Transcript
White Paper
Long Term Evolution Protocol
Overview
Document Number: LTEPTCLOVWWP
Rev 0
Overview
Long term evolution (LTE) is the next step forward in cellular 3G services. LTE technology is a based on a 3GPP
standard that provides for a downlink speed of up to 150 megabits per second (Mbps) and an uplink speed of up to 50
Mbps. Fixed wireless and wired standards are already approaching or achieving 100 Mbps or faster, and LTE is a way
for cellular communications to operate at that high data rate.
This paper provides an introduction to how the LTE protocol stack operates. Because the final 3GPP specification will
cover tens of thousands of pages, this paper touches only on the highest levels of protocol operation. The paper
discusses the history and application requirements that determine the functions and priorities of LTE, examines the
protocol stack in terms of the time domain and in terms of information moving through the stack, and finally discusses
more specialized aspects of the standard such as scheduling and quality of service, management and control functions,
handovers and power save operation.
Contents
1 Introduction 1
1.1 Protocol Design 1
1.2 History and Scope of
3GPP Standards 1
2 LTE Architecture 2
2.1 How the MAC Sees the PHY 2
2.2 Frames and Packet Timelines:
LTE Downlink 3
3 Life of an LTE Packet 4
3.1 Life of an LTE Packet: Downlink 4
3.1.1 MAC Layer 4
3.1.2 Hybrid ARQ 4
3.1.3 MAC Channels 5
3.1.4 MAC Downlink Mapping 5
3.1.5 MAC Format Selection and
Measurements 6
3.1.6 RLC Layer 6
3.1.7 PDCP Layer 7
3.2 Life of an LTE Packet: Uplink 8
3.2.1 PDCP Layer 9
3.2.2 RLC Layer 9
3.2.3 MAC Layer 9
3.3 Life of a Packet: Conclusion 11
4 LTE Protocol Operation 12
4.1 Scheduling 12
4.1.1 Downlink Scheduling 12
4.1.2 Downlink Scheduling
with HARQ 13
4.1.3 Uplink Scheduling
with HARQ 14
4.2 QoS Architecture 15
4.3 Management and Control Functions 15
4.4 Handover and Roaming 16
4.4.1 Handover Measurement 16
4.4.2 Handover: Neighbor Lists 17
4.5 Power Save Operation 17
4.5.1 DRX and DTX
17
4.5.2 Long and Short DRX
17
5 Conclusion
18
1 Introduction
Long Term Evolution (LTE) is important because it will bring up to a 50x performance improvement and much better
spectral efficiency to cellular networks. LTE is different from other technologies that call themselves 4G because it is
completely integrated into the existing cellular infrastructure for 2G and 3G. This allows seamless handoff and complete
connectivity between previous standards and LTE. LTE is in trials now and should see commercial deployment by
2010.
This paper provides an overview of the MAC for 3GPP™ Long Term Evolution (LTE) also referred to as E-UTRAN, with
a focus on the handset or User Equipment (UE). The protocol stack functions consist of the Medium Access Control
(MAC), Radio Link Control (RLC), Packet Data Convergence Protocol (PDCP), and Radio Resource Control (RRC).
LTE is the latest generation of the 3GPP standards. The LTE standard specifies an IP-only network supporting data
rates up to 150 Mbps. These high data rates will enables new applications and services such as voice over IP,
streaming multimedia, videoconferencing or even a high-speed cellular modem.
1.1 Protocol Design
The LTE standard grew out of the Global System for Mobile Communications (GSM) and Universal Mobile
Telecommunications System (UMTS) standards, commonly called 3G. Voice communication was primary application,
with data added recently. Mobility and seamless handoff were requirements from the start, as was a requirement for
central management of all nodes.
LTE speeds will be equivalent to what today’s user might see at home on a fast cable modem. The LTE standard is
designed to enable 150 Mbps downlink and 50 Mbps uplink over a wide area. While 150 Mbps is LTE’s theoretical top
uplink speed, each user’s bandwidth will depend on how carriers deploy their network and available bandwidth.
Supporting high rates while minimizing power is a key design challenge.
The LTE physical layer is unique because it has asymmetrical modulation and data rates for uplink and downlink. The
standard is designed for full-duplex operation, with simultaneous transmission and reception. The radio is optimized for
performance on the downlink, because the transmitter at the base station has plenty of power. On the uplink, the radio
is optimized more for power consumption than efficiency, because while processing power has increased, mobile
device battery power has stayed essentially constant.
1.2 History and Scope of 3GPP Standards
The standardization process for LTE comes out of the Third Generation Partnership Project (3GPP), which was
developed out of GSM cellular standards. Work on LTE has been going on since 2004, building on the GSM/UMTS
family of standards that dates from 1990. Stage 2 of the standard, the functional descriptions, was completed in 2007.
The standards bodies are currently finalizing Stage 3, which is the detailed specifications. Stage 3 is expected to be
completed by the end of 2008.
The LTE standard was designed as a completely new standard, with new numbering and new documents.it does not
build on the previous series of UMTS standards. Earlier elements were only brought in if there was a compelling reason
for them to exist in the new standard. There is no requirement for backward compatibility or error interoperability, for
example, because LTE will operate in different spectrum using a different physical layer and different coding. However,
The architecture may often appear similar because the standards were created by similar standards bodies.
The entire LTE system is specified by a large number of 3GPP working groups which oversee everything from the air
interface to the protocol stack and the infrastructure network. This paper focuses on protocols specified by RAN2, a
3GPP Radio Area Network working group1. LTE is a departure from historical cellular and telecom operations, which
were circuit switched. LTE is the first GSM/3GPP standard that is fully IP and packet-based. Much of the complexity of
UMTS that deals with circuit switching is not carried into LTE; this has allowed some simplifications and optimizations of
1 For more information about RAN2, visit the 3GPP RAN2 web pages at http://www.3gpp.org/tb/ran/RAN2/RAN2.htm.
Freescale Semiconductor, Inc. Long Term Evolution Protocol Overview
the architecture. LTE provides a packet switched model at the SAP, but retains a circuit switched model at the PHY.
The physical layer itself maintains the continuous connection model, especially on the downlink, where there is
continuous transmission.
2 LTE Architecture
Figure 2.1 provides a high-level view of LTE architecture. This is a snapshot of the part that most closely interacts with
the UE, or mobile device. The entire architecture is much more complex; a complete diagram would show the entire
Internet and other aspects of network connectivity supporting handoffs among 3G, 2G, WiMAX, and other standards.
This particular device shows the eNodeB, which is another name for the base station, and the interfaces between the
eNodeB and UEs. The E-UTRAN is the entire network, which is the “official” standards name for LTE.
Figure 2.1. LTE Architecture Overview
UE
.
eNB: Enhanced Node B, or base
station
.
UE: User Equipment
.
EPC: Evolved Packet Core
o
MME: Mobility Management
Entity (Control Plane)
o
SAE: System Architecture
Evolved (User Plane)
.
E-UTRAN: Evolved Universal
Terrestrial Radio Access Network
2.1 How the MAC Sees the PHY
The LTE PHY is typically full duplex. LTE is designed primarily for full duplex operation in paired spectrum. To contrast,
WiMAX operates in half duplex in unpaired spectrum, where information is transmitted in one direction at a time. LTE
can support TDD operation in unpaired spectrum; however, it is not a primary focus of the design. The PHY operates
continuously for downlink with interspersed sync, providing multiple channels simultaneously with varying modulation.
The downlink channel operates as a continuous stream. Unlike IEEE® 802 family standards, there is no relation
between the air interface.transmitted frames on the air.and the actual service data unit (SDU) packets that are
coming from the top of the protocol stack. LTE uses the concept of a resource block, which is a block of 12 subcarriers
in one slot. A transport block is a group of resource blocks with a common modulation/coding. The physical interface is
a transport block, which corresponds to the data carried in a period of time of the allocation for the particular UE. Each
radio subframe is 1 millisecond (ms) long; each frame is 10 milliseconds. Multiple UEs can be serviced on the downlink
at any particular time in one transport block. The MAC controls what to send in a given time.
The LTE standard specifies these physical channels:
.
Physical broadcast channel (PBCH)
o
The coded BCH transport block is mapped to four subframes within a 40 ms interval
o
40 ms timing is blindly detected, i.e. there is no explicit signaling indicating 40 ms timing
o
Each subframe is assumed to be self-decodable, i.e. the BCH can be decoded from a single reception,
assuming sufficiently good channel conditions
Long Term Evolution Protocol Overview Freescale Semiconductor
.
Physical control format indicator channel (PCFICH)
o
Informs the UE about the number of OFDM symbols used for the PDCCHs
o
Transmitted in every subframe
.
Physical downlink control channel (PDCCH)
o
Informs the UE about the resource allocation of PCH and DL-SCH, and Hybrid ARQ information related to
DL-SCH
o
Carries the uplink scheduling grant
.
Physical Hybrid ARQ Indicator Channel (PHICH): Carries Hybrid ARQ ACK/NAKs in response to uplink
transmissions.
.
Physical downlink shared channel (PDSCH): Carries the DL-SCH and PCH
.
Physical multicast channel (PMCH): Carries the MCH
.
Physical uplink control channel (PUCCH)
o
Carries Hybrid ARQ ACK/NAKs in response to downlink transmission
o
Carries Scheduling Request (SR)
o
Carries CQI reports
.
Physical uplink shared channel (PUSCH): Carries the UL-SCH
.
Physical random access channel (PRACH): Carries the random access preamble
2.2 Frames and Packet Timelines: LTE Downlink
Figure 2.2 shows a time domain view. At the bottom are radio frames. A full frame is 10 ms but we normally think in
terms of the 1-ms subframe, which is the entity that contains the transport block. Within the transport block is the MAC
header and any extra space (padding). Within that there is the RLC header, then within the RLC header there can be a
number of PDCPs. There is a somewhat arbitrary relationship between the IP packets coming in, which form the SDUs,
and how the RLC PDUs are formed. Therefore you can make the maximum effective use of radio resources in a fixed
period of time.
Figure 2.2. Time domain view of the LTE downlink
Each sub-frame
contains 14 OFDM
Symbols
PDCP
Packet Data Convergence
Protocol
RLC
Radio Link Control
MAC
Medium Access Control
PHY
Physical Layer
UE
PDCP SDUs
RLC SDUs
MAC SDUs
PDCP PDUs
RLC PDUs
MAC PDUs
Data
PDCP Hdr
RLC Hdr
n
Data
PDCP Hdr PDCP Hdr
n+1 n+2
Data
RLC Hdr
Transport Block
Padding
IP Packets
MAC Hdr
Freescale Semiconductor, Inc. Long Term Evolution Protocol Overview
3 Life of an LTE Packet
This section traces the flow of a packet through the sub-layers of the LTE stack. The downlink direction (from network
to terminal) is covered first. An uplink packet is then described, highlighting any differences.
3.1 Life of an LTE Packet: Downlink
Stacks are always shown graphically with the higher layers on the top, so the downlink flow progresses upward through
the stack in Figure 3.1. Start by delivering a transport block from the physical layer to the MAC . that contains the
information that was decoded off the air in the previous radio subframe. There can be an arbitrary relationship between
what’s in the transport block and the actual packets that are being delivered to higher layers.
Figure 3.1. Life of an LTE Packet: Downlink
Host
RLC
RRC
MAC
Layer 1
Layer 3
PDCP
MAC PDUs
RLC PDUs
MAC Ctrl
L1 Ctrl
Layer 2
PDCP PDUs
RRC PDUs
PS NAS
RLC Ctrl
User Traffic
Radio Bearers
Logical Channels
Transport Channels
Downlink
Flow
PDCP Ctrl
Physical Channels
PHY (DL OFDM, UL SC-FDMA)
The transport block, delivered from the PHY to the MAC, contains data from the previous radio subframe. It may
contain multiple or partial packets, depending on scheduling and modulation.
3.1.1 MAC Layer
The MAC, represented in the MAC block in Figure 3.1, is responsible for managing the hybrid ARQ function, which is a
transport-block level automatic retry. It also performs the transport as a logical mapping.a function that breaks down
different logical channels out of the transport block for the higher layers. Format selection and measurements provide
information about the network that is needed for managing the entire network to the radio resource control.
3.1.2 Hybrid ARQ
The Hybrid Automatic Repeat-reQuest (HARQ) process, done in combination between the MAC and the PHY,
retransmits transport blocks (TBs) for error recovery. The PHY performs the retention and re-combination (incremental
redundancy) and the MAC performs the management and signaling.
The MAC indicates a NACK when there’s a transport block CRC failure; the PHY usually indicates that failure.
Retransmission is done by the eNodeB or the sender on the downlink using a different type of coding. The coding is
sent and maintained in buffers in the eNodeB. Eventually, after one or two attempts, there will be enough data to
reconstruct the signal. In HARQ operation, the retransmission does not have to be fully correct. It has to be correct
enough that it can be combined mathematically with the previous transport block in order to produce a good transport
Long Term Evolution Protocol Overview Freescale Semiconductor
block. This is the most efficient way of providing this ARQ function. It does operate at the transport block level, there is
another ARQ process mechanism operating at the RLC.
These are the basic steps of the HARQ process:
.
MAC sends “NACK” message when TB fails
CRC
.
Transport blocks with errors are retained
.
PHY retransmits with different puncturing code
.
Retransmission combined with saved transport
block(s)
.
When correct transport block is decoded, MAC
signals “ACK”
.
Multiple HARQ processes can run in parallel to
retry several outstanding TBs
3.1.3 MAC Channels
Figure 3.2. Simplified HARQ operation
Bad Bits
First TB
CRC = Fail
Retry TB
CRC = Fail
HARQ Combined
CRC = Good
Logical channels exist at the top of the MAC. They represent data transfer services offered by the MAC and are defined
by what type of information they carry. Types of logical channels include control channels (for control plane data) and
traffic channels (for user plane data).
Transport channels are in the transport blocks at the bottom of the MAC. They represent data transfer services offered
by the PHY and are defined by how the information is carried, different physical layer modulations and the way they are
encoded.
3.1.4 MAC Downlink Mapping
When a valid transport block is available from the HARQ process, the transport channels are mapped to logical
channels. Figure 3.3 shows the physical layer control channel at the bottom of the picture; working up, it terminates in
the MAC layer. It is used for scheduling, signaling and other low-level functions. The downlink shared channel contains
both a transport channel for paging and for downlink. The physical broadcast channel goes all the way through for
broadcast.
Multicast channels are grayed out in Figure 3.3 because they are not being specified in Release 8 of the LTE standard.
These channels will be re-addressed in Release 9.
Figure 3.3. MAC Downlink Mapping
BCCH PCCH CCCH DCCH DTCH MCCH MTCH
Logical
Channels
MAC
Downlink
Transport
Channels
Downlink
Physical
Terminates
in MAC Layer
PHY
PBCH PDSCH PDCCH PMCH
Channels
aka L1/L2
control channel
BCH PCH DL-SCH MCH
.
Logical Channels
o
PCCH: Paging Control Channel
o
BCCH: Broadcast Control Channel
o
CCCH: Common Control Channel
o
DCCH: Dedicated Control Channel
o
DTCH: Dedicated Traffic Channel
o
MCCH: Multicast Control Channel
o
MTCH: Multicast Traffic Channel
.
Transport Channels
o
PCH: Paging Channel
o
BCH: Broadcast Channel
o
DL-SCH: Downlink Shared Channel
o
MCH: Multicast Channel
Freescale Semiconductor, Inc. Long Term Evolution Protocol Overview
Transport channels use different modulations and coding. Paging and broadcast channels must be received
everywhere in the cell, so they must use robust modulation. The DL-SCH can be optimized by the UE.
Figure 3.3 illustrates the following logical channels:
.
Dedicated Traffic Channel (DTCH): A point-to-point channel, dedicated to one UE, for the transfer of user
information. A DTCH can exist in both uplink and downlink.
.
Broadcast Control Channel (BCCH): A downlink channel for broadcasting system control information.
.
Paging Control Channel (PCCH): A downlink channel that transfers paging information. This channel is used
when the network does not know the location cell of the UE.
.
Common Control Channel (CCCH): Uplink channel for transmitting control information between UEs and
network. This channel is used by the UEs having no RRC connection with the network.
.
Dedicated Control Channel (DCCH): A point-to-point bi-directional channel that transmits dedicated control
information between a UE and the network. Used by UEs that have an RRC connection.
3.1.5 MAC Format Selection and Measurements
The format selection process prepares the physical layer to be ready for the coding and modulation of the next
transport block. On downlink, the MAC in the UE interprets the transport format. The eNodeB includes information in
each transport block that specifies the format using a Modulation Coding Scheme (MCS) for the next transport block;
this can change dynamically. The MAC configures the PHY for the next TB.
The MAC coordinates measurements from the local PHY to the RRC regarding local status and conditions, and the
RRC reports back to the eNodeB using control messages. From the eNodeB to the RRC, the RRC controls local PHY
modulation and configuration settings.
MAC measurements support downlink scheduling. Rates and radio conditions at the UE are used by the eNodeB.
Buffer status reports and other types of signaling are carried back to the higher layers through RRC messaging. If the
rate is high, fewer time slots are needed to send data. Both uplink and downlink are fully scheduled by the eNodeB.
3.1.6 RLC Layer
The RLC layer is illustrated by the RLC block in Figure 3.1 on page 4. It performs segmentation and reassembly and
operates in three modes: transparent mode (TM), acknowledged mode (AM) and unacknowledged mode (UM). These are
used by different radio bearers for different purposes. The RLC provides in-sequence delivery and duplicate detection.
3.1.6.1 RLC Segmentation
The segmentation process involves unpacking an RLC PDU into RLC SDUs, or portions of SDUs. The RLC PDU size is
based on transport block size. RLC PDU size is not fixed because it depends on the conditions of channels which the
eNodeB assigns to every UE on the downlink. Transport block size can vary based on bandwidth requirements,
distance, power levels or modulation scheme. The process also depends on the size of the packets, e.g. large packets
for video or small packets for voice over IP. If an RLC SDU is large, or the available radio data rate is low (resulting in
small transport blocks), the RLC SDU may be split among several RLC PDUs. If the RLC SDU is small, or the available
radio data rate is high, several RLC SDUs may be packed into a single PDU. In many cases, both splitting and packing
may be present. This is illustrated in Figure 3.4.
Long Term Evolution Protocol Overview Freescale Semiconductor
Figure 3.4. RLC Segmentation
Downlink
Flow
3.1.6.2 RLC In-order Delivery
The RLC ensures in-order delivery of SDUs. Out-of-order packets can be delivered during handover. The PDU
sequence number carried by the RLC header is independent of the SDU sequence number (i.e. PDCP sequence
number). An RLC SDU is built from (one or more) RLC PDUs for downlink. Packet order is corrected in the RLC using
sequence numbers in the RLC header.
3.1.6.3 RLC Modes
As mentioned above, the RLC operates in three modes. Transparent mode is used only for control plane signaling for a
few RLC messages during the initial connection. There is effectively no header; it simply passes the message through.
Unacknowledged and acknowledged modes use the RLC header and indicate whether or not the ARQ mechanism is
involved.
ARQ applies to an RLC SDU, while HARQ applies to a transport block. These interactions may contain a partial SDU,
one SDU, or multiple SDUs. ARQ can be used for TCP/IP or critical information; it also might use unacknowledged
mode for voice over IP or when there’s no time for a retry because of latency requirements; in voice over IP, for
example, a packet that does not arrive the first time is useless, and the higher layers make up for the difference. ARQ,
unlike the HARQ, applies to the SDU at the top of the RLC. If the HARQ transmitter detects a failed delivery of a TB.
for example, maximum retransmission limit is reached.the relevant transmitting ARQ entities are notified and potential
retransmissions and re-segmentation can be initiated at the RLC layer for any number of affected PDUs.
3.1.7 PDCP Layer
The Packet Data Convergence Protocol (PDCP) layer is illustrated by the PDCP block in Figure 3.1 on page 4. In
earlier versions of GSM and 3GPP standards, the PDCP was only used for the packet data bearers, and the circuit-
switched bearers connected directly from the host to the RLC layer. Because LTE is all-packet, this is now a place for
higher-layer functions that sit above the packing and unpacking that goes on in the RLC.
PDCP functions in the user plane include decryption, ROHC header decompression, sequence numbering and
duplicate removal. PDCP functions in the control plane include decryption, integrity protection, sequence numbering
and duplicate removal. There is one PDCP instance per radio bearer. The radio bearer is similar to a logical channel for
user control data.
3.1.7.1 PDCP Header Compression
Header compression is important because VoIP is a critical application for LTE. Because there is no more circuit
switching in LTE, all voice signals must be carried over IP and there is a need for efficiency. Various standards are
being specified for use in profiles for robust header compression (ROHC), which provides a tremendous savings in the
amount of header that would otherwise have to go over the air. These protocols are designed to work with the packet
loss that is typical in wireless networks with higher error rates and longer round trip time. ROHC is defined in IETF RFC
3095, RFC4815, and RFC 3843.
Freescale Semiconductor, Inc. Long Term Evolution Protocol Overview
Figure 3.5. PDCP Header Compression
Payload size for VoIP
e.g. G.723.1 codec
IP Header Payload
Token Payload
40 bytes
1 -4 bytes
24 bytes
3.1.7.2 Ciphering and Integrity Protection
Ciphering, both encryption and decryption, also occurs in the PDCP. Security has to occur below the ROHC because
the ROHC can only operate on unencrypted packets. It cannot understand an encrypted header. Ciphering protects
user plane data, radio resource control (RRC) data and non access stratum (NAS) data. Processing
order in the PDCP is as follows: For the downlink, decryption occurs first, then ROHC Higher layers
decompression. For the uplink, ROHC compression occurs first, then encryption.
Details of LTE security architecture are still being defined. The 3GPP System Architecture Working
Group 3 (SA3) is responsible for security and has decided to use either Advanced Encryption
Standard (AES) or SNOW 3G algorithms. Specific modes for AES are still being determined; AES is
a block cipher and has to use specific operational modes to operate in a streaming mode.
3.2 Life of an LTE Packet: Uplink
LTE processes on the uplink side are often similar to processes on the downlink side. Key differences are that the peak
data rate is half that of downlink; access is granted by the eNodeB; there are changes in logical channels and transport
channels; and random access is used for initial transmissions. The PHY uses SC-FDMA for the uplink because it has a
lower peak-average ratio, which allows a more power-efficient transmitter in the UE.
Figure 3.6. Life of an LTE Packet: Uplink
ROHC
Security
UL DL
Host
RLC
RRC
MAC
Layer 1
Layer 3
PDCP
MAC PDUs
RLC PDUs
PDCP Ctrl
MAC Ctrl
L1 Ctrl
Layer 2
PDCP PDUs
PS NAS
RLC Ctrl
User Traffic
Radio Bearers
Logical Channels
Transport Channels
Uplink
Flow
RRC PDUs
Physical Channels
PHY (DL OFDM, UL SC-FDMA)
Long Term Evolution Protocol Overview Freescale Semiconductor
3.2.1 PDCP Layer
PDCP functions are symmetrical for the uplink and the downlink. The PDCP block in Figure 3.6 illustrates the process.
The functions are the same for the header compression and encryption, but they occur in reverse order. Uplink
processing includes header compression and encryption.
3.2.2 RLC Layer
The RLC functions, shown in the RLC block in Figure 3.6, are also symmetrical for the uplink and the downlink. Instead
of removing the headers, now headers are applied. There is still a need to support transparent mode, unacknowledged
mode and acknowledged mode. The uplink process concatenates rather than segments the SDUs into transport blocks.
Segmentation is only done when it is needed to fit SDUs into a transport block.
3.2.2.1 Concatenation
Concatenation is the process of packing an RLC SDU into a size appropriate for transport blocks. The RLC PDU size is
chosen based on the transport block size for the radio bearer. If the RLC SDU is large, or the available radio data rate
is low, the RLC SDU may be split into several RLC PDUs. If the RLC SDU is small, or the available radio data rate is
high, several RLC SDUs may be packed into a single PDU. In many cases both splitting and packing occur. This is
shown in Figure 3.7.
Figure 3.7. RLC Concatenation in the Uplink
Uplink
Flow
3.2.3 MAC Layer
MAC functions in the uplink, illustrated in the MAC block in Figure 3.6, are significantly different in the uplink and
downlink. Uplink functions include random access channel, scheduling, building headers and transport format selection.
3.2.3.1 Transport Format Selection
The MAC determines the transport format.how to use it, how to pack the information in, and what modulation and
coding are available and configure the PHY appropriately to be ready to transmit. The Uplink Shared Channel (ULSCH)
is the primary transport channel. Format variables are modulation and coding, which determine data rate. The
MAC determines the capacity of a transport block based on the transport format.
3.2.3.2 MAC Uplink Channel Mapping
The Common Control Channel (CCCH), Dedicated Control Channel (DCCH) and Dedicated Traffic Channel (DTCH)
are all mapped to the UL-SCH.
All MAC transmissions on the UL-SCH must be scheduled by the Random Access Channel (RACH) procedure. When
the UE is not connected, no transmit slots are ever scheduled. The RACH provides a means for disconnected devices
to transmit. Transmitting on the UL-SCH requires a resource allocation from the eNodeB, and time alignment to be
current. Otherwise the RACH procedure is required.
Freescale Semiconductor, Inc. Long Term Evolution Protocol Overview
Figure 3.8. MAC Uplink Channel Mapping
CCCH DCCH DTCH
RACH UL-SCH
PRACH PUSCH PUCCH
UplinkLogicalChannelsUplinkTransportChannelsUplinkPhysicalChannelsGenerated
in MAC Layer
Generated
in MAC Layer
aka L1/L2
control channel
3.2.3.3 Random Access Procedure
The RACH procedure is used in four cases:
.
Initial access from disconnected state (RRC_IDLE) or radio failure
.
Handover requiring random access procedure
.
DL or UL data arrival during RRC_CONNECTED after UL PHY has lost synchronization (possibly due to
power save operation)
.
UL data arrival when are no dedicated scheduling request (PUCCH) channels available
Timing is critical because the UE can move different distances from the base station, and LTE requires microsecond-
level precision; the speed-of-light propagation delay alone can cause enough change to cause a collision or a timing
problem if it is not maintained.
There are two forms of the RACH procedure: Contention-based, which can apply to all four events above, and non-
contention based, which applies to only handover and DL data arrival. The difference is whether or not there is a
possibility for failure using an overlapping RACH preamble.
3.2.3.4 Contention-based Random Access
Figure 3.9 illustrates the four steps of the contention-based
Figure 3.9 Contention-based Procedure
RACH procedure.
1.
Random access preamble: sent on a special set of
physical layer resources, which are a group of
subcarriers allocated for this purpose
o
Uses Zadoff-Chu sequence, a CDMA-like coding,
to allow simultaneous transmissions to be
decoded
o
6-bit random ID
2.
Random access response
o
Sent on Physical Downlink Control Channel
(PDCCH)
o
Sent within a time window of a few TTI
o
For initial access, conveys at least RA-preamble
identifier, timing alignment information, initial UL
grant, and assignment of temporary C-RNTI
Long Term Evolution Protocol Overview Freescale Semiconductor
o
One or more UEs may be addressed in one response
3.
Scheduled transmission
o
Uses HARQ and RLC transparent mode on UL-SCH
o
Conveys UE identifier
4.
Contention resolution: The eNodeB uses this optional step to end the RACH procedure
3.2.3.5 Non-Contention Based Random Access
In the non-contention based random access procedure, there is no chance of a preamble collision because the code is
pre-assigned by the eNodeB. Figure 3.10 illustrates the three steps in this procedure.
Figure 3.10. Non-Contention Based Procedure
1.
Random access preamble assignment: the eNodeB assigns the 6 bit preamble code
2.
Random access preamble: the UE transmits the assigned preamble
3.
Random access response
o
Same as for contention based RA
o
Sent on PDCCH (Physical Downlink Control Channel)
o
Sent within a time window of a few TTI
o
Conveys at least the timing alignment information and initial Ul grant for handover, and the timing
alignment information for DL data arrival. In addition, RA-preamble identifier if addressed to RA-RNTI on
L1/L2 control channel.
o
One or more UEs may be addressed in one response
3.3 Life of a Packet: Conclusion
Figure 3.11 shows a different view of the LTE protocol stack. This view shows multiple instances of the RLC and PDCP,
which are usually different radio bearers. The radio bearers are characterized by parameters describing the type of
information and QoS transported over the radio interface. One of these groupings might be for voice over IP, another
for a video stream, another for best-effort file transfer or some background processing or another for the control plane.
Scheduling is global, because it controls the relative priority of the radio bearers and logical channels.
Freescale Semiconductor, Inc. Long Term Evolution Protocol Overview
Figure 3.11. LTE Protocol Stack
Downlink
PHY
Uplink
4 LTE Protocol Operation
This section moves away from the flow of packets and examines the LTE protocol on a system level.
4.1 Scheduling
The eNodeB allocates physical layer resources for the uplink and downlink shared channels (UL-SCH and DL-SCH).
Resources are composed of Physical Resource Blocks (PRB) and Modulation Coding Scheme (MCS). The MCS
determines the bit rate, and thus the capacity, of PRBs. Allocations may be valid for one or more TTIs; each TTI interval
is one subframe (1 ms).
Semi-persistent scheduling reduces control channel signaling. If every allocation was individually signaled, the
overhead would be unacceptable. In an application such as voice over IP, for example, a downlink frame occurs every
10 to 20 milliseconds. If each downlink frame were signaled individually, it would cause a lot of traffic on the control
channel and the control channel would need a lot more bandwidth than necessary. Semi-persistent scheduling lets you
set up an ongoing allocation that persists until it is changed. Semi-persistent schedules can be configured for both
uplink and downlink.
4.1.1 Downlink Scheduling
The PDCCH carries the Cell Radio Network Temporary Identifier (C-RNTI), which is the dynamic UE identifier. The CRNTI
indicates that an upcoming downlink resource has been demultiplexed by the MAC and passed on to higher layers
and is now scheduled for this UE. Semi-persistent scheduling periodicity is configured by RRC. Whether scheduling is
dynamic or semi-persistent is indicated by using different scrambling codes for the C-RNTI on PDCCH. The PDCCH is a
very low-bandwidth channel; it does not carry a lot of information compared to the downlink shared channel.
Figure 4.1. Dynamic Scheduling
RLC and PDCP
DL-SCH
MAC
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
D D
De mux De mux
PDCCH
Long Term Evolution Protocol Overview Freescale Semiconductor
Figure 4.2 adds semipersistent scheduling information to the information presented in Figure 4.1. Here, the RRC
configures some of the semipersistent scheduling. This illustration shows a four-TTI example. The first time it actually
occurs there is signaling on the PDCCH. After that, every four TTIs there is a transmission which occurs without any
signaling on the control channel. You can still use dynamic scheduling at the same time for other purposes if necessary;
this carries on until changed by another indication on the control channel.
Figure 4.2. Semipersistent and Dynamic Scheduling
RLC and PDCP
MAC
DL-SCH
PDCCH
4 5 6 7 8 9 10 1112 13 14 1516 1718 19 20
De-mux
D S
De-mux De-mux De-mux
RRC
semi
persistent:
every 4 TTIs
De-mux
S SD
4.1.2 Downlink Scheduling with HARQ
Figure 4.3 continues to build on Figure 4.2. Again, the C-RNTI is found on the PDCCH, indicating that an upcoming
downlink resource is scheduled for this UE.
Figure 4.3. Downlink Scheduling with HARQ
DL-SCH
PDCCH
RLC and up
MAC
4 5 6 7 8 9 10 1112 13 14 1516 17 1819 20
semi
persistent:
every 3 TTIs
PUCCH
De-mux De-mux
D S S S S S
De mux De-mux De mux
Figure 4.4 adds an illustration of the ACK/NACK process to Figure 4.3. HARQ generates an ACK or NACK, which is
sent on L1/L2 control channel (PUCCH) on subframe n+4, for each downlink transport block. Here there is a negative
acknowledgement, so a subframe needs to be retransmitted using HARQ. The retransmission is signaled dynamically
and downlinked, then it is decoded and sent up to higher layers. Finally the subframe has to be acknowledged again.
The process can become fairly complicated when both acknowledgements and semipersistent scheduling are involved.
Freescale Semiconductor, Inc. Long Term Evolution Protocol Overview
Figure 4.4. ACK/NACK Process in Downlink Scheduling
RLC and up
MAC
DL-SCH
PDCCH
PUCCH
4 5 6 7 8 9 10 1112 13 14 1516 1718 19 20
semi
persistent:
every 3 TTIs
De-mux De-mux
ACK ACK ACK
N = 4
De-mux De-mux
D
ACK
De-mux
NACK
Retr
DS S S S S
De-mux
4.1.3 Uplink Scheduling with HARQ
As with the downlink, uplink scheduling information is found on the PDCCH. The C-RNTI indicates that an upcoming
uplink resource is scheduled for this UE in 4 TTI. The 4 TTI delay between when uplink slot availability is present to
when it actually has to be sent gives the UE time to dequeue, determine the proper priority and determine the best way
to pack that transport block with information based on the QoS requirements of the scheduler that it’s running locally.
Figure 4.5 illustrates this process.
Figure 4.5. Uplink Scheduling with HARQ
RLC and up
MAC
TX in
Slot 8
De-queue
Multiplex
TX in
Slot 10
De-queue
Multiplex
4 5 6 7 8 9 10 1112 13 14 15 16 171819 20
PHICH
PDCCH
UL-SCH
ACK NACK
N = 4 N = 4
eNB Response UE Response
Figure 4.6 builds on Figure 4.5 to show the ACK/NACK process. The Physical HARQ Indicator Channel (PHICH) is a
special channel for providing feedback from the eNodeB back to the UE on the HARQ process for the uplink. It carries
ACK/NACK messages for uplink data transport blocks. HARQ is synchronous, with a fixed time of 4 TTI from uplink to
ACK/NACK on the downlink from the eNodeB. The eNodeB responds back with an opportunity to retransmit which is
then scheduled and retransmitted. Although this illustration does not show the positive acknowledgement after that, it
would occur.
Long Term Evolution Protocol Overview Freescale Semiconductor
Figure 4.6. ACK/NACK Process in Uplink Scheduling
RLC and up
MAC
TX in
Slot 8
De-queue
Multiplex
TX in
Slot 10
De-queue
Multiplex
TX in
Slot 20
Re
transmit
4
5
6 7 8 9 10 1112 13 14 1516 171819 20
PHICH
PDCCH
UL-SCH
ACK NACK
N = 4 N = 4
eNB Response UE Response eNB Response
4.2 QoS Architecture
LTE architecture supports “hard QoS,” with end-to-end quality of service and guaranteed bit rate (GBR) for radio
bearers. Just as Ethernet and the internet have different types of QoS, for example, various levels of QoS can be
applied to LTE traffic for different applications. Because the LTE MAC is fully scheduled, QoS is a natural fit.
Evolved Packet System (EPS) bearers provide one-to-one correspondence with RLC radio bearers and provide support
for Traffic Flow Templates (TFT). There are four types of EPS bearers:
.
GBR Bearer . resources permanently allocated by admission control
.
Non-GBR Bearer . no admission control
.
Dedicated Bearer . associated with specific TFT (GBR or non-GBR)
.
Default Bearer . Non GBR, “catch-all” for unassigned traffic
4.3 Management and Control Functions
UE management and control is handled in the Radio Resource Control (RRC). Functions handled by the RRC include
the following:
.
Processing of broadcast system information, which provides information that allows a device to decide if it
wants to connect to the network or not from the access stratum (AS) and non access stratum (NAS)
o
The access stratum is the functional grouping consisting of the parts in the infrastructure and the UE, and
the protocols between these parts, specific to the access technique (i.e. the way the specific physical
media between the UE and the infrastructure is used to carry information). The access stratum provides
services related to the transmission of data over the radio interface and the management of the radio
interface to the other parts of UMTS
.
Paging, which indicates to a device in idle mode that it might have an incoming call
.
RRC connection management between the UE and the eNodeB
.
Integrity protection and ciphering of RRC messages (RRC uses different keys than the user plane)
.
Radio Bearer control (logical channels at the top of the PDCP)
.
Mobility functions (handover during active calls, and cell reselection when idle)
.
UE measurement reporting and control of signal quality, both for the current base station and other base
stations that the UE can hear
.
QoS management maintains the uplink scheduling to maintain QoS requirements for different active radio
bearers
Freescale Semiconductor, Inc. Long Term Evolution Protocol Overview
There are two RRC states:
.
RRC_Idle . the radio is not active, but an ID is assigned and tracked by the network
.
RRC_Connected . active radio operation with context in the eNodeB base station
Figure 4.7 shows a state diagram of functions in the two RRC states. Connected mode measures, transmits and
receives data. The idle state handles cell reselection, network selection, receiving and paging. In discontinuous
reception (DRX), low power mode is configured in both active mode and idle mode. This is a much longer time period in
idle mode; active mode is much shorter, based on activity level.
Figure 4.7. RRC States
.RRC_Idle
. Cell re-selection
..Measurements
. Network (PLMN) selection
..The UE has an ID which
uniquely identifies it in a tracking
area
..No RRC context is stored in the
eNB
. Monitor system information
broadcast
. Receive Paging
. DRX configured by NAS
.RRC_Connected
. Control plane
..eNB context and
E-UTRAN ..RRC connection
..E-UTRAN knows UE’s cell
..Network can transmit and/or
receive data to/from UE
..Network controlled mobility
..Neighbor cell measurements
. User Plane
..UE can transmit and/or receive
data to/from network
..UE monitors control signaling
channel
..UE also reports CQI and feedback
information to eNB
..DRX period configured according
to UE activity level
4.4 Handover and Roaming
Handover is an important function that maintains seamless connectivity when transitioning from one base station to
another. There are two types of handover: intra-RAT, which is within one radio access technology (i.e. LTE-to-LTE from
one eNodeB to another), and inter-RAT between radio access technologies. Inter-RAT could be between LTE and GSM
or 3G WCDMA, 3GPP2, WiMAX or even wireless LAN. These non-LTE handovers are being defined for the LTE
standard. These involve higher layers and often different radio modems, but call continuity is guaranteed with up to 100
milliseconds of disruption when a call is transferred using techniques such as mobile IP or operations in software layers
above the modem stack.
Handover occurs in the active state; it is controlled by the network (the eNodeB).The network uses measurements from
the UE and its own knowledge of the network topology to determine when to handover a UE, and to which eNodeB.
Cell re-selection occurs in the idle state; it is controlled by the UE.
4.4.1 Handover Measurement
In a single-radio architecture, it is challenging to monitor other networks while the receiver is active, because they are
on different frequencies. The radio can only receive on one channel at a time. The radio needs to listen to other
frequencies to determine if a better base station (eNodeB) is available.
In the active state, the eNB provides measurement gaps in the scheduling of the UE where no downlink or uplink
scheduling occurs. Ultimately the network makes the decision, but the gap provides the UE sufficient time to change
frequency, make a measurement, and switch back to the active channel. This can normally occur in a few TTIs. This
has to be coordinated with the DRX, which also causes the system to shut off the radio for periods of time to save
power.
Long Term Evolution Protocol Overview Freescale Semiconductor
4.4.2 Handover: Neighbor Lists
The LTE network provides the UE with neighbor lists. Based on the network knowledge configuration, the eNodeB
provides the UE with neighboring eNB’s identifiers and their frequency. During measurement gaps or idle periods, the
UE measures the signal quality of the neighbors it can receive. The UE reports results back to the eNodeB and the
network decides the best handover (if any), based on signal quality, network utilization, etc.
4.5 Power Save Operation
In wireless data communication, the receiver uses significant power for the RF transceiver, fast A/D converters,
wideband signal processing, etc. As LTE increases data rates by a factor of 50 over 3G, wireless device batteries are
still the same size, so substantial improvements in power use are necessary to operate at these very high rates and
wide bandwidths. Some of that savings comes from hardware, some from system architecture and some from the
protocol.
Wireless standards employ power save mechanisms. The objective is to turn off the radio for the most time possible
while staying connected to the network. The radio modem can be turned off “most” of the time while the mobile device
stays connected to the network with reduced throughput. The receiver is turned on at specific times for updates.
Devices can quickly transition to full power mode for full performance.
4.5.1 DRX and DTX
LTE power save protocols include Discontinuous Reception (DRX) and Discontinuous Transmission (DTX). Both
involve reducing transceiver duty cycle while in active operation. DRX also applies to the RRC_Idle state with a longer
cycle time than active mode. However, DRX and DTX do not operate without a cost: the UE’s data throughput capacity
is reduced in proportion to power savings.
The RRC sets a cycle where the UE is operational for a certain period of time when all the scheduling and paging
information is transmitted. The eNodeB knows that the UE is completely turned off and is not able to receive anything.
Except when in DRX, the UE radio must be active to monitor PDCCH (to identify DL data). During DRX, the UE radio
can be turned off. This is illustrated in Figure 4.8.
Figure 4.8. DRX Cycle
DRX Cycle
On Duration
UE is monitoring PDCCH
4.5.2 Long and Short DRX
In active mode, there is dynamic transition between long DRX and short DRX. Long DRX has a longer “off” duration.
Durations for long and short DRX are configured by the RRC. The transition is determined by the eNodeB (MAC
commands) or by the UE based on an activity timer.
Figure 4.9 shows the DRX cycle operation during a voice over IP example. A lower duty cycle could be used during a
pause in speaking during a voice over IP call; packets are coming at a lower rate, so the UE can be off for a longer
period of time. When speaking resumes, this results in lower latency. Packets are coming more often, so the DRX
interval is reduced during this period.
Freescale Semiconductor, Inc. Long Term Evolution Protocol Overview
Figure 4.9. Long and Short DRX
DRX Level
Long
DRX
Short
DRX
Continuous
Level 1 (RRC Configured)
Level 2 (RRC Configured)
Voice
Activated
Period
Silent
Period
Realtime packets (e.g. VoIP) transmitted to UE
5 Conclusion
The principal of LTE is that the LTE network, like all cellular systems, is designed to operate in scarce and valuable
licensed spectrum. This means that it is highly optimized and a lot of complexity is necessary for the highest possible
efficiency. When the standards body has to choose between efficiency and simplicity, they always choose efficiency to
make the best use of this spectrum.
LTE uses all the time on the downlink for conveying data; the downlink PHY is fully scheduled so there are no gaps due
to arbitration or contention except for the initial access on the random access procedure. The downlink carries multiple
logical channels over one link, so a lot of information is multiplexed together in one transport block, as opposed to other
networks where any given packet is only carrying one type of information at a given time, such as in a control plane or a
user plane.
Long Term Evolution Protocol Overview Freescale Semiconductor
How to Reach Us:
Home Page:
www.freescale.com
Web Support:
http://www.freescale.com/support
USA/Europe or Locations Not Listed:
Freescale Semiconductor, Inc.
Technical Information Center, EL516
2100 East Elliot Road
Tempe, Arizona 85284
+1-800-521-6274 or +1-480-768-2130
www.freescale.com/support
Europe, Middle East and Africa:
Freescale Halbleiter Deutschland GmbH
Technical Information Center
Schatzbogen 7
81829 Muenchen, Germany
+44 1296 380 456 (English)
+46 8 52200080 (English)
+49 89 92103 559 (German)
+33 1 69 35 48 48 (French)
www.freescale.com/support
Japan:
Freescale Semiconductor Japan Ltd.
Headquarters
ARCO Tower 15F
1-8-1, Shimo-Meguro, Meguro-ku,
Tokyo 153-0064, Japan
0120 191014 or +81 3 5437 9125
support.japan@freescale.com
Asia/Pacific:
Freescale Semiconductor Hong Kong Ltd
Technical Information Center
2 Dai King Street
Tai Po Industrial Estate
Tai Po, N.T., Hong Kong
+800 2666 8080
support.asia@freescale.com
For Literature Requests Only:
Freescale Semiconductor
Literature Distribution Center
P.O. Box 5405
Denver, Colorado 80217
1-800-441-2447 or 303-675-2140
Fax: 303-675-2150
LDCForFreescaleSemiconductor@hibbertgroup.com
Information in this document is provided solely to enable system and software implementers
to use Freescale Semiconductor products. There are no express or implied copyright
license granted hereunder to design or fabricate any integrated circuits or integrated circuits
based on the information in this document.
Freescale Semiconductor reserves the right to make changes without further notice to any
products herein. Freescale Semiconductor makes no warranty, representation or guarantee
regarding the suitability of its products for any particular purpose, nor does Freescale
Semiconductor assume any liability arising out of the application or use of any product or
circuit, and specifically disclaims any and all liability, including without limitation
consequential or incidental damages. “Typical” parameters which may be provided in
Freescale Semiconductor data sheets and/or specifications can and do vary in different
applications and actual performance may vary over time. All operating parameters,
including “Typicals” must be validated for each customer application by customer’s technical
experts. Freescale Semiconductor does not convey any license under its patent rights nor
the rights of others. Freescale Semiconductor products are not designed, intended, or
authorized for use as components in systems intended for surgical implant into the body, or
other applications intended to support or sustain life, or for any other application in which
the failure of the Freescale Semiconductor product could create a situation where personal
injury or death may occur. Should Buyer purchase or use Freescale Semiconductor
products for any such unintended or unauthorized application, Buyer shall indemnify and
hold Freescale Semiconductor and its officers, employees, subsidiaries, affiliates, and
distributors harmless against all claims, costs, damages, and expenses, and reasonable
attorney fees arising out of, directly or indirectly, any claim of personal injury or death
associated with such unintended or unauthorized use, even if such claim alleges that
Freescale Semiconductor was negligent regarding the design or manufacture of the part.
Freescale and the Freescale logo are trademarks or registered trademarks of Freescale
Semiconductor, Inc. in the U.S. and other countries.
All other product or service names are the property of their respective owners.
ⓒ Freescale Semiconductor, Inc. 2008.
Document Number: LTEPTCLOVWWP
REV 0