in 3G system increases as the

number of sectors increases, four 30-degree beams coverage can substitute one

120-degree one, resulting in increasing capacity by 2 to 4 times. Users are switched

between beams without the requirement of any special auxiliary channel.

There are several options to apply SA in 3G. Beam handover SA is an option in starting

phase. In network design, SA can reduce the external network interference (such as

one frequency interference, adjacent frequency interference and other-system

interference) and the internal network interference as well. The order of magnitude

depends on the amount of beams.

3G Handover Design

Introduction

As the mobile station (MS) gets out of the service cell and goes into another service cell,

the link between the former base station and the MS will be substituted for the link

between the new base station and the MS.

Handover, in mobility management, is mostly performed by RRC layer protocol in 3G.

1. Protocol state

The UE state can be classified into IDLE state and CONNECTED state. The IDLE state

can be classified into UTRAN IDLE, GPRS IDLE and GSM IDLE, three of which has

CONNECTED state. The UTRAN CONNECTED state can be classified into four states:

URA-PCH, CELL-PCH, CELL-FACH and CELL-DCH. Handover, generally speaking, is

that UE is transferred from one communication connection to another one in the

CONNECTED state. In this text, handover refers to that UE in CELL-DCH state, unless

otherwise specified.

2. Handover Classification

According to the setup and release of radio link between MS and network, handover

can be classified into softer handover, soft handover and hard handover.

Soft handover refers to that a MS begins to connect with a new base station, the

communication between the mobile station and the former base station is still on. Soft

handover is only applying to the CDMA cells with the same frequency.

The difference between soft handover and softer handover is, softer handover is

performed in the same NodeB where the maximum gain ratio combination of diversity

signals are implemented, while soft handover is performed between two NodeBs,

diversity signals selective combination in RNC.

Hard handover consists of intra-frequency handover, inter-frequency handover and

inter-system handover. Note that soft handover refers to the intra-frequency handover,

but not all intra-frequency handovers are soft handovers. If the target cell and the

former cell have the intra-frequency but belong to different RNC, and there is no Iur

interface between RNCs, the intra-frequency hard handover will occur. Besides, the

internal code handover in the same cell is also hard handover.

Inter-system hard handover consists of the handover between FDD mode and TDD

mode, the handover between WCDMA system and GSM system in R99, and the

handover between WCDMA and cdma2000 in R2000.

The startup compress mode is required to measure inter-frequency and inter-system

for inter-frequency hard handover and inter-system hard handover.

According to the purpose, handover can be classified into edge handover, urgent

handover at poor quality, urgent handover at quick level decrease, interference

handover, velocity sensitivity handover, charge handover, layered/leveled handover,

etc.

The typical process of handover is measurement control → measurement report →

handover decision→ handover implementation → new measurement control.

In the phase of measurement control, the network informs UE of the measurement

parameters through the sent measurement control message. In the phase of

measurement report, the measurement report message is sent to the network by UE. In

the phase of handover decision, the network makes a handover decision according to

the measurement report. In the phase of handover implementation, UE and the

network go along the signalling procedure and make a response to signalling.

Measurement Procedures

In WCDMA system, there are intra-frequency measurements, inter-frequency

measurements, inter-RAT measurements, traffic volume measurements and

UE-internal measurements.

The same type of measurements may be adopted in different functions or processes of

UTRAN, such as cell reselection, handover and power control. The UE shall support a

number of measurements running in parallel. The UE shall also support that each

measurement is controlled and reported independently of every other measurement.

Cells that the UE is monitoring (e.g. for handover measurements) are grouped in the

UE into three different categories:

1.Cells, which belong to the active set. User information is sent from all these cells.

The cells in the active set are involved in soft handover or softer handover.

2.Cells, which are not included in the active set, but are monitored according to a

neighbour list assigned by the UTRAN belong to the monitored set.

3.Cells detected by the UE, which are neither included in the active set nor in the

monitored set belong to the detected set. Reporting of measurements of the detected

set is only required for intra-frequency measurements made by UEs in CELL_DCH

state.

In IDLE mode, the UE shall perform measurements according to the measurement

control information included in System Information Block Type 11 on BCCH. In

CELL-FACH, CELL-PCH or URA-PCH state, the UE shall perform measurements

according to the measurement control information included in System Information

Block Type 12 on BCCH, while in CELL-DCH state, according to the measurement

control information transmitted by UTRAN.

The measurement results will pass two smoothness processing. The first processing is

on the physical layer, with an aim to filter fast fading and report the measurement

results from physical layer to higher layer; The second processing is done before event

evaluation, when the higher layer takes weighted average on the measurement results

reported from the physical layer, based on the time, to confirm the coefficient of filter.

1. UE Measurement

ü  P-CCPCH RSCP

RSCP, Received Signal Code Power, is a measured received power of a code from

P-CCPCH in TDD cell. The reference point of RSCP is the antenna connector at UE.

ü  SIR

S/N is defined as (RSCP/ISCP) × (SF/2).Measuring SIR should be in DPCCH after the

combination of wireless link. The reference point of SIR is the antenna connector at UE.

Where:

RSCP, Received Signal Code Power, is a received power of pilot bit in a code.

ISCP, Interference Signal Code Power, is a received signal interference measured in

pilot bits. Only non-orthogonality of the interference is concerned in measurement.

SF=Spreading Factor.

ü  P-CPICH RSCP

Received Signal Code Power is a code power measured in P-CPICH. The reference

point of RSCP is the antenna connector at UE. If transmission diversity is adopted in ,

the received code power from each antenna should be measured separately, then

added, and sequentially be the power of the whole received codes in P-CPICH.

ü  UTRA carrier RSSI

RSSI = Received Signal Strength Indicator, a broadband received power within relative

channel width. The measurement will be taken at the downlink of UTRAN. The

reference point of RSSI is the antenna connector at UE.

ü   GSM carrier RSSI

RSSI = Received Signal Strength Indicator, a broadband received power within relative

channel width. The measurement will be taken at the BCCH carrier of GSM. The

reference point of RSSI is antenna connector at UE.

ü   CPICH Ec/No

Ec/No refers to ratio of the received energy of each code to noise power density in a

channel. Ec/No has something in common with RSCP/RSSI. The measurement will be

taken at basic CPICH. The reference point of Ec/No is the antenna connector at UE. If

basic CPICH adopts transmission diversity, the received energy of each code (Ec) from

each antenna should be measured separately. The value of adding the received energy

of each code on basic CPICH will be Ec.

ü  BLER of transmission channel c

It refers to the evaluation of Block Error Rate (BLER) of transmission channel. The

evaluation of BLER is based on CRC of each transmission block after the combination

of wireless link. Only the transmission channel with CRC requires the evaluation of

BLER. In the connection mode, BLER can be measured in any transmission channel.

In the idle mode, the BLER on the transmission channel PCH should firstly be

measured if BLER needs measuring.

ü  UE transmitting power

It refers to the transmitting power of the whole UE at a carrier. The reference point of

UE transmitting power is the antenna connector at UE.

ü  In UE, in addition to the measurements above mentioned, there are the

measurements in the aspects of time and order, which would not be described

here.

2. RNC Measurement

ü  RSSI

RSSI, Received Signal Strength Indicator, refers to a broadband received power within

UTRAN uplink carrier channel bandwidth at the access point of UTRAN. The reference

point of RSSI measurement is the antenna connector.

ü  SIR

S/N is defined as (RSCP/ISCP)×SF. Its measurement should be taken in DPCCH after

the combination of wireless link at Node B. In the compression code, SIR should not be

measured at transmitting interval. The reference point of SIR measurement is the

antenna connector.

Where:

RSCP, Received Signal Code Power, refers to a received power in a code.

ISCP, Interference Signal Code Power, refers to a received signal interference. Only

non-orthogonality of interference is concerned in measurement.

SF refers to the spreading factor used in DPCCH.

ü  SIRerror

SIRerror = SIR – SIRtarget_ave, where:

SIR refers to the SIR measured at UTRAN in dB.

SIRtarget_ave = refers to the average value of SIRtarget during a period of time. This period

of time is the same as that when counting the average value of SIRerror. The average

value of SIRtarget is arithmetic average, SIRtarget_ave in dB.

ü  Transmitting carrier power

The transmitting carrier power is the ratio (0…100%) of the whole transmitting power to

the maximum transmitting power, where the whole transmitting power [W] is related to

the average power [W] of a carrier at an access point of UTRAN. The maximum

transmitting power is related to the average transmitting power [W] of a carrier at an

access point of UTRAN under the condition that each cell is on the maximum

power .The measurement may be taken at any transmitting carrier from the access

point of UTRAN .The reference point of measuring transmitting carrier frequency is the

antenna connector. The carrier frequency of each branch should be measured in case

of transmitting diversity.

ü  transmitting code power

Transmitting code power goes under the condition of given carrier, given scrambling

and channel code. You can take measurements at DPCCH of any specific wireless link

from the access point of UTRAN, to show the pilot bit power at DPCCH. All time slots

should be involved in the measurement of transmitting power in the compression mode.

For example, the time slot of transmitting interval should be involved. The reference

point of measuring transmit code power is the antenna connector. The transmitting

code power [W] of each branch should be measured and added in case of transmitting

diversity.

ü  BER of transmission channel

BER of transmission channel is to evaluate the mean bit error rate of DPDCH data after

the combination of wireless link. The BER of transmission channel (TrCH) is as a result

of measuring the puncture bits of channel coding input terminal at Node B. The

evaluation of transmission channel BER may be reported at the end of each TTI at

TrCH. The reported transmission channel BER should be a BER evaluation at the latest

TTI of the current TrCH. Only the BER of transmission channel through channel coding

are needed to be reported.

ü  BER of physical channel

BER of physical channel is to evaluate the mean bit error rate of DPCCH data after the

combination of wireless link at Node B. BER of physical channel may be reported at the

end of each TTI of all sent transmission channels. The reported BER of physical

channel should be a mean BER evaluation at the latest TTI of each transmission

channel.

Other measurements: round trip time, transmission time delay, leading accesses,

etc.

Co-frequency Handover

The WCDMA handover algorithm is briefed as follows. WCDMA soft handover

algorithm adopts Ec/Io of pilot CPICH to be a measurement value of handover, which is

reported to RNC through three-layer signaling.

The following terms are used for describing handover:

Active set: The cells in active set are connected to MS in the form of soft handover.

Neighboring set/Monitoring set: They both list the cells which are measured by MS

continually, but the pilot Ec/Io in these cells are not mature enough to enter active set.

WCDMA handover algorithm

WCDMA soft handover algorithm is briefed as shown in Figure :

WCDMA soft handover algorithm scheme common mechanism

If during the period of ΔT, pilot _Ec/Io> optimum pilot _Ec/Io reports range hysteresis

event 1A, and active set is not full, the cell enters active set, which is called event 1A or

wireless link addition.

If during the period of ΔT, pilot _Ec/Io> optimum pilot _Ec/Io reports range hysteresis

event 1B, the cell will be deleted from active set, which is called event 1B or wireless

link deletion.

If during the period of ΔT, active set is full, optimum alternate pilot _Ec/Io> former worst

pilot _Ec/Io + _Ec/Io + hysteresis event 1C, the strongest alternate cell (the strongest

cell in monitoring set) will substitute for the weakest cell in active set. Such event is

called event 1C or the combination of wireless link addition and deletion. Supposing

there are at most two cells in active set as shown in Figure 5-20.

Where:

Reporting range indicates a threshold value of soft handover;

Hysteresis event 1A indicates adding hysteresis;

Hysteresis event 1B indicates removing hysteresis;

Hysteresis event 1C indicates replacing hysteresis;

ΔT indicates trigger time;

Optimum pilot _Ec/Io indicates the highest value of cell measurement in active set;

Former worst pilot _Ec/Io indicates the highest value of cell measurement in active set;

Optimum alternate pilot _Ec/Io indicates the highest value of cell measurement in

monitor set;

Pilot _Ec/Io is a measured and filtered value.

Handover between WCDMA System and GSM System

The WCDMA criteria and GSM criteria support the bidirectional handover between

WCDMA and GSM. As a result of coverage and load balancing, such handovers are

used. In the early WCDMA configuration, it is necessary to hand off to GSM system for

continuous coverage, and the handover from GSM to WCDMA can be used to reduce

the load in GSM cell. Due to load, the bidirectional handover is significant as the service

of WCDMA network grows. The inter-system handover is triggered by source

RNC/BSC. From the point view of received system, the inter-system handover is

similar to the inter-RNC handover or the inter-BSC handover.

1. Compression mode

If WCDMA adopts a way of continuous sending and receiving, but no WCDMA signal

gap is generated, MS can not take inter-system measurement through one receiver. In

this regard, the compression mode is significant for the inter-frequency measurements

and the inter-system measurements.

The introduction of compression mode is to take alien frequency measurement or

alien-system measurement at FDD. The reason for that is a set of transceiver can only

work on a group of transceiver frequency at the same time. If you want to measure the

signals with other frequencies, you should power off the transceiver and hand off the

frequency to target frequency for measurement. To ensure the normal transmission of

downlink signals, the former signals should be transmitted during the transmission time

left, which is called downlink compression mode. As the measurement frequency is

close to the uplink transmitting frequency, to ensure good measurement, the uplink

signal transmit should be stopped at the same time, which is called uplink compression

mode.

The compression mode is shown in the following figure.

Compression Mode Diagrammatic Sketch

Fast power control can not be used during the period of compression mode gap, so

some interleaved gain will be lost; In this regard, during the compression frame, higher

Eb/No is demanded to decrease capacity. The process of typical inter-system handover

is as follows:

The inter-system handover trigger is implemented at RNC, for example MS is out of

WCDMA coverage range;

RNC commands MS to begin with the inter-system measurements in compression

mode;

RNC chooses the target GSM cell based on the MS measurements;

RNC sends a handover command to MS.

The handover from GSM system to WCDMA system sources from the BSC of GSM.

Due to discrete transmission and receiving, the compression mode is not required for

the measurement value of WCDMA from GSM.

Inter-frequency Handover in WCDMA

Most of UMTS operators have 2 ~ 3 available FDD carrier, where: one frequency is

enough to operate, while the other ones will be used to meet the demands of increasing

capacity. We can adopt two different ways on how to use the frequencies: For the site

with high capacities, several frequencies can be used at the same site, or, different

frequencies can be used in macro-cell layer and micro-cell layer. These schemes

should be supported for the inter-frequency handover at WCDMA carrier.

The same as the inter-system handover, inter-frequency handover requires the

compression mode measurement.

Handover Design

The soft handover design consists of the configuration of the soft handover-related

parameters and the control of soft handover rate. As WCDMA adopts soft

handover-related threshold, it ensures a relatively stable configuration of the

parameters such as the threshold. But the control of the handover rate is identical with

IS-95, which is about 30% to 40%, because too much soft handover will not only

increase the cost on the radio resources, but also reduce the capacity of the down link

when the soft handover is increased to a certain degree.

On the down links, the system interference will be increased along with the increase of

the soft handover link. In case that the system interference exceeds the diversity gain

of the soft handover, the soft handover will bring no benefit to the system capacity. In

this regards, a well-prepared design is demanded before performing the soft handover

in WCDMA. We can keep the soft handover rate in a suitable range by providing

enough diversity in the up/down link.

Parameters with regard to the network performance:

Reporting Range: It is used to set the events 1a and 1b, namely the parameter R in

formulas 1a-1 and 1a-2, 1b-1 and 1b-2. The bigger the R value is, the wider the soft

handover area is. That is because the bigger the R value is, the easier it is to access

ACTIVE SET.

W, which is used to calculate the cell quality of the active set is the value adopted for

different cells. You will use it when you calculate the formulas 1a-1, 1a-2, 1b-1 and1b-2.

Hystersis: The magnetic hysteresis value in the event report. Like in GSM, the purpose

of introducing this value is to avoid the Ping-pong effect as possible. If the value is set

too big will result in that the handover may occur difficultly, but if it is set too small, the

Ping-pong effect may not be avoided.

Reporting deactivation threshold: The maximum number of cells in the active set when

the event is effective, is less than the maximum number of cells in the active set by 1. It

is actually used to confirm the maximum number of cells in the active set (only for 1A

event). If the value is set too big, the system interference may exceed the diversity gain

of the soft handover, otherwise. If it is set too small, it may fail to fully use the diversity

gain of soft handover.

Reporting activation threshold: The minimum cell number of the active set when the

event is effective (only for 1C event).

Time to trigger: Try to avoid the impact of fast fading. If this value is too big, the

handover may be delayed while if it is too small, the handover may occur frequently.

Amount of reporting: The maximum amount of reporting after the event report changes

to the cycle report. It is often used together with the Reporting interval.

Reporting interval: Reporting cycle after the event report changes to the cycle report. It

is used together with the amount of reporting. In using it, we should try to avoid

over-adding the signaling flow.

Reporting Cell Status: It is used to indicate the cell composition principle of the

measured result, including the maximum number of reporting cells and the attributes of

the reporting cell.

WCDMA Power Control Planning

Introduction

In the WCDMA system, radio resource management includes power management,

mobility management, load management, channel allocation and reconfiguration, and

AMR mode control. Of which, the power management is a link of great importance.

Power is the ultimate radio resource of the WCDMA system, so the only method to

make fully use of the radio resources is to strictly control the use of power.

In terms of power management, the QoS of a subscriber can be improved by increasing

the transmit power of the subscriber; however, such improvement may result in

deteriorating other subscriber’s receiving quality due to the self-interference feature of

the CDMA system. WCDMA adopts the broadband spreading technology with all

signals sharing the same spectrum and the signal energy of each MS is allocated within

the frequency band, thus for other MSs it is a kind of wideband noise. Therefore, the

use of power in the CDMA system is conflicting.

In addition, there are such effects as shadow, multi-path fading and remote loss in the

radio environment. The position of a cellular MS in the cell is random and changes

frequently, so the path loss will fluctuate greatly, especially in the multi-cell DS/CDMA

system, where all the cells adopt the same frequency. Theoretically, the address codes

allocated by different subscribers are orthogonal, but in fact it is hard to guarantee them,

thus causing mutual interference among the channels and serious “near-far effect” and

“corner effect”. Near-far effect occurs in the uplink. If all the subscribers in the cell

transmit signals to the BS with the same power, then the signals of the MS near the BS

are strong while the signals of the MS far from the BS are weak. In such a case, the

weak signals will be masked by the strong signals. Corner effect occurs in the downlink.

When the MS is at the corner of the cell, the interference will be twice more than that in

the vicinity of the cell. When the interference is severe, the communication quality of

the MS will be lowered promptly.

Therefore, on the basis of ensuring QoS for subscribers, how to effectively control

power, how to reduce the transmit power as much as possible, and how to reduce the

system interference and increase the system capacity are the key to WCDMA

technologies. The WCDMA system has such functions as forward power control (i.e.

control of the BS transmit power) and reverse power control (i.e. control of the MS

transmit power}, of which the reverse power control is especially important, because

with it, the system capacity and communication quality may be ensured and the fading

and near-far effect may be avoided to a great extent.

Principles of Power Control Implementation

1. Features of fast power control

The mode of power control implementation in the WCDMA system is greatly different

from that in the GSM system. Fast power control is a very important concept integrated

in the WCDMA system.

The radio propagation environment is severe. In the typical cellular mobile

communication environment, the transmitting signals between the BS and the MS

usually reach respective receivers after many times of reflections, dispersions and

refractions. In this way, it is easy to cause multi-path fading of the signal. Fast fading will

cause great impact on the receiving quality of the slow mobile receiver. In the GSM

system, the MS reports the measurement result every 480 ms, and the frequency of the

power control does not exceed twice per second. Therefore, for the GSM system, the

multi-path fading is counteracted through frequency hopping. For the WCDMA system,

in the uplink the DPCCH will divide a 10 ms radio frame into 15 timeslots, each of which

includes a power control command (TPC_cmd).As the speed of power control is higher

than that of fast fading, the receiving quality of the slow MS is effectively ensured.

In other words, the fast power control brings some gain to the slow MS by avoiding fast

fading. Table 7-1 gives a comparison of the required Eb/Io values and required relevant

transmit power changes for the slow and fast power control in the case of three different

motion conditions.

Changes of slow and fast power control in the case of three different conditions

Another two advantages of fast power control are that it can quickly adjust the power of

the MS to avoid far-near affect to a great extent and the fast adjustment of the power

reduces the interference to other cells and MSs.

2. Power control implementation

In the WCDMA system, power control may be divided into inner loop power control and

outer loop power control.

The inner loop power control is to converge the received SIR to the target SIR by

controlling the transmit power of physical channels. In the WCDMA system, relevant

power adjustment commands are sent out by estimating the received Eb/No (ratio of bit

energy to interference power spectrum density). There is certain mapping relationship

between Eb/No and SIR. For instance, for the 12.2 kbps voice service, the typical value

of Eb/No is 5.0 dB. If the chip rate is 3.84 Mcps, the processing gain will be 10 log10

(3.84M/12.2k) = 25 dB. So, the SIR is -20 dB (= 5 dB-25 dB), that is, the

Carrier-to-Interference Ratio (C/I) is more than –20 dB.

The outer loop control mechanism is to dynamically adjust the SIR target value of the

inner loop control, so as to ensure that the communication quality always meets the

requirements (i.e. the specified FER/BLER/BER value).The outer loop control is

conducted in the RNC. The radio channels are complex, so the power control based

only on the SIR value cannot reflect the real quality of the links. For instance, based on

the same FER, the requirements of static subscribers, low speed subscribers (3 km/H)

and high speed subscribers (50 km/H) for SIR are different. The communication quality

is finally measured with FER/BLER/BER, so it is necessary to dynamically adjust the

SIR target value according to the actual FER/BLER value.

The inner power control may be subdivided into open loop power control and closed

loop power control. The former aims to providing the estimates of the initial transmit

power. It estimates the path loss and the interference level according to the

measurement result, so as to calculate the process of initial transmit power. In the

WCDMA system, the open loop power control is adopted in both the uplink and

downlink.

In the WCDMA-FDD system, the fast fading conditions in the uplink and downlink are

absolutely irrelevant because the frequency spacing between the uplink and the

downlink is large. Therefore, the path loss estimates obtained through the open loop

power control according to the downlink signals are inaccurate for the uplink. The

method to solve this problem is to introduce the fast closed loop power control

mechanism.

The closed loop power control mechanism is to rapidly adjust the power in the

uplink/downlink during the communication period, thus making the link quality

converged to the target SIR. Two algorithms may be adopted for the closed loop uplink

power control in 3GPP protocol. In the two algorithms, the step length of the uplink

power control is 1 dB or 2 dB. In the DPCCH, the step adjustment of the power control

is Δdpcch = Δtpc*TPC_cmd.TPC_cmd is the synthesized TPC command from different

algorithms. The power of DPDCH is set according to the power offset between the

DPDCH and the DPCCH.

Differences between the two modes are: The open loop is not closed. It estimates the

downlink interference according to the uplink interference, or estimates the uplink

interference according to the downlink interference. In comparison, the closed loop is a

closed feedback loop. The initial transmit power of the open loop power control is set by

the RNC (uplink) or the UE (downlink), while the closed loop power control is completed

by Node B with RNC only giving the target SIR value of the inner loop power control.

Basic structure of the closed loop power control mechanism

3. SSDT (Site Selection Diversity Transmission)

In the soft handover, there are two or more BSs in the downlink transmitting signals to

an UE simultaneously, which occupies additional system resources (transmit power),

causing additional interference and reducing the forward capacity. Therefore, careful

selection of the power control algorithm during the soft handover is important to

improve the system capacity. Another algorithm of the power control in the soft

handover is SSDT (site selection diversity transmission), according to which, the BS

with minimum path loss will transmit signals, while other BSs will only receive signals of

the uplink and transmit DPCCH. In this way, the total transmit power and additional

interference may be reduced. SSDT is an optional macro diversity method in the

software handover mode.

The specific SSDT implementation method is as follows: Firstly, the UE selects a cell

from ACTIVE SET as the PRIMARY CELL, and all other cells fall into the NON

PRIMARY CELL. SSDT is to transmit signals from the PRIMARY CELL in the downlink,

so as to reduce the interference resulted from the multi-channel transmission in the soft

handover mode. Secondly, it is required to implement the site address selection

promptly if there is no network intervention, so as to maintain the advantages of the soft

handover. To select PRIMARY CELL, a temporary identity code should be allocated to

each cell. Then, the UE will notify other cells in ACTIVE SET of the identity code of

PRIMARY CELL on a regular basis. NON PRIMARY CELL selected by the UE will turn

off the transmit power, and the identity code of PRIMARY CELL will be transmitted via

the uplink FBI domain of the uplink. The SSDT activation, stopping and ID code

allocation are implemented by the upper signaling.

SSDT is initiated by the network according to ACTIVE SET of the soft handover. Once

it is determined to adopt SSDT, the network will notify the cell and the UE of the

message that SSDT is activated in the period of current soft handover. Otherwise, TPC

will still operate in the usual mode, that is, each cell controls the transmit power

according to the TPC instruction of the uplink. The allocation of temporary identity code

should be implemented by the network and notified to all the cells in ACTIVE SET and

the UE for site address selection.

The UE measures the Received Signal Code Power of Common Pilot Channels (RSCP

of CPICHs) transmitted by the cells within ACTIVE SET on a regular basis to select the

PRIMARY CELL. The cell with highest RSCP of CPICHs is the PRIMARY CELL.

Planning of the Power Control Parameters

In the 3G system, the design criteria of the network planning is based on the SIR

optimization and the activity set management. How to set proper RSCP of CPICHs,

SIR target value of various services and handover area (changes to the activity set

scope), and how to determine the coverage and quality of each service area are the

mandatory tasks of the network planning.

In the WCDMA system, the inner loop power control is implemented by NODE-B. The

inner loop power control makes convergence to the target SIR, which is determined by

the outer loop power control. Therefore, the power control parameters planning is

mainly reflected by the outer loop parameters planning. With the research to and

experiment on the relevant parameter settings of the outer loop power control, the outer

loop power control can satisfy the requirements for the control accuracy and the control

speed.

Specific parameters involving the outer loop power control are as follows:

ü  Time factor of BLER report: The target BLER value divided by the time factor is the

pieces N to be measured;

ü   BLER measurement report parameters;

ü  Maximum pieces to be observed: This parameter is used to control the upper limit

of the pieces N to be measured;

ü  SIR-converged lagging value: Check the SIR lagging value (one of measurement

report parameters) converged by SIR;

ü  Control parameters of the SIR measurement report: SIR measures the filter factor

used by the SIRerr;

ü  Uplink outer loop power control parameters: SIR variation range, SIR adjustment

factor, SIR target value falling step length, SIR maximum falling step length;

ü  Uplink soft capacity control parameters: voice quality level and corresponding

BLER values;;

ü  Default CPICH power downlink power balancing parameters: Trigger/Stop the

threshold of DPB process, adjustment period and proportion of the downlink

power balancing

ü  Downlink outer loop power control parameters: Trigger and stop the thresholds of

downlink outer loop power control;

ü  Inner loop power control parameters: Initial SIR value, adjustment step length,

algorithm mode selection;

The above parameters are provided by OM, and there is a close link among them.

WCDMA Radio Network Structure and Resource Planning

Basic Network Structure

1. Network structure

The basic network structure of WCDMA has been described in the prior chapter. It is

divided into core network and access network. This chapter introduces the structure

features of UTRAN and some key technologies and network parameters affecting the

radio network structure from the perspective of network planning.

Types of areas and the relationships among them

1) Types of areas include:

􀁺 Location Areas;

􀁺 Routing Areas;

􀁺 UTRAN Registration Areas;

􀁺 Cell Areas.

2) Relationships among the areas are shown as Figure :

Relationships among the areas

The classification of location areas and routing areas in the WCDMA system is similar

to that in the GSM and GRPS systems.

2. Cell structure

The Node B of Huawei WCDMA system supports omnidirectional, 31, 32, 34, 61, 62

and 64 (combined cabinet) cell configurations. The cell structure of WCDMA is similar

to that of GSM with different titles. “Cell” is similar to the “BS” in the GSM, while

“SECTOR” is equal to the cell in the GSM system.

For instance: What are 3*1, 3*2, 3*4, 6*1, 6*2 and 6*4?

The first digit is the number of sectors supported by each cell; the second digit is the

number of carrier frequencies supported by each sector. 3*1 means the BS supports 3

sectors, each of which has one carrier frequency; 6*4 means the BS supports 6 sectors,

each of which has 4 carrier frequencies.

The cell structure planning is to evenly provide high bit rate within the cell area, or the

data rate of the cell boundary may be lower than that of the area near the BS, thus the

cell area will be larger.

The number of cells is calculated according to the capacity and the link budget. A

network may be coverage-limited or capacity-limited. Capacity limited means the

maximum cell radius cannot support the total traffic flow. Then, the number of cells may

be calculated according to the number of subscribers supported by the cell per sq.km.

Coverage limited means there is enough capacity in the cell to support all the traffic flow.

Then, the number of required BSs may be calculated according to the maximum cell

area.

3. Hierarchical structure of the air interface

According to the protocol, the air interface may functionally form a hierarchical structure.

From bottom to top, they are physical layer, link layer and network layer. The physical

layer fulfills the coding, modulation and spread spectrum of the physical channel. The

link layer may be subdivided into two sub-layers: Medium Access Control (MAC) and

Link Access Control (LAC).The former determines the resources provided by the

physical layer, while the latter completes the establishment, maintenance and release

of the logic link connection. The network layer includes such functions as call control,

mobility management and radio resource management.

4. Channel allocation and reconfiguration

The channel allocation includes the following types:

Connection-oriented channel configuration:

Fundamental Channel Configuration

(FRC) and Dynamic Channel Reconfiguration (DCCC)

Cell-oriented channel configuration:

Cell code resource allocation, cell channel

resource allocation and uplink scramble allocation

Of which:

Fundamental channel configuration: Allocate the channel types and bandwidth

according to the service request; and configure the parameters of each layer of the

channel according to the QoS.

Dynamic channel configuration: During the communication, dynamically change the

channel configurations according to current service status, including the channel types

and the parameters of each layer of the channel

Cell channel resource allocation: Common channel is the resource in the cell, including

RACH, FACH, DSCH and CPCH.

Cell code resource management: Cell downlink code resource allocation policy and

code resource maintenance.

Uplink scramble allocation: The uplink scramble includes the scramble reserved for the

common channels RACH and CPCH, and the scramble allocated to the UE with

dedicated channel.

If RRC is connected or RAB sets up the request, the fundamental channel configuration

entity determines the channel type according to the service type and rate requirements,

and configures parameters of each layer of the channel according to the QoS. It makes

a request to the call admission control entity for admission control according to the

channel configuration parameter (QoS).If it is permitted, then go ahead; or the process

of the channel setup is failed.

The dedicated channel is used to allocate the uplink scrambles; while the cell code

resource maintenance entity is used to allocate the downlink channel code. The

common channel is used to allocate the common channel parameters.

If the channel is set up successfully, the dynamic channel configuration entity will

monitor the traffic flow for the specific service, and make dynamic adjustment to the

channel parameters.

The cell channel resources are allocated according to the current cell service and the

load conditions, so as to adjust the resource configuration of the cell common channel

and optimize the system performances. The cell code resource management entity is

used to maintain the cell code resources.

Hierarchical Network Structure

1. Basic concept of the hierarchical network structure

Similar to the GSM system, the cell may be divided into macro cell (umbellate cell),

micro cell and pico cell according to the features and scope of the cell services. The

macro cell, micro cell and pico cell form a hierarchical network structure (HCS).

The third generation mobile communication system should be able to support the

services with wide coverage in various radio operation environments. The cell type

varies with the requirements: Continuous coverage should be ensured for the large cell,

while the small cell needs high spectral efficiency and capacity. The cell with small

coverage is used to the terminal with low mobility and high capacity, while the cell with

large coverage is used to the terminal with high mobility and low capacity. In addition,

cells should be able to operate on other cells of different types. The coverage area of

micro cell is hundreds meters, while the coverage area of macro is one or just over one

kilometers. In the rural area, it can provide the micro cell with continuous coverage and

the fast mobile subscribers with services. The pico cell covers an indoor scope with the

radius of several meters. The satellite cell provides global continuous coverage. The

traffic should be based on the minimum available cell.

There are two methods to design multi-layer cell in the CDMA system: Different

hierarchical cells operating on the same band, or different hierarchical cells operating

on the different bands. The multi-layer structure also can be applied to the

multi-operator environment.

2. Micro cell and macro cell with the same frequency

The frequency reusability factor is 1. The processing gain of the system enables

subscribers to bear the interference from the cells of different layers. The intra-layer

interference is controlled by the power control, while the inter-layer interference is

controlled by the spatial isolation. Generally, the attenuation of the micro cell is larger

than that of the macro cell, because its antenna is lower. Soft handover can offset the

attenuation valley of the boundary of micro cell.

3. Micro cell and macro cell with different frequencies

It is easy to manage when the cells of different layers adopt different frequencies,

because there is no interference among the layers. The disadvantage of this method is

that it needs large spectrum. At least 15 MHz bandwidth is required if the WCDMA

system is divided into three layers. The impact on the total spectral efficiency from the

non-linear power amplifier depends on the neighboring channel interference and the

link performance deterioration. The increase of the neighboring channel interference

will reduce the spectral efficiency.

Although different carrier frequencies are adopted in the multi-layer cells, interference

will occur between neighboring channel carriers if the capacity is high.

4. Antennas selection and parameters setting of the hierarchical network

Similar to the GSM system, the traffic of the hierarchical network should be deployed as

much as possible to the cell with minimum coverage area; that is, the macro cell is used

to satisfy the requirements of the system for wide coverage, while the micro cell and the

pico cell are used to absorb the traffic and the data traffic.

For this purpose, as to the engineering parameters setting, the antenna and the

transmit power of the macro cell are high; while the antenna and the transmit power of

the micro cell are low; as to the software parameters setting, it is easier for the MS to

access the micro cell and the pico cell. Most of the data traffic is convergent in the pico

cell, so higher QoS of the pico cell should be ensured for higher service rate.

Mobility Management

1. Cell selection and re-selection

1) MS status

According to the protocol, five statuses are available for the MS (UE): IDLE,

CELL_DCH, CELL_FACH, CELL_PCH and URA_PCH.

In the CELL_DCH status, the cell-crossing is judged by the measurement report, and

the location is updated by the handover process.

In the CELL_FACH and CELL_PCH statuses, the cell-crossing is judged by the UE cell

reselection, and the location is updated by CELL UPDATE.

In the URA_PCH status, the cell-crossing is judged by the UE URA, and the location is

updated by URA UPDATE.

2) Mobility management policy in the IDLE status

When the UE starts up, it will carry out PLMN selection, cell selection and location

registration.

Upon the completion of the cell selection, the cell re-selection will be carried out. If a

new cell is selected to stay, the location registration will be carried out for the new

location area entry.

If access and immediate cell evaluation are required, then initiate the access in the

optimized cell.

If the MS is in the CELL_DCH status, the UE crosses the cell through the handover

flow.

3) Potential subscriber control

The stay cell selection by the MS is determined by adjusting the parameters of the cell

reselection, so as to adjust the cell load direction and achieve the load self-adaptive

adjustment.

2. Random access procedure

Random access procedure is a process: A MS requests the access system, then the

network responses and allocates a service channel to the MS. The random access is

carried out when the MS begins to transmit power; or when synchronization loss for

some reasons occurs; or when message packets should be transmitted. The random

access is fulfilled after following steps are completed: 1) synchronization between the

code and frame; 2) search for the cell parameters, such as the random access code; 3)

evaluation of the downlink path loss and random access to the initial power level.

The optimal criterion for the random access procedure is the process rate and the low

transmission power. The requirement for the random access procedure speed is

determined by the requirement for the initial synchronization time. The number of

access channels is depending on the involved access load. In addition, it also will be

affected by the information transmitted in the random access status. Too high

transmission power will reduce the capacity of the CDMA system, and the transmit

power in the random access status cannot be controlled by the fast closed loop power

control, so it is most important to make the total transmit power in the random access

status minimum. If the initial transmission power is lowest, there is a long time for the

access attempt. On the other hand, the high transmission power in the initial access will

cause interference to other subscribers during the fast synchronization. The least

information that needs to be transmitted in the random access attempt is the identity

numbers of the MSs of some types. A kind of typical random access information

includes the pre-field, the synchronization and the data. The data should include at

least the MS identity number, while the pre-field is the unmodulated wideband spread

spectrum signal.

3. Call Admission Control

Call Admission Control is a part of the load management.

The call admission control algorithm is used to accept new calls as many as possible

on the basis of ensuring the existing QoS. Its principle is: Current Status of the Cell

Resources + Service Request ⇒ YES/NO.

The current status of the cell resources is depending on the uplink interference and the

downlink load; while the requested service is depending on the QoS.

4. SNRS migration

The structure shown in the left figure may occur for some reasons, such as the

handover, cell update, URA update, RRC reconnection and direct retry. To save Iur

interface resource and reduce the time delay, it is required to migrate the Iu interface as

shown in Figure 7-8, that is, SNRS migration. The SNRS migration may effectively

reduce the traffic of the Iur interface and improve the adaptability of the system.

SNRS migration

Factors Affecting the Network Structure

1. Universal antenna and smart antenna (narrow-beam antenna)

Smart antenna is widely used in the WCDMA system. In some relevant articles, it is

also called self-adaptive antenna or narrow-beam antenna.

The smart antenna adopts the concept of SDMA, monitoring and extracting the space

information of every subscriber via the self-adaptive array antenna. It separates the

signals of different directions without any interference according to the differences of

the antenna array in the incident signal direction. In fact, it makes the communication

resources are no longer restricted by the time domain, the frequency domain and the

code domain, and extended to the space domain.

The advantages of the smart antenna are: The result of antenna beam forming is equal

to increase the gains of the antenna; the antenna beam forming algorithm may take the

multi-path transmission into consideration, avoiding the multi-path communication from

affecting the digital wireless communication system with the performance enhanced;

the antenna beam forming greatly reduces the multiple access interference. In this way,

the communication capacity can be expanded in multiple. It also can improve the

channel multiplexing rate of the communication system and the BS coverage area, and

solve the increasingly serious interference problems (like common channel and

multi-path fading). In addition, it can optimize the network structure.

2. The transmission models of GSM, CDMA and WCDMA and the affects of radio

transmission on the system structure

From the perspective of signal transmission, in the same frequency band range, the

signals of the GSM, CDMA and WCDMA transmitted in the space have the same

features, including the path loss, the slow and fast fading from the transmitter to the

receiver.

However, the transmission bandwidth in the WCDMA reaches 5M or more, so the

multi-path fading performance is powerful. Its signal frequency band is far larger than

the relevant bandwidth of the channel. The multi-path components may be separated,

making fully use of the multi-path diversity receiving technology.

3. Parameters

The network planning structure is realized by setting proper network engineering

parameters and the network functional parameters. Different antenna gains, antenna

heights, antenna types, network connection parameters, power control parameters,

handover parameters and service rates are set for the cells of different layers of the

hierarchical network.

4. Power control

The power control plays an important role in the CDMA network performance and the

network capacity.

5. Coverage

The maximum cell coverage is determined by the link budget. Besides the data rate

and the Eb/N performance, such specific factors as the cable loss, the antenna gain

and the receiver noise should be calculated. In addition, the affects of the soft handover

gain and asymmetric traffic should be taken into consideration. Different service

coverage area has different service rate requirements. The design basis of the

hierarchical network is as follows:

Outdoor in the rural: Terminal speed 250 kmph, 144 Kpbs at least, 384 Kpbs optimal

Outdoor in the urban or suburb: Terminal speed 150 Kmph, 384 Kpbs at least, 512

Kpbs optimal

Indoor or outdoor with a small area: Terminal speed 10 Km, 2 Mbps at least.

Real-time fixed time delay: BER--, time delay 20-300 ms.

Non-real time variable time delay: BER--, time delay 150 ms.

Erl/km2 may be adopted for in a geometrical area. The data traffic may be Mbps/km2.

The BS adopts multi-subscriber detection technology to provide a favorable coverage

and lower the transmit power of the MS. The increase of the data rate will reduce the

coverage area of the uplink. It differs from the narrowband system.

Radio Resource Planning

1. WCDMA frequency resource

The W-CDMA spectral efficiency is related to the link performance. According to the

theoretical analysis and the emulation, the uplink capacity is 2 to 2.5 times of the

downlink capacity. Besides the antenna diversity in the BS, the uplink adopts the

multi-subscriber signal detection technology, which provides almost two times capacity

than the common receiver. In the downlink, two BSs transmit to the same MS the

signals, which are not orthogonal, but only cause multi-path diversity. For the spectral

efficiency, the bandwidth of 15- to 20-MHz is required for each cell to support an

effective 2-Mbps subscriber.

2. Relationship between the resource planning and the network structure

The WCDMA carrier interval is 200 KHz, ranging from 4.2 MHz to 5.4 MHz. The carrier

intervals are adopted according to the interference to obtain proper protection for the

neighboring channels. The bandwidth of 15 MHz may be divided for the use of three

cells. The interval between different operators may be longer to avoid interference

among them.

3G Network Capacity Estimation

Introduction

In the WCDMA system, the capacity of the uplink is lower that of the downlink, because

the BS has better receiving technologies than the MS, such as the antenna diversity

and multi-subscriber detection. In the UMTS, the downlink capacity is considered more

important than the uplink capacity because the asymmetrical traffic is closely related to

download services. In the 3G, more consideration is given to the downlink capacity. The

factors causing the differences between the capacity uplink and the downlink capacity

are the orthogonal code and the BS transmit diversity. WCDMA system adopts long

extended code to distinguish cells in the downlink and subscribers in the uplink.

Downlink Orthogonal Code

The downlink orthogonal code will affect the capacity, so it is considered to adopt

irregular short code, which is orthogonal in the case of one path. Part of the

orthogonality will disappear in case of multi-path, causing mutual interference among

the subscribers in the cell. In the GSM system, there is no mutual interference in the

same cell, because the time domain is orthogonal.

Link Budget

To estimate the maximum area of the cell, it is required to carry out RLB ( Radio link

budget) calculation. In the RLB, it is necessary to consider such factors as antenna

gain, cable loss, diversity gain and fading margin. The RLB calculation result is the

maximum allowed transmission path loss, based on which the cell radius and the

number of required sites may be obtained. Compared with the TDMA-based radio

access system (like the GSM system), the WCDMA system has some special problems

in the link budget, including interference margin, fast fading margin, transmit power

increasing and soft handover gain.

Capacity and Coverage Analysis

If the maximum allowable path loss of a cell is known, it is easy to calculate the

coverage area of the cell with the known transmission model. If the coverage area of

the cell is known, it is required to select such sites as the channel elements, sector and

carrier frequency and site density (cell radius) to configure, so as to satisfy the traffic

requirements. The cell radius is also closely related to the number of access

subscribers. Therefore, the coverage is correlative with the capacity, and the network

operator should know the subscriber distribution and the growth trend, because it will

directly affect the coverage. The network should be configured properly to meet the

traffic requirements and reduce the network cost as much as possible. The number of

carrier frequencies, number of sectors, cell load, number of subscribers and cell radius

will affect the final result.

Soft Capacity

The number of required cells may be calculated with the available spectrum,

subscribers amount forecast and traffic density information. The traffic density is

indicated by erl. The calculation is conditional on the given congestion. If the

congestion is caused by the hardware, then the result may be obtained in Table B. If the

maximum capacity is caused by the interference, then the capacity is defined as soft

capacity. For the system with soft capacity restriction, it cannot be calculated with

Ireland Table. The total channel capacity is larger than the average number of the

channels of each cell. The neighboring cells share a part of interference, so more traffic

may be processed if the congestions are the same. If the interference from the

neighboring cell is less, there will be more available channels in the middle cell as

shown in Figure

Interference sharing in the WCDMA

If the cell has few channels, that is, there are high bit rate real-time subscribers, then

the average load should be reduced to ensure low congestion. With the reduction of

average load, there is additional capacity to be provided to the neighboring cell for use.

This part of capacity is borrowed from the neighboring cell, so the interference sharing

provides soft capacity. It is important for the high bit rate real-time subscribers to

connect as shown in the figure.

The soft capacity depends on the transmission environment (i.e. network planning).

The value of α is of great importance, and determined by the equipment radio resource

management algorithm.

In the WVDMA system, all subscribers share the interference source in the space

channel, so the analysis cannot be conducted separately. The mutual affects among

subscribers result in the changes of transmit power, which in turn cause further

changes and repeating mutual affects. Forecast processing repeats unless it is stable.

In the WCDMA system, the uplink/downlink fast power control, soft handover/softer

handover and orthogonal downlink channels will affect its performance. Unlike the

GSM, the BS sensitivity is depending on the number of subscribers and the rate of

subscribers. It is constant in the GSM.

The interference planning and capacity planning are even more important in the 3G.

Planning Conclusion

The WCDMA cell capacity (at full load) is inversely proportional to the cell coverage

radius. Under certain I/C, reducing the cell radius can improve the cell capacity.

Essentially, it is to offset the cell capacity loss caused by the system noise. Reducing

the cell capacity can increase the coverage radius. Of course, the coverage area

should be subject to the radio coverage conditions. In the cell planning, it is required to

determine proper coverage radius to meet the cell capacity requirements.

The cell coverage requirements for the capacity in the urban and rural are different. In

the urban, there are many hot areas and the unit area traffic demand is high. In such a

case, to solve the capacity problem is a main task. Whereas, in the rural area, the traffic

demand is low and the main task is to solve the coverage problem. The CDMA system

features soft capacity and can satisfy this requirement.

The unit area capacity may be improved by the cell splitting and multi-sector. Such

improvement may be realized easier in the CDMA cell, which dynamically changes the

cell coverage by controlling the pilot transmit power.