3GPP Technology Blog

Transmission bandwidth configuration: Number of RBs for given BW

2013-05-22T11:56:00.002-07:00

Transmission bandwidth configuration N_RB in E-UTRA channel bandwidths

Channel bandwidth BW_Channel [MHz]	1.4	3	5	10	15	20
Transmission bandwidth configuration N_RB	6	15	25	50	75	100

Channel Bandwidth in LTE

2012-06-17T17:50:00.003-07:00

The MIB in LTE tell the UE the prevalent Channel Bandwith in number of resource blocks

value of dl-bandwidth as in MIB corresponds to :

6 = 1.4 MHz
15 = 3 MHz
25 = 5 MHz
50 = 10 MHz
75 = 15MHz
100 = 20 MHz

Definition of channel bandwidth and transmission bandwidth configuration for one E‑UTRA carrier

source 3GPP TS 36.101

E-UTRA operating bands

2012-06-17T17:46:00.005-07:00

E-UTRA operating bands

E‑UTRA Operating Band

Uplink (UL) operating band
BS receive
UE transmit

Downlink (DL) operating band
BS transmit
UE receive

Duplex Mode

F_{UL_low} – F_{UL_high}

F_{DL_low} – F_{DL_high}

1920 MHz

–

1980 MHz

2110 MHz

–

2170 MHz

FDD

1850 MHz

–

1910 MHz

1930 MHz

–

1990 MHz

FDD

1710 MHz

–

1785 MHz

1805 MHz

–

1880 MHz

FDD

1710 MHz

–

1755 MHz

2110 MHz

–

2155 MHz

FDD

824 MHz

–

849 MHz

869 MHz

–

894MHz

FDD

6¹

830 MHz

–

840 MHz

875 MHz

–

885 MHz

FDD

2500 MHz

–

2570 MHz

2620 MHz

–

2690 MHz

FDD

880 MHz

–

915 MHz

925 MHz

–

960 MHz

FDD

1749.9 MHz

–

1784.9 MHz

1844.9 MHz

–

1879.9 MHz

FDD

1710 MHz

–

1770 MHz

2110 MHz

–

2170 MHz

FDD

1427.9 MHz

–

1447.9 MHz

1475.9 MHz

–

1495.9 MHz

FDD

699 MHz

–

716 MHz

729 MHz

–

746 MHz

FDD

777 MHz

–

787 MHz

746 MHz

–

756 MHz

FDD

788 MHz

–

798 MHz

758 MHz

–

768 MHz

FDD

Reserved

FDD

Reserved

FDD

704 MHz

–

716 MHz

734 MHz

–

746 MHz

FDD

815 MHz

–

830 MHz

860 MHz

–

875 MHz

FDD

830 MHz

–

845 MHz

875 MHz

–

890 MHz

FDD

832 MHz

–

862 MHz

791 MHz

–

821 MHz

FDD

1447.9 MHz

–

1462.9 MHz

1495.9 MHz

–

1510.9 MHz

FDD

3410 MHz

–

3490 MHz

3510 MHz

–

3590 MHz

FDD

2000 MHz

–

2020 MHz

2180 MHz

–

2200 MHz

FDD

1626.5 MHz

–

1660.5 MHz

1525 MHz

–

1559 MHz

FDD

1850 MHz

–

1915 MHz

1930 MHz

–

1995 MHz

FDD

...

1900 MHz

–

1920 MHz

1900 MHz

–

1920 MHz

TDD

2010 MHz

–

2025 MHz

2010 MHz

–

2025 MHz

TDD

1850 MHz

–

1910 MHz

1850 MHz

–

1910 MHz

TDD

1930 MHz

–

1990 MHz

1930 MHz

–

1990 MHz

TDD

1910 MHz

–

1930 MHz

1910 MHz

–

1930 MHz

TDD

2570 MHz

–

2620 MHz

2570 MHz

–

2620 MHz

TDD

1880 MHz

–

1920 MHz

1880 MHz

–

1920 MHz

TDD

2300 MHz

–

2400 MHz

2300 MHz

–

2400 MHz

TDD

2496 MHz

2690 MHz

2496 MHz

2690 MHz

TDD

3400 MHz

–

3600 MHz

3400 MHz

–

3600 MHz

TDD

3600 MHz

–

3800 MHz

3600 MHz

–

3800 MHz

TDD

NOTE 1: Band 6 is not applicable

source 3GPP TS 36.101

A trip around the HSPA world

2011-12-29T22:12:00.000-08:00

I am writing this post to illustrate the path that a byte of data takes to go around the HSPA network. The details are conforming to Rel 6 of 3GPP specifications. This is just an overview, for details please refer the 3GPP specification.

The dialogue is started from the mobile device(it can be a HTTP req, email push etc), where a few bytes travel through the L3, L2 and PHY layers in the Uplink direction to reach the network nodes and routed to the application end point. There is a response from the application end point which consists of bytes destined to the Mobile(IP address).The bytes in the DL again have to go through the network nodes and the protocol layers before they reach the mobile.

Since Application end point can be any where in the internet, the start/end of the journey is from/to the GGSN on the network side.

The user starts a application such as Mobile Originated email(or HTTP req from a browsing session, ...) on the mobile, which triggers data in the Uplink. The application submits the data (as an IP packet in most cases) to the mobile modem to be transmitted in Uplink. It is assumed here that the RRC Connection and the Radio Links are already in place and the UE is in DCH state. The data is submitted under a PDP Context. A PDP Context is assigned in the GGSN where in an IP address is assigned for the connection. This connection is associated to a Radio bearer which enables the data transfer between L3 end points between mobile and network. Radio Bearer is like a service provided by the L3 to transfer data across the Air interface up to the L3 on the other end.

The modem inputs this data to the string of Radio Interface Protocols(RIP) to be packaged suitably to be transmitted over the air. Each layer of the Radio Interface Protocol plays a role in efficient and reliable transfer of the data across the Air Interface. The unit of data entering a layer is called the Service Data Unit(SDU) and after processing the unit of data exiting the layer is the Packet Data Unit(PDU). So the PDU of a upper layer becomes the SDU of the lower layer as the data traverses the RIP. The figure below shows the architecture of the RIP.

The first station in this packaging is the PDPC layer. Packet Data Convergence Protocol(PDCP) compresses the IP header in the packet. This enables more efficient use of the radio resources by eliminating some of the redundant information in the IP header(20 bytes for IPv4). This is all the more effective in case of streaming or file transfer where the packets travel between two fixed end points(IP addresses). PDCP compresses the header, attaches a header of its own(containing information about the algorithm and a sequence number of the PDU). The PDCP PDU is sent to the RLC.

Radio Link Control(RLC) layer maps the data from the Radio bearers onto the Logical Channels. The RLC does the segmentation/concatenation of RLC SDUs, ciphers and deciphers the data on RLC AM and UM mode, and provides for re transmission of the PDUs which were NACKed. Depending upon the logical channel type required, a RLC entity is setup(TM, UM or AM). Bidirectional logical channels which require reliable transfer are setup with a AM RLC entity. TM is for real time applications (like voice call), which donot require re transmission and there is no overhead of RLC header(transparent). There are not many applications which utilise the UM mode. The IP packet could be segmented depending on the RLC PDU size configured for the RLC AM entity, in which case the IP packet will be carried in multiple RLC PDUs. These RLC PDU(s) are forwarded to the MAC layer on Logical Channels.

Medium Access Control(MAC) is the Link layer protocol. In Rel 6, the MAC layer is composed of MAC-e(on UE and NodeB), MAC-d(on UE and NodeB) and MAC-es(on RNC) in the Uplink direction(other MAC entities can also be present depending on configuration and transport channel to be used). The MAC plays many important roles in the reliable and efficient data transmission. It maps logical channels to transport channels, selects transport format for the transmission in each TTI, schedules the data according as the logical channel priority, among other functions.In this case the transport channel is the E-DCH Channel. The IP packet which started as part of an email from the mobile user, and has been delivered to MAC in form of RLC PDU(s) is now resting in MAC priority queues, waiting for Uplink resources (Scheduling Grants) to be scheduled for transfer on a transport block. When the moment arrives(the logical channel is scheduled), the RLC PDU(s) and formed into a MAC-e PDU to be delivered to the Physical Layer. Here again the RLC PDUs could be segmented to fit into a MAC-e PDU or several RLC PDUs could be concatenated to form a MAC-e PDU.

The UE side MAC architecture is illustrate in the figure below:

The Physical Layer(PHY) maps the transport channels onto Phycical channels. In this case the Physical channel is the E-DPDCH channel. The data received in a transport block from MAC is to be transmitted onto one radio frame(2ms or 10ms long, depending on configuration).

The radio transmissions are received at the NodeB and decoded. The successfully decoded transport blocks are forwarded to the MAC-e. MAC-e besides processing the MAC-e SDU does other important jobs like handles UE feedback of the channel quality and outstanding traffic, schedule UL resources among UEs participating in UL E-DCH transfer according as the UE priority. Here the RLC SDUs are extracted from the MAC-e PDU and are delivered to RLC (via MAC-es and MAC-d). MAC-d deciphers the PDUs MAC-es resides on the RNC and does macro-diversity combining (combining the data received from multiple Radio Links on separate NodeB in case of SHO) and seuencing of the RLC SDUs. The RLC SDUs are sent to RLC on the logical channel which they belong to.

At RLC, the SDUs meet the RLC entity which is corresponding to its counterpart on the mobile. The IP packet is still hidden in the RLC SDUs. Here RLC SDUs are deciphered and then concatenated if required to retrieve the PDCP SDU, which is sent to the PDCP.

The PDCP decompresses the IP header and sends of the IP packet to the NAS(Non Access Stratum) nodes for routing to the destination. The IP packet is sent over the IuPS interface on a GTP (GPRS Tunneling Protocol) tunnel. The packet reached GGSN after going through SGSN. Direct Tunneling provides to bypass the SGSN and deliver the IP packet directly to GGSN.

At GGSN the IP packet is associated to a PDP context. Here the IP header is modified such that the outside world can send back the response to the correct place. After this modification, it is routed to its destination(the mail server). The response from the server comes back to GGSN where it is tagged to a PDP context, and sent off onto the GTP tunnel to the RNC.

From here starts the journey of the IP packet in the downlink towards its destination on the Mobile which started the dialogue by triggering the Mobile Originated email.

The processing in the downlink is spread over RNC and NodeB, with RLC and some functions of MAC(MAC-d) on RNC and MAC-hs and PHY on the NodeB.

The PDCP gets the IP packet for transmission in Downlink. The role of PDCP is same as in UL; compressing the IP header and attaching another header. Next is RLC, where a RLC AM entity has been instantiated to handle the logical channel for the DL data. It makes the RLC PDUs and sends them off to MAC-d.

MAC-d does ciphering and flow control towards MAC-c(if there is any) or FP(Framing Protocol). Framing protocol is required to synchronise RNC with multiple NodeBs. There is time alignment and Transport Channel synchronisation between the FP on RNC and NodeB. Time synchronisation makes adjustments for the difference in clock. Transport Channel sync corrects the timing of the arrival of data from RNC to NodeB(done on a Transport Channel basis). This is required to ensure availability of data at the instance it has to be transmitted(CFN) and to limit the buffering of data on NodeB. Next stop is the MAC-hs.

MAC-hs (high speed) is the MAC entity handling the HS-DSCH transport channel in DL. It has many improvements over its predecessor. Adaptive Modulation and coding allows for adaption to the channel conditions, Hybrid ARQ(HARQ) provides for fast retransmission and soft combining with incremental redundancy. Another function is fast scheduling, where in the MAC-hs allocates the available WCDMA codes among the competing UEs, based on the channel quality feedback from the UE, UE priority, pending data, and other factors. The IP packet that has hit this acceleration station, finds itself becoming a part of thebig, roomy MAC-hs PDU which is assigned one HARQ process for transmission to the Mobile.

The MAC architecture from the UTRAN side is shown below in the figure.

After reaching the Mobile, the MAC-hs PDU traverses through MAC, RLC and PDCP, finding its familiar form of the IP packet. Once fully formed, the IP packet is delivered to the Application depending on the destination IP address or Transport port number. In this case this is the response from the mail server. It will take many such IP packets to successfully transmit the email from the Mobile. All of those will have to wade through the intricate layers of the Radio Interface Protocols in order to reach their destinations.

Link Layer and App Layer Speeds

2011-12-28T13:27:00.000-08:00

The throughput experienced at the application layer is always lower than the Link layer speed (which is in most cases advertised). The speeds published by the Network Service Providers is reduced because of number of factors.

For definition:

Link Layer Speed: is the throughput at the Link Layer. This is the rate with which a L2 packet is transferred between the corresponding L2 entities.
Application Layer Speed: is the speed experienced by the user. The rate at which the data useful to the application is transferred between the Application End points.

The extent of reduction in the Application layer speed depends upon:

IP and transport protocol overhead
UMTS Layer 2, Layer 3 Header overheads
Segmentation/Concatenation
Amount of Padding

The IP carries almost all of the data traffic and adds a IP header. The transport protocol adds another header and extra bytes to the data to be transmitted.

The UMTS architecture has multiple entities on the UE and network side interacting with each other. These require some control information in order to transfer data among each other.

The L2 (MAC) and L3(RLC) headers add to the over head. Next is the efficiency of the algorithm at MAC which packs the data in MAC PDUs. This algorithm should be smart enough to keep the amount of padding to a minimum.

The figure above shows the overhead added at each layer. Though not to scale, it highlights all the extra bytes that the MAC PDU is carrying for the Application data bytes that it is carrying.
Also, here it is assumed that the IP packet is transmitted as a whole, which is seldom the case. At most of the time, the IP packet might be segmented into RLC PDUs and then one RLC PDU segmented into MAC PDUs, depending on the TB size selected by MAC.

This overhead is even more pronounced when the MAC PDU size is small. Therefore it is important that RLC PDU size and the MAC PDU sizes be chosen in a way that the segmentation is minimized.

Lets take an example to see the magnitude of this overhead:
For a application using 384 kbps PS RAB

Higher layer	RAB/Signalling RB		RAB
RLC	Logical channel type		DTCH
	RLC mode		AM
	Payload sizes, bit		320
	Max data rate, bps		384 000
	AMD PDU header, bit		16
MAC	MAC header, bit		0
	MAC multiplexing		N/A
Layer 1	TrCH type		DCH
	TB sizes, bit		336
	TFS	TF0, bits	0x336
		TF1, bits	1x336
		TF2, bits	2x336
		TF3, bits	4x336
		TF4, bits	8x336
		TF5, bits	12x336
		TF6, bits	N/A (alt. 16x336)
		TF7, bits	N/A (alt. 20x336)
		TF8, bits	N/A (alt. 24x336)
	TTI, ms		10 (alt. 20)
	Coding type		TC
	CRC, bit		16
	Max number of bits/TTI after channel coding		12 684 (alt. 25 368)
	RM attribute		110 to 150

Lets assume for simplicity a 5 byte ping. Each ping packet will be carried in a IP packet with ICMP header and will be fit inside one Transport Block for carrying over the air. The application data transferred in each 336 bit of Transport Block is:

ping size = 5 * 8 = 40 bits

The header sizes are:

336 - 20*8(IP header) - 8*8(ICMP header) - 16 (MAC-header) - 16 (RLC AM header) - 40 (padding) = 40 bits.

Optimizing Modulation for MIMO Operation

2011-12-10T13:38:00.001-08:00

MIMO Operation s marred by Inter Channel Interference and Inter Antenna Synchronisation, because it involves simultaneous transmission/reception on Multiple Antennas. The transmission is pre-coded according to the PMI (Precoding Matrix Indicator) which is used at the receiver to identify the transmission from each of the antennas.

The ICI mitigation and synchronisation requirement make the receptor design too complicated and also limit the gain to 1.5X rather than the theoretical 2X for a 2X2 MIMO.

An improvisation in the Modulation used for MIMO operation is proposed in a IEEE paper which will not require simultaneous transmission/reception. Only one antenna will transmit and one antenna will receive the signal. This will remove the ICI and the requirement for antenna synchronisation, while still giving increased throughput with the use of MIMO (the increase will be less than the theoretical conventional MIMO, but will match the practical gain).

The new modulation technique uses the antenna as a modulation unit (called Space Modulation).
The Space Modulated Symbol is formed by the combination of the conventional modulation symbol (QPSK or 16-QAM) and the index of the antenna used for transmission.

Example for No of Tx Antenna = 4

With 4 antennas the transmission can be from antenna 1,2,3 or 4. The symbols therefore are:

With QPSK used for Modulating the bit stream, the no of bits in the Space Modulated Symbol = 2(QPSK)X2=4. This is same as 16-QAM. Therefore using Space Modulation on top of QPSK will result in 2X data speed, same as in 16-QAM.

(QPSK Symbol)

4 Tx Antenna adding 2 bits to the Space Modulated Symbol

The increase in the data rate is logarithmic with the number of Tx Antennas in use.As compared to MIMO, where the increase in data rate is linear with no of antennas.

MIMO

New Data Rate = (No of Antenna)x(Original Data Rate)

MIMO with Space Modulation

New Data Rate = log2(No of Antenna)x(Original Data Rate)

For DL only MIMO operation, the design of the UE receiver can be greatly simplified. Multiple Rx antennas will provide diversity gain, and a single Rx antenna will also work.

Multiple Operator Core Network

2011-12-10T12:24:00.001-08:00

This is a way to share the network resources among multiple operators. The same network infrastructure is used to transmit/receive for different operators.
The MIB contains a list of multiple PLMN Identities. The UE reads the MIB and selects a PLMN to register to, based on its subscription.

MIB Contents:

.........
Common PLMN
..........

mibPLMN-Identity BOOLEAN,
multiplePLMNs SEQUENCE (SIZE(1..5))

PLMN1
PLMN2

PLMN1, PLMN2, etc are the multiple PLMNs in the cell. If mibPLMN-Identity is st to TRUE, the supporting UE can include the common PLMN identity also in the list of PLMNs to choose from. If it is FALSE, the supporting UE does not attempt to register on the common PLMN.

The terms associated with MOCN and their definitions are:

Conventional network: A PLMN consisting of radio access network and core network, by which only one serving operator provides services to its subscriber. Subscribers of other operators may receive services by national or international roaming.

Common PLMN: The PLMN-id indicated in the system broadcast information as defined for conventional networks, which non-supporting UEs understand as the serving operator.

Core network operator: An operator that provides services to subscribers as one of multiple serving operators that share at least a radio access network. Each core network operator may provide services to subscriber of other operators by national or international roaming.

Gateway Core Network: A network sharing configuration in which parts of the core network (MSC/SGSNs) are also shared.

Multi-Operator Core Network: A network-sharing configuration in which only the RAN is shared.

Non-supporting UE: A UE that does not support network sharing in the sense that it ignores the additional broadcast system information that is specific for network sharing. In other specifications, the term "network sharing non-supporting UE" may be used.

Supporting UE: A UE that supports network sharing in the sense that it is able to select a core network operator as the serving operator within a shared network. In other specifications, the term "network sharing supporting UE" may be used.

Downlink Speed calculation in LTE

2011-11-26T21:15:00.001-08:00

LTE is said to have speed upto 100 mbps in Downlink, following is the calculation for it:

The speed is the number of bits that can be transmitted in DL in one second with all the resources assigned to a single user. This is the theoretical speed (for political purposes), practical speeds will be less that this as the control, broadcast channels, reference signals have to be accommodated in the same resources.

The speed is calculated starting with the 100 Resource Blocks for 20 MHz. Each Resource Block has 12 Resource Elements. With 1 sub-frame of size 1 ms, there are 1000 sub-frames in 1 sec. Each subframe can carry 7 OFDMA symbols for Normal Cyclic Prefix. If 64 QAM modulation is used, each symbol can communicate 6 bits and with IQ multiplexing the number of bits can be doubled.

Therefore the possible speed is:

100 Resource Blocks (for 20 MHz bandwidth)
X 12 Resource Elements per RB
X 1000 sub-frames per second (1 subframe = 1 ms)
X 7 symbols per sub-frame (Normal Cyclic Prefix)
X 6 (1 symbol in 64 QAM represents 6 bits)
X 2 (IQ multiplexing)

= 100.8 Mbps speed in DL

Mapping of Logical Transport and Physical Channels in LTE

2011-11-23T12:40:00.001-08:00

The following figures show the mapping of Logical channels to transport channels and Physical Channels in LTE in Uplink

The following figures show the mapping of Logical channels to transport channels and Physical Channels in LTE in DownLink

RNTIs in LTE

2011-11-23T12:08:00.001-08:00

Radio Network Temporary Identifier (RNTI) are used in LTE to

Associate the data in UL and DL with a Transport channel: Like the P-RNTI and SI-RNTI identify the Paging and the System Information Channels
address a particular UE on a shared Transport Channel; like the C-RNTI identifies the data for a UE on the DL-SCH channel

On the shared channels, the UL-SCH and the DL-SCH, the C-RNTI is used to initialise the scrambling code that is used to scramble the data stream before modulation. This allows the data to be multiplexed on the shared transport channel. The intended UE can un-scramble the data using the C-RNTI assigned to it and receive the data.

LTE simplifies the use of RNTI as there is no need to define URNTI, ERNTI, HRNTI, SRNTI in LTE, since there is only one shared MAC entity.

RNTI values

Value (hexa-decimal)	RNTI
0000	N/A
0001-003C	RA-RNTI, C-RNTI, Semi-Persistent Scheduling C-RNTI, Temporary C-RNTI, TPC-PUCCH-RNTI and TPC-PUSCH-RNTI (see note)
003D-FFF3	C-RNTI, Semi-Persistent Scheduling C-RNTI, Temporary C-RNTI, TPC-PUCCH-RNTI and TPC-PUSCH-RNTI
FFF4-FFFD	Reserved for future use
FFFE	P-RNTI
FFFF	SI-RNTI

NOTE: The values corresponding to the RA-RNTI values of a cell’s PRACH configuration are not used in the cell for any other RNTI (C-RNTI, Semi-Persistent Scheduling C-RNTI, Temporary C-RNTI, TPC-PUCCH-RNTI or TPC-PUSCH-RNTI).

The table below gives the usage of the RNTIs in LTE.

RNTI usage

RNTI	Usage	Transport Channel	Logical Channel
P-RNTI	Paging and System Information change notification	PCH	PCCH
SI-RNTI	Broadcast of System Information	DL-SCH	BCCH
RA-RNTI	Random Access Response	DL-SCH	N/A
Temporary C-RNTI	Contention Resolution (when no valid C-RNTI is available)	DL-SCH	CCCH
Temporary C-RNTI	Msg3 transmission	UL-SCH	CCCH, DCCH, DTCH
C-RNTI	Dynamically scheduled unicast transmission	UL-SCH	DCCH, DTCH
C-RNTI	Dynamically scheduled unicast transmission	DL-SCH	CCCH, DCCH, DTCH
C-RNTI	Triggering of PDCCH ordered random access	N/A	N/A
Semi-Persistent Scheduling C-RNTI	Semi-Persistently scheduled unicast transmission (activation, reactivation and retransmission)	DL-SCH, UL-SCH	DCCH, DTCH
Semi-Persistent Scheduling C-RNTI	Semi-Persistently scheduled unicast transmission (deactivation)	N/A	N/A
TPC-PUCCH-RNTI	Physical layer Uplink power control	N/A	N/A
TPC-PUSCH-RNTI	Physical layer Uplink power control	N/A	N/A

RA-RNTI: The Random Access RNTI is used on the PDCCH when Random Access Response messages are transmitted. It unambiguously identifies which time-frequency resource was utilized by the UE to transmit the Random Access preamble.

References:
[1] 36.421
[2] 36.211

Physical Layer Bit Processing in LTE - 2

2011-11-12T20:17:00.001-08:00

Continued from the last post

The transport channels are mapped to Physical Channels, and after the transport channel processing, as described in last post, the Physical Channel Processing takes place. This is again for LTE Downlink.

Scrambling, Modulation, Layer Mapping, Precoding and Resource Element Mapping

The baseband signal representing a downlink physical channel is defined in terms of the following steps:

- scrambling of coded bits in each of the code words to be transmitted on a physical channel

- modulation of scrambled bits to generate complex-valued modulation symbols

- mapping of the complex-valued modulation symbols onto one or several transmission layers

- precoding of the complex-valued modulation symbols on each layer for transmission on the antenna ports

- mapping of complex-valued modulation symbols for each antenna port to resource elements

- generation of complex-valued time-domain OFDM signal for each antenna port

Scrambling

Scrambling the bits make them less prone to interference as the process make the value of bits to be transmitted to be psuedo random.

The scrambling code used is a length 31 gold sequence which yields 2^31 puseddo random sequences. If the input code word is :

b^q(0), b^q(1),b^q(2) .......... b^q(M^q_bit- 1); where M^q_bit is the number of bits in the code word q.

and scrambling code is:

c^q(0), c^q(1),c^q(2) .......... c^q(M^q_bit- 1)

the scrambling is done as shown in figure below.

Modulation

Modulation converts the bits in the code words into complex-valued symbols. One symbol can represent n bits depending on the modulation type. (n=2 for QPSK, n=4 for 16 QAM, n=6 for 64 QAM).

The signal s(t) is split into Re and Img parts which are multiplied by the carrier and 90 degree shifted carrier and then summed up to give the compled valued modulation symbol.

For each code word, the block of scrambled bits
b^q~(0), b^q~(1),b^q~(2) .......... b^q~(M^q_bit- 1)

shall be modulated using one of the modulation schemes in table below, resulting in a block of complex-valued modulation symbols
d^q(0), d^q(1), d^q(2) .......... d^q(M^q_bit- 1).

Physical channel	Modulation schemes
PDSCH	QPSK, 16QAM, 64QAM
PMCH	QPSK, 16QAM, 64QAM

Modulation schemes

The complex valued modulation symbol, onto which the bits in the code word are mapped, is specified by the amplitude of the Real component(I) and the Imaginary component(Q).

Modulation Symbol = I + jQ

For QPSK, the modulation symbols are:

QPSK Modulation

The I, Q for different modulation schemes are given in the tables below

Layer Mapping

The complex-valued modulation symbols for each of the code words to be transmitted are mapped onto one or several layers. Complex-valued modulation symbols
d^q(0), d^q(1), d^q(2) .......... d^q(M^q_bit- 1),

for code word q shall be mapped onto the n layers (equal to the number of antenna ports), with M^layer_symb is the number of modulation symbols per layer

The following table shows how the bits in the code word are mapped onto different layers of transmission for spatial multiplexing.

For 1 layer and 1 code word (single antenna port and single data stream); the mapping is strainghtforward, with the symbols from the code word added to the layer in order. Note that number of symbols in code word 0 (M⁽⁰⁾_symb) is equal to the number of symbols that the layer can take ( M^layer_symb).

For 2 layer and 2 code words (two antenna ports and dual data stream); one data stream is mapped onto one antenna port and the other data stream on the pther port. Note that number of symbols in code word 0 (M⁽⁰⁾_symb) or code word 1 (M⁽¹⁾_symb) is equal to the number of symbols that the each layer can take ( M^layer_symb).

For 2 layer and 1 code word (two antenna ports and single data stream); one data stream is distributed over the two layers. Note that number of symbols in code word 0 (M⁽⁰⁾_symb) is twice that of the number of symbols that the layer can take ( M^layer_symb).

Other combinations are shown in the able below:

Precoding

Precoding is done to adapt to the channel conditions. The transmission is precoded according to the frequency response of the channel. The precoding is done in such a way that original signal (before precoding was applied) is received at the Rx end after the transmission undergoes the frequency response of the channel.

In LTE a feedback mechanism is in place for MIMO, where the UE signals back to the eNB one of the three possible precoding matrix to be applied during precoding of the signal.

Mapping to resource elements

For each of the antenna ports used for transmission of the physical channel, the block of complex-valued symbols y^p(0), y^p(1), y^p(2) .......... y^p(M^ap_symb- 1) shall be mapped in sequence starting with y^p(0) to resource elements (k,l) which meet all of the following criteria:

-they are in the physical resource blocks corresponding to the virtual resource blocks assigned for transmission, and

-they are not used for transmission of PBCH, synchronization signals or reference signals, and

-they are not in an OFDM symbol used for PDCCH as described in the post on PDCCH.

The mapping to resource elements (k,l) on antenna port p not reserved for other purposes shall be in increasing order of first the index k over the assigned physical resource blocks and then the index l, starting with the first slot in a subframe.

Physical Layer Bit Processing in LTE - 1

2011-11-12T15:56:00.001-08:00

The bits that are carried on the radio waves undergo a lot of processing before these are ready for transmission. The processing makes them resilient to the attenuation, and also packs them in a way to make them economical to transmit over the air.

The processing starts right from the point where they leave the MAC layer and continues till they are transformed to a voltage to be transmitted on the Tx antenna. Corresponding processing happens when the radio waves hit the Rx antenna till the point these are ready to be delivered to the MAC layer.

We will look at the processing for transmitting the bits in LTE below(Downlink):

CRC Attachment, Channel coding, Rate Matching and Interleaving

The first stage of processing is the Transport Channel Processing. The parameters for the processing are defined by the transport channel on which the data is to be transmitted and are given by the MAC layer(MAC layer maps logical channels to transport channels). The data/control streams from the MAC layer are subject to the following processing:

CRC Attachment
Channel coding
Rate Matching and
Interleaving

The bit sequence resulting from this stage is input to the next stage.

Transport Channel Processing

CRC Attachment

CRC attachment allows for detection of bits which are received in error. If the CRC fails, the bits have to be retransmitted.

The CRC length can be 8, 16 or 24 according to configuration. The CRC generator polynomial for parity bits length L is:

- g_CRC24A(D) = [D²⁴ + D²³ + D¹⁸ + D¹⁷ + D¹⁴ + D¹¹ + D¹⁰ + D⁷ + D⁶ + D⁵ + D⁴ + D³ + D + 1] and;

- g_CRC24B(D) = [D²⁴ + D²³ + D⁶ + D⁵ + D + 1] for a CRC length L = 24 and;

- g_CRC16(D) = [D¹⁶ + D¹² + D⁵ + 1] for a CRC length L = 16.

- g_CRC8(D) = [D⁸ + D⁷ + D⁴ + D³ + D + 1] for a CRC length of L = 8.

For the CRC Computation the input bits are left shifted by L(CRC length).The CRC generator polynomial is then XORed with the input bits by right shifting the polynomial in each step until all the input bits are zero. The resulting L bits are the parity bits which are appended to the input bits to result in the output bits, as shown in the figure below.

CRC Computation

At the receiver side the received bits are divided by the CRC polynomial and if the result is zero, the CRC check is PASS.

The CRC is used on the MIB to indirectly communicate one of the three Antenna Configuration in use in the DL. The CRC of the MIB data transmitted undergoes a special mask to encode this information. The UE does trial and error method until it gets zero after division with three possible CRC polynomials to determine the mask applied and thus knows the DL antenna configuration in use.

For Code block segmentation and code block CRC attachment, please refer cluse 5.1.2 of Specification 36.212

Channel Coding

Channel coding makes the bits resilient to the channel attenuation. Although this results in reduction of the data rate, it is essential to combat the attenuation.

The bit sequence input for a given code block to channel coding is denoted by c₀, c₁, c₂, ....... c_K-1, where K is the number of bits to encode. After encoding the bits are denoted by d⁽ⁱ⁾₀, d⁽ⁱ⁾₁, d⁽ⁱ⁾₂, d⁽ⁱ⁾_{3,
.................} , d⁽ⁱ⁾_D-1, where D is the number of encoded bits per output stream and i indexes the encoder output stream. The relation between c_K and d⁽ⁱ⁾_K and between K and D is dependent on the channel coding scheme.

The following channel coding schemes can be applied to TrCHs:

- tail biting convolutional coding;

- turbo coding.

The values of D in connection with each coding scheme:

- tail biting convolutional coding with rate 1/3: D = K;

- turbo coding with rate 1/3: D = K + 4.

The range for the output stream index i is 0, 1 and 2 for both coding schemes.

Rate Matching and Interleaving

The 3 coded streams from the channel coding are subject to rate matching to match the bit rate of the transport channel.

The rate matching for channel coded transport channels is defined per coded block in 36.212 and consists of interleaving three information bit streams dk(0), dk(1) and dk(2), followed by the collection of bits and the generation of a circular buffer, as shown in figure below. The circular buffer based rate matching provides a simple method of generating puncturing patterns with good performance.

For the next stage of processing, go to the post on Physical Channel Processing.

Physical hybrid ARQ indicator channel (PHICH)

2011-11-07T21:00:00.000-08:00

The characteristics of PHICH are:

The PHICH carries the hybrid-ARQ ACK/NAK
Multiple PHICHs mapped to the same set of resource elements constitute a PHICH group
PHICHs within the same PHICH group are separated through different orthogonal sequences.
Modulation used is BPSK

PHICH channel is like using CDMA over OFDM, where a UE is assigned a PHICH group(identifying the resource elements carrying the PHICH) and a orthogonal sequence(just like a CDMA code, which allows PHICH to be multiplexed over as many UEs as are the orthogonal codes within the group). For Normal CP SF=4 codes are used and for Extended CP, SF=2 codea are used.

Number of PHICH groups (for Normal CP) = Ceiling [Ng*(num-dl-RB/8)]

where Ng = {1/6, 1/2,1,2} as defined by higher layers

num-dl-RB = Number of Resource Blocks in DL

Therefore for 20 MHz (num-DL-RB = 100) if Ng = 2 then

number of PHICH groups = ceiling (2*(100/8)) = 25

The orthogonal codes used for Normal CP are SF=4 codes and after BPSK modulation and using IQ multiplexing,

num of users in each PHICH group = 8

Therefore the number of UEs that can be supported on PHICH on 20 MHz and Normal CP = 25*8 = 200

PCFICH and PDCCH

2011-11-06T10:56:00.000-08:00

Physical Control Format Indicator Channel (PCFICH)

As the name suggests, this channel communicates the number of OFDMA symbols that the PDCCH will be transmitted on, to the UE.
The PDCCH can take either {1,2,3} OFDMA symbols for BW > 1.4 MHz and values {2,3,4} for BW 1.4 Mhz.
The two bits required to carry one of the 4 values is encoded into 32 bits. QPSK modulation is used which yields 16 bits to be transmitted.
The 16 bits for the PCFICH are spread over 16 REs in 4 groups of 4 REs each. This provides added protesction from frequency selective fading.
The PCFICH shall be transmitted on the same set of antenna ports as the PBCH.
The PCFICH is spread over the whole range of sub-carriers which provides it with some imunity to the frequency selective fading

PCFICH and PDCCH Channel

Physical downlink control channel(PDCCH)

The PDCCH is the control channel which is specified by the PCFICH, as shown in the figure above
The channel carries downlink assignments for data on the Physical Downlink Shared Chaneel (PDSCH)
The modulation used in QPSK
The number of OFDMA symobols occupied by PDCCH in the first slot of each sub frame is given by PCFICH (in absence of PCFICH, the PDCCH is also absent). Let the number of OFDMA symbols be L.

PDCCH Resource Element Mapping

The block of quadruplets to be transmitted on the PDCCH are mapped to Resource Elements (k',l').

(k',l') where k' = 0 ... num-dl-rb*12 and l' = 0 ... L

The algorithm for the mapping is:

1) Initialize m'=0 (resource-element group number)

2) Initialize k' =0

3) Initialize l' = 0

4) If the resource element m' represents a resource-element group and the resource-element group is not assigned to PCFICH or PHICH then perform step 5 and 6, else go to step 7

5) Map symbol-quadruplet w(m') to the resource-element group represented by (k',l') for each antenna port p

6) Increase m' by 1

7) Increase l' by 1

8) Repeat from step 4 if , where l' < L, where L corresponds to the number of OFDM symbols used for PDCCH transmission as indicated by the sequence transmitted on the PCFICH

9) Increase k' by 1

10) Repeat from step 3 if k' < num-dl-rb*12

Reference Signals in LTE

2011-11-05T21:46:00.000-07:00

Downlink Reference Signals

LTE provides for reference signals to support Uplink and Downlink transmission. Reference signals occupy pre-defined Resource Elements on the OFDMA Resource Grid. These do not carry any user data and only facilitate transmission of user data.

The transmission on the reference signal is used for channel estimation. Since the sequence that is transmitted on the downlink reference signal is known by the UE, it can compute the Channel Response for that Resource Element. The Channel Response for the other Resource Elements can then be approximated.

The hopping of the reference signals on Resouce Elements based on a known Pseudo Random Sequence is being considered. This will better approximate the ever changing radio conditions.

UE also utilises the reference signals for CQI measurement and reporting.

Three types of downlink reference signals are defined:

- Cell-specific reference signals, associated with non-MBSFN transmission

- MBSFN reference signals, associated with MBSFN transmission

- UE-specific reference signals

There is one reference signal transmitted per downlink antenna port.

Cell-specific reference signals

Cell-specific reference signals shall be transmitted in all downlink subframes in a cell supporting non-MBSFN transmission. In case the subframe is used for transmission with MBSFN, only the first two OFDM symbols in a subframe can be used for transmission of cell-specific reference symbols.
Cell-specific reference signals are transmitted on one or several of antenna ports 0 to 3.
Cell-specific reference signals are defined for sub-carrier separation of 15 KHz only.
The resource elements which are used for the downlink reference signal on one ports are left unused for transmission on other antenna ports. This reduces the interference and helps the UE to decode the reference signals better.

Mapping of downlink reference signals (normal cyclic prefix).

MBSFN reference signals

MBSFN reference signals shall only be transmitted in subframes allocated for MBSFN transmissions. MBSFN reference signals are transmitted on antenna port 4.
MBSFN reference signals are defined for extended cyclic prefix only.

Mapping of MBSFN reference signals (extended cyclic prefix, carrier spacing 15 KHz)

Mapping of MBSFN reference signals (extended cyclic prefix, carrier spacing 7.5 KHz)

UE-specific reference signals

UE-specific reference signals are supported for single-antenna-port transmission of PDSCH and are transmitted on antenna port 5.
UE specific reference signals are present and are a valid reference for PDSCH demodulation only if the PDSCH transmission is associated with the corresponding antenna port.
UE-specific reference signals are transmitted only on the resource blocks upon which the corresponding PDSCH is mapped.

Mapping of UE-specific reference signals (normal cyclic prefix)

LTE Broadcast Channel

2011-11-05T14:19:00.000-07:00

LTE Broadcast Channel

The LTE Broadcast Channel is carried on the PBCH and PDSCH. The PBCH carries the Master Information Block of the BCH. The UE reads the MIB at the time of cell (re)selection. The SIBs are carried on the PDSCH and are read by the UE when required.

The information carried in the MIB is crucial in the initial cell selection process, and consists of:

DL Bandwidth (3 bits) – tells the System bandwidth (1.4MHz …. 20 MHz)
PHICH Configuration (3 bits)
System Frame Number (8 bits)
Spare bits (10 bits)

The MIB also indirectly communicates one of the three Antenna Configuration in use in the DL. The CRC of the MIB data transmitted undergoes a special mask to encode this information. The UE does trial and error method to determine the mask applied and thus knows the DL antenna configuration in use.

Mapping of PBCH onto the OFDM Resource Grid

The Physical BCH channel is restricted to the 72 sub carriers about the DC in the resource grid.
The PBCH is transmitted in the first four OFDMA symbols of the 1^st slot of each radio frame.
The PBCH uses QPSK modulation
The scrambling code of PBCH is a cell specific code determined by the UE from the P-SCH/S-SCH.
The TTI of PBCH is 40 ms, i.e. the UE has to decode PBCH in 4 consecutive radio frames to get the information transmitted in the MIB.
The Downlink Reference Signals override the PBCH in Resource Elements reserved for them.
The PBCH is transmitted on each antenna port.

LTE Physical Broadcast Channel

LTE Synchronization Channels and Cell Search

2011-11-05T12:29:00.000-07:00

LTE Synchronization Channels

LTE, like UMTS , follows a hierarchical approach to cell search. This makes the process easier for the UE.

There are two Synchronisation channels defined for cell search procedure. The two channels carry the information about cell ID between them. After the UE has obtained the information from these two synchronization channels, it has the Physical Cell ID of the cell and is Radio Frame, Subframe and Slot aligned with the cell.

Primary Synchronisation channel (P-SCH) carries the primary code with value 0-2. This indicates the cell ID within a cell ID group. The sequence to be transmitted on the primary synchronization signal corresponding to one of the cell-Ids is generated from a frequency-domain Zadoff-Chu sequence, which results in 3 root indexes, one each for the 3 cell-Ids.

Secondary Synchronisation channel (S-SCH) carries the secondary code with value 0-167. This indicates the cell ID group, one from 168 possible groups.

The above arrangement allows for 168x3 = 504 unique Physical cell IDs. The UE computes the Physical Cell ID as follows:

Physical Cell ID = 3 x (cell-ID group) + cell-ID

Mapping of P-SCH and S-SCH onto the OFDM Resource Grid

The Synchronisation channels are restricted to the 62 sub carriers about the DC in the resource grid. This is done to allow the UE to do cell search with FFT size of 64, which is much easier for the UE. The position of the P-SCH and S-SCH on the OFDMA grid is shown in the figure below. This figure is for LTE FDD (Frame Structure 1 with 7 OFDMA symbols/slot).

LTE Synchromisation Channels in the OFDMA Resource grid

The following points are shown in the figure above:

The P-SCH is transmitted in the last OFDMA symbol of the 0th and 11th slot i.e. in the 1st and 5th sub frame.
The S-SCH is transmitted in the second last OFDMA symbol of the 1st and 11th slot i.e. in the 1st and 5th sub frame.
In other words, P-SCH occupies the last symbol of the slot=0 and slot=11, and S-SCH immediatel precedes it on the second last symbol of the slot=0 and slot=11.
Both P-SCH and S-SCH are transmitted on 62 subcarriers centered about the DC. This is irrespective of the DL bandwidth of the LTE system. The DL bandwidth in use is communicated on the Broadcast Channel(PBCH) inside the MIB.
The periodicity of P-SCH and S-SCH is 5ms.
Since the position of the P-SCH and S-SCH is fixed in the sub frame and slot structure of the LTE Radio Frame, the UE gets aligned to the slot level after reading the P-SCH and S-SCH

The cell search procedure tells the UE the Physical cel ID of the cell. The next step is reading the master Information Block (MIB) on the Physical Broadcast Channel (PBCH).

Characteristics of LTE Physical layer

2011-10-29T19:01:00.000-07:00

The table below provides some of the specifications of the LTE Physical Layer.

Transmission BW		1.25 MHz	2.5MHz	5MHz	10 MHz	15 MHz	20 MHz
Slot Duration		0.5 ms
Sub Carrier Spacing		15 KHz
Sampling Freq		1.92 MHz	3.84 MHz	7.68 MHz	15.36 MHz	23.04 MHz	30.72 MHz
FFT Size		128	256	512	1024	1536	2048
CP Length usec	Short	5.21x1, 4.69/9 x 6	5.21x1, 4.69/9 x 6	5.21x1, 4.69/9 x 6	5.21x1, 4.69/9 x 6	5.21x1, 4.69/9 x 6	5.21x1, 4.69/9 x 6
CP Length usec	Long	16.67	16.67	16.67	16.67	16.67	16.67

Transmission BW: LTE supports varying bandwidth from 1.25 MHz to 20 MHz, allowing the operators to utilize BW chunks available. The operator can use chunks of 1.25 MHz at different frequencies rather than a 5 mHz(which is required in UMTS).

The Radio Frame of 10ms is divided into 10 sub frames of 1 ms each. Each sub frame has two slots of 0.5ms each.

Sub carrier spacing: It is difference in the center frequency of two adjacent sub carriers.

T-FFT: This is one FFT period and is equal to 66.667 us.

FFT size: This is the size of the FFT employed in the IFFT and FFT.

Sampling Frequency: With the FFT size of 128, this means 128 samples are taken in each FFT period. Therefore, sampling frequency = 128/66.667 us = 1.92 MHz. Note that this is half of 3.84 MHz, which is the chip rate in UMTs). Sampling frequency for other BW sizes can be calculated likewise.

Cyclic Prefix Length: Length of the prefix attached to each OFDM symbol before transmission is 4.69 us or 16.67 us depending on Short/extended prefix in use.

Ts: This is the unit of time in LTE and is equal to 1/(1500x2048) = 0.52086 us. Every time duration is expressed in terms of Ts.

LTE X2 (inter eNB) Handover

2011-10-23T19:28:00.000-07:00

The LTE standards come with well defined interface between eNB(X2), which makes handover between eNB possible.

LTE X2-based Handover

The steps of the handover as in the figure above are described below:

(1) The old eNB upon receiving the appropriate measurement reports from the UE will initiate the Handover procedure by sending a HANDOVER_REQ to the new eNB.This requires intelligent eNB which are aware of the other eNB in the vicinity.

(2) The new eNB sends a HANDOVER_ACK. This will contain a RRC_CONNECTION_RECONFIG message containing the parameters of the new cell for the UE.It will also have the GTP-U TEID for the data connection for DL data to be sent from the old eNB.

(3) The old eNB relays the RRC_CONNECTION_RECONFIG message to the UE.

(4) At this point, the old eNB starts forwarding the DL data destined for the UE to the new eNB using the GTP-U TEID from step 2.

(5) The UE does a Random Access Procedure in the new cell under new eNB to get the Timing Advance and UL grant.

(6) UE sends the RRC_CONNECTION_RECONFIG_COMPLETE message as a handover confirmation to the new eNB.

(7) The new eNB will now ask the MME to switch the GTP-U tunnel for DL data connection from the old eNB to the new eNB by sending the PATH_SWITCH_REQ message which has the new GTP-U TEID.

(8) The MME will in turn send the UPDATE_UP_REQ having the GTP-U TEID from the last step to S-GW.

(9) The UL and DL data starts in the new eNB. The new eNB however has to buffer the UL data till the data connection to the S-GW is switched completely.

(10) (11)The old eNB forwards the last packet to new eNB before the swicth.

(12) The S-GW will confirm the switch by sending UPDATE_UP_RESP to the MME.

(13) MME in turn confirms the switch to the new eNB with PATH_SWITCH_ACK.

(14) Now that the new eNB has completely taken over the role of the eNB serving the UE, it sends a class 2 one-sided message (UE_RESOURCE_REL) to the old eNB.

(15) The data on new data link continues.

CSoHSPA : CS Voice over HSPA

2011-10-18T22:04:00.000-07:00

The new features added in Rel7 for HSPA have made CS voice to be run over HSPA possible. It would have been possible to map CS voice over HSDPA and HSUPA in Rel6, but the resources at both the UE and the network would have been wasted, and it was not commercially viable.

The implementation till Rel 6 was to have a HSPA bearer for PS traffic and establish DCH bearer for CS voice. PS only traffic would be mapped to HSDPA in DL and HSUPA in UL and at the onset of CS call, everything would be moved to DCH. And when the CS call is disconnected, there would be radioBearerReconfiguration to move the PS to HSPA again.

The new features in UTRAN that have made this possible are:

Continuous Packet Connectivity: Allows the UE to be in CELL_DCH state for most of the time. UE can be configured with DRX cycles in this state and the battery resources conserved. HS-SCCH less operation on HS-DSCH channel saves DL codes. The HS-DSCH is switched to operation with HS-SCCH channel when the required data rate increases. With CPC, everything is mapped to HSPA, there is no DPCCH or DPDCH channel in operation in the DL.

enhanced Fractional-DPCH: This feature allows upto 7 UEs to be multiplexed on one DPCCH channel in the Downlink(for the calculation read enhanced F-DPCH). This F-DPCH channel carries TPC commands for the UE and is required along with HSDPA assignment in the downlink. Therefore, to have for example 35 active simultaneous voice calls, 5 such codes are required.

Non-Scheduled transmission on HSUPA: voice calls required guaranteed data rate(depending on codec, like 12.2 kbps). So, non scheduled grant will be used to transmit the voice packets in Uplink. Network allocates non-scheduled grant to the UE and makes provision in the UL resources (power transmitted by the UE) for the combined non scheduled grants for all the UEs in a cell. To read more about Non scheduled transmission go the MAC Scheduling paper.

Guaranteed QoS by the MAC-(e)hs Scheduler: MAC scheduler will have to be more robust in scheduling the voice packets on a priority basis and guarantee the codec rate in downlink. MAC Scheduling paper.

The case for HSPA+

2011-10-17T11:00:00.001-07:00

GSM has been around from as long as the mobile phone has been in use. It was well suited for voice only and there are still new GSM networks being put in places where the predominant traffic is voice traffic. EDGE was introduced as a GSM evolution to cater to data traffic. Later HSDPA and HSUPA revolutionised the data traffic by allowing high speeds in Uplink and Downlink.
But as the need for higher data rates is increasing, LTE is providing a viable option to the operators. Even higher data rates are being researched upon in LTE Advanced.
Among all this HSPA is again being revolutionised with the HSPA+ technology in 3GPP Rel 7 and beyond.
HSPA+ has promised 4G speeds and has new features that look great to both the network operators and the UE manufacturers.

With only a few LTE enable UEs available in selected markets, LTE has still a long way to go before there can be widespread deployment. Therefore HSPA will be around for quite some time and HSPA+ will match it with LTE for the data rates. This means HSPA+ and LTE will coexist for a long time until cheap LTE UEs are in the market and network vendors have their networks ready for LTE.

HSPA+ Features

	New Feature	Benefits
Rel 7	Higher Order Modulation in DL – 64 QAM	DL speed - 21 mbps
	Higher Order Modulation in UL – 16 QAM	UL speed – 11.5 mbps
	MIMO + 16 QAM	DL speed with 2x2 + 16 QAM - 42
	CPC	- Huge battery saving in CELL_DCH - reduced call setup time - always ON UE
	eFACH	HSDPA data possible in CELL_FACH
	Enhanced F-DPCH	- DL codes saved for signaling - Increased cell capacity in the cell
	E-DPCCH power boosting	Allows network to configure UE with parameters to boost the E-DPCCH w.r.t. DPCCH when the data rate threshold is reached.
Rel 8	MIMO + 64 QAM	DL speed with 64 QAM – 28
	DL Dual Cell	DL speed with 64 QAM – 42
	eFACH	HSDPA + HSUPA data possible in CELL_FACH
	CSoHSPA	- Made possible by enhanced F-DPCH, CPC - CS traffic mapped to HSDPA/HSUPA - cell capacity increased tremendously - network has to manage only HSPA traffic
Rel 9	DL Dual Cell + MIMO	DL speed with (2x2 MIMO + 64 QAM) in each cell – 84 mbps
Rel 9	UL Dual Cell	Dual cell with 16 QAM – 23 mbps

Comparison of LTE Security and UMTS Security

2011-10-15T15:12:00.000-07:00

LTE has inherited most of the Security architecture from UMTS, some enhancements have also been made. While there are lot of places to find the overview of the LTE Security architecture, a side-by-side comparison with UMTS will be helpful for peopple coming from UMTS background.

Security Elements

Authentication and Key Agreement

UMTS

based on UMTS-AKA

key derivation from the UMTS Authentication Quintuplet(RAND, CK, IK, RES, AUTN)

LTE

based on EPS-AKA

key derivation from the UMTS Authentication Quintuplet(RAND, CK, IK, RES, AUTN)
Key Derivation Functions use HMAC-SHA-256
K(ASME) computed from CK, IK, and AUTN, which is used for Integrity and Encryption keys

Integrity

UMTS

Integrity protection mandatory for only few RRC messages.
Integrity Protection Algorithm

UIA1: Kasumi

LTE

Integrity Protection mandatory for all messages after (and including) Security Mode Command

Integrity Protection Algorithm

128-EIA 1: based on SNOW 3G
128-EIA 2: based on AES-128

Encryption

UMTS

Encryption Algorithm

UEA0: no encryption.
UEA1: Kasumi.

LTE

Encrption done independently at two levels

NAS - for EMM and ESM messgaes
PDCP-SRB(1 and 2) and DRB (1 .. 11)

two SECURITY MODE COMMANDS for two sets of keys

{K(NAS-enc), K(NAS-integrity)}
{K(RRC-enc), K(RRC-integrity)}

Encryption Algorithm

128-EEA 0: No Encrption
128-EEA 1: based on SNOW 3G
128-EEA 2: based on AES-128

Ec/No description

2011-10-06T20:59:00.000-07:00

Abbreviation for received Energy per Chip / power density in the band

The Ec (RSCP)/ No (RSSI - total receive power) is the received energy per chip divided by the power density in the band. No includes the power of specified cell as it indicates total receive power (as a result, Ec/No is deteriorated by increasing 'No')

Ec/No (dB) = RSCP(db) - RSSI (db)

Since, RSSI > RSCP, therefore Ec/No is -ve.

LTE Security Architecture

2011-10-06T17:20:00.001-07:00

This is a paper from NOKIA about LTE Security Architecture

Download Here

LTE Random Access Procedure

2011-10-06T11:46:00.000-07:00

Highlights of Random Access Procedure in LTE:

Contention based access and Non-contention based
Function of the MAC Layer
Provide Timing Advance for UL transmission to the UE
Resource Assignment for transmission of AS message on UL-SCH

Contention Resolution:

To avoid two or more UEs trying to access the RACH channel, the LTE design provides upto 64 different RACH configurations which the UEs in a cell can use. The configuration ID to be used to transmit the RACH Preamble is determined by the higher layer. The algorithm which computes the configuration index to be used in upcoming RACH transmission should be random enough to minimise the chances of collision.

Timing Advance for UL transmission to the UE

The BCH provides the Frame and Slot synchronisation but does not provide the round trip delay. The UE must know how much early it should transmit in UL to compensate the trip delay, so that the transmission arrives at the slot boundary at eNB. This timing advance is provided by the eNB in the Random Access Response.
The UE must do a Random Access when in RRC Connected Mode, the time for synchronisation expires.

Resource Assignment for transmission of AS message on UL-SCH

The Random Access Response message provides the UL-grant for transmission of the AS message. The UE utilises the UL-grant to send a AS message.

Random Access Procedure - Contention based

Figure: Contention based RA Procedure

(1) UE selects a P-RACH configuration ID(out of 64) to be used in P-RACH transmission of the RA Preamble. The RA Preamble is computed and transmitted for the corresponding RA-RNTI.
If there is no response from the network, the UE re attempts with higher transmission power.
Note: the Random Access RNTI(RA-RNTI) is used by the UE to identify the DL-SCH transmission destined for it on the DL-SCH channel.

(2) eNB sends a Radom Access Response on the DL-SCH, using the RA-RNTI sent by the UE earlier. It sends the Timing Advance Info, UL grant for the UE on UL-SCH and a temporary C-RNTI. temp C-RNTI is used by the UE for UL-SCH scheduled tx.

(3) The UE uses the UL grant on UL-SCH to transmit the Access Stratum Message(e.g. RRC Connection Setup Request) along with the temp C-RNTI to identify itself.
The UE can indicate if it requires more resources in the UL using the MAC-Buffer Status Report(BSR).

(4) The eNB sends the Contention Resolution message on the DL-SCH using the temp C-RNTI. This message contains the C-RNTI for the UE. After the UE receives the contention Reslution, it replaces temp-CRNTI with this CRNTI and uses it for all future UL transmissions.

The contention based RA Procedure is used for:

transition from RRC_IDLE to RRC Connected(RRC Connection Establishment)
in RRC_CONNECTED when UL data has to be sent and UL Synchronisation state is “non-synchronised”

Note: the UL synchronisation state is changed to non-synchronised after the expiry of the TimeAlignmentTimer. At this point a UL Time Alignment must be done using RA Procedure.

Random Access Procedure - Non Contention based
The non-contention based Random Access procedure is initiated by the network.

Figure: Non-Contention based RA Procedure

(1) The network sends the RA Preamble Assignment Info to the UE on DL-SCH(using the C-RNTI)
(2) UE sends the assgined RA Preamble on the assigned P-RACH configuration ID
(3) Network sends the Random Access Response with Timing Advance Info.

The non-contention based RA Procedure is used by the network for:

When timing advance is required for UE positioning.
in RRC_CONNECTED when DL data has to be sent and RA Procedure is required.
Inter system or intra LTE Handover