=====WARNING: THIS POST IS INCOMPLETE &UNREVIEWED..I AM IN THE PROCESS OF MAKING IT COMPLETE AND READABLE===========
What is the need for LTE technology /What are the major features of LTE?
Higher bandwidth ( 300 Mbps DownLink, 75 Mbps UL)
All IP network, leading to lower operating costs
Lesser Number of nodes thereby reducing CAPEX.
Lower call setup delay time ( ~100 ms)
Lower user plane latency (~10 ms rtt)
Spectral Efficiency: 3 to 4 times compared with Rel6 in DL & 2 to 3 times compared with R6 in UL.
Mobility:Highly optimised for 0-15 kmph;Between 15 & 120, high performance, should support mobility till 500 kmph
Coverage:Targeted performance for 5 km radius cells, with slight degradation 30 km should be supported; upto 100 km support is required
No circuit switched domain/ Everything is in PS.
Architecture of LTE Network:
Ref:
http://www.ltesimplified.blogspot.com/2012/03/lte-architecture.html
LTE/E-UTRAN + EPC/SAE ==> EPS
UE:-User Equipment. The device with which the user accesses LTE network.
eNodeb:-Evolved Nodeb, contact point for the UE in LTE network.Does RRM and Scheduling.Has the entire access stratum implementation on the network side.
MME:- Control entity in the core network; Functionalities related to connection management and mobility management.Registration,detach,pdn connectivity,disconnect,activation of beareres etc. Selection of S-GW is also done by MME. It interacts with SGSN while 3gpp handovers. Selection of different MME or SGSN is done by MME during handover. Security is also the responsibility of MME.When the UE attaches to network, a SAE-TMSI (S-TMSI) is allocated to the UE.UE context which is addressed by S-TMSI is stored in MME, MME downloads the subscription profile from HSS and stores it. When UE moves to ECM IDLE state also this is kept.. Interface between MME & SGW is S11, with SGSN it is S3, with HSS it is S4, With eNB it is S1-MME
SGW:-User Plane towards the eNB is directed by the SGW; Packet router marking the DiffServ for maintainig QOS. Does mobility anchoring for handovers within 3gpp including LTE-LTE handovers
Once a handover is executed successfully and the associated UE has left the S-GW, the old S-GW will send one or more “end marker” packets to the source eNB, source SGSN, or sourceRNC of the handover to assist the reordering of user plane packets in these network elements.
For inter-3GPP mobility, S-GW acts as
the terminating point of the S4 interface (and routes
the traffic between the 2G/3G SGSN and the PDN-GW of the EPC). ie
the S-GW acts as the GGSN. Keeps the S5/S8 bearers alive even when rrc idle,thus when a paging happens buffers the packets.
SGW interfaces with PGW using S5/S8, eNB S1-U,S12 with RNC/SGSN
PGW: EPS Gateway that interfaces with the external network.Allocates the IP address to the UE,Does mobility anchoring during handover with non 3gpp networks.Usually has a policy control enforcement function, that determines the flows based on the qos policies. Does the filtering of users IP packets to different QOS based bearers. as per the TFT. Qos Enforcement for GBR bearers. During roaming, can connect to eith local/home PGW. Gx with PCRF, SGi with operator services/packet network.
HSS:Home Subscriber System. holds customer information like the subscription profile. It also contain dynamic information like MME to which UE is connected, APNs connected etc.
PCRF: Policy Charging and Rules Function, provides the QOS Authorisation (QCI & bit rates) and ensures that it is in accordance with the users subscription profile.
What is a bearer?
A bearer is an IP flow between the Gateway and the UE with a defined QoS.E-UTRAN and UE together setup and release bearers and required by the applications.
What are the states RRC-IDLE,RRC-CONNECTED? EMM-REGISTERED/DEREGISTERED? ECM-CONNECTED/IDLE?
In RRC Connected state, UE is known at cell level and there is a context available in eNB. UE can directly reached by the network. It will always check for scheduling assignment in the control channel.
In ECM connected state, there is a signalling connection between EPC and the UE. It happens after the UE is in RRC Connected state and UE activated a PDN Connection.There is an S1 connection present.
When UE goes to ECM IDLE,EPS bearer contexts are still available(may be out of sync). ie contexts are still maintained in P-GW. When rrc connection is terminated, UE moves to ECM IDLE as well.
EMM Registered state:UE goes to EMM registered when the UE attaches to the network. UE is known to MME in registered state and is not known in emm deregistered state.
What is S1-Flex?
Multiple MMEs/SGWs control the same geographical area, being connected to a mesh of eNodebs. This is called MME/SGW pool and the area is called a pool area. During attach procedure, eNB selects the MME to which the UE will attach.
Describe the LTE protocol stack:
L1-L1 is based on OFDMA in downlink and SC-FDMA in Uplink. Layer1 is concerned with actual transmission of the broadband signal.
MAC:- MAC layer is mainly concerned with the transmission of pdus, scheduling and dynamic resource allocation.
RLC:- RLC is responsible for ensuring the QOS, by flow control, error detection and retransmissions.
PDCP:Packet Data Convergence protocol is responsible for header compression, security (ciphering and integrity) and storing and forwarding of the packets during handover.
RRC:-RRC is used for assigning/modifying/releasing the bearers,configuring and reporting of measurements, handovers,security,configuring the lower layers,direct transfer of NAS messages,paging etc.
NAS LAYERS:
eMM: eMM is concerned with the mobility procedures- ie attach, detach and tracking area update procedures.
eSM:-eSM is concerned with the EPS bearers and PDN connectivity. Main procedures supported are the activation of default and dedicated (EPS)bearers,bearer resource modification,pdn connectivity/dsiconnect.
[Figure TBA]
What is the difference between PDN Connection and EPS Bearer?
A PDN Connection is the connection between a UE & PDN GW. An EPS bearer is a PDN connection + a defined QOS (QCI)+Aggregate Guaranteed and Maximum Bit rate (optional) + TFT
There can be multiple EPS Bearers associated with a PDN connection.
What is a TFT?
Traffic Flow Template is the set of all packet filter associated with an EPS bearer.A packet filter may be associated with a protocol or an ip address or port address.
What are the identities in LTE world?
Two of the identities from UMTS/GSM world are still used in LTE world
1. IMSI which uniquely identifies the subscriber or specifically SIM card. IMSI consists of MCC+MNC+MSIN
2.IMEI which identifies the device. IMEI has a TAC which identifies the device manufacturer and a SN which identifies the device.It may have a version number too
In addition to that the main identities are
1.GUTI :- Similar to TMSI in UMTS, GUTI-Global Unique Temporary Identifier is a temporary identifiier.GUTI will be allocated during attach procedure and will be used for identifying the UE thereafter
GUTI consists of MCC+MNC+MMEID(MME-GROUP & MME-Code)+M-TMSI
MME-Code and M-TMSI together is called S-TMSI and is used in paging.
2.Radio Network Temporary Identifiers (RNTIs) : RNTIs are similar to the ones in HSPA. Since shared channels are used for data transfer in LTE, we need some identifier to say this data belongs to this particular UE. RNTIs are for that purpose.
Various RNTIs are present,
C-RNTI:-
RA-RNTI :- RA response will contain the RA-RNTI
P-RNTI :- To a group of UEs, derived from the IMSI
SI-RNTI:- Indicates which resource blocks carry SIBs.
3.TAI: Identifies Tracking Area; A tracking area is a collection of one or more E-UTRAN cells. How they are mapped depends on proprietery radio planning. One cell can be part of multiple TAs as well. Paging happens at TA level
4.E-UTRAN CGI- Cell Global Identity consists of PLMN id (MCC+MNC) and CI (Cell Id is a 28 bit string, leftmost bits uniquely identify eNodeb) Thus CGI uniquely identifies the E-UTRAN cell.
=====================================================================
Define QOS Architecture?
Ref LTE Pictures.
There can be 256 individual E-RABs. Default bearers with QCI 9, dedicated bearers will have other values like 5 for IMS (non conversional video)
Describe LTE Security?
Double security for NAS messages- ciphered and integrated by MME/UE as well.
RRC messages are ciphered and integrity protected.
Security Procedures:
1.UE sends the security capabilities in the initial NAS message (Attach)
2.MME sends the diameter authentication information request to HSS
3.HSS has the secret Key K, and from this it generates authentication vectors and sends to the MME in diameter authentication information response to HSS (KASME,AUTN,RAND,XRES)
4.MME generates NAS Encryption & Integrity key and Kenb from Kasme.
5.MME sends authentication request with autn and rand, ue checks the MAC part of AUTN to see if it is acceptable, if so it generates xres and sends to MME, MME verifies the response.
6.On succesful response, MME initiates NAS security mode command-specifying the security and integrity algorithms to be used. (optional)
7.MME will send the security context (
Kenb) with Nodeb as part of initial S1AP context setup message. eNB will generate ciphering and integrity keys for rrc and ciphering key for UP activate the rrc security mode (sends algorithms to use and the start parameter)
What is the multiple access mechanism in LTE?
The multiple access mechanism in LTE DL is OFDMA and SC-FDMA in UL. Botha are based on OFDM-Orthogonal Frequency Division Multiplexing.In OFDM, entire frequency band is divided into multiple orthogonal subcarriers. This eliminates the need for guard band. Orthogonality comes from IFFT.
In OFDM systems,all channel estimation,equalisation and interpretation are done in frequency domain, time domain is just for transmission.
=>LTE can have bandwidth between 1.4MHz to 20 MHz.Available bandwidth is divided into multiple subcarriers.
Reference subcarriers are used for equalisation of individual subcarriers.
=>To prevent intersymbol interference,a guard period called cyclic prefix is used.End of OFDM symbols are copied in front of the OFDM symbols to be transmitted. Two CPs are defined, normal CP and extended CP.
=>OFDM symbols show a high dynamic range after converting to time domain, which leads to high PAPR-which will lead to faster battery draining and complex RF receiver.So SCFDMA is used.In SCFDMA,DFT spread is used to overcome this.
Describe the scheduling in LTE?
In LTE, each subcarrier is 15 KHz. A modulated subcarrier is called resource element.
A group of 12 adjacent subcarriers is called a resource block.This is the smallest unit of scheduling in LTE.
A user can be allocated between 1 RB till the maximum bandwidth. If the user is allocated adjacent frequencies, it is called localised scheduling.If nonadjacent freq is allocated it is distributed scheduling.
DL=>Both Distributed and Localised. UL=>Only localised for PAPR reason
1 RB spans 1 scheduling period, 1 subframe which is 1 ms; 2 slots.
1 subframe has 14 ofdm symbols for normal CP,12 for extended CP
Localised and Distributed virtual resource block :- Distributed Virtual resource block,RB split into two, first slot equalent to physical RB.Second slot is hopped on another physical resource block.
Describe DL Slot Structure?
One radio frame is 10 ms which is divided into 10 subframes. Two types of frames, type1 in FDD, Type2 in TDD
First OFDM symbols in a frame is allocated by PDCCH, In area of PDCCH, embedded are PCFICH and PHICH. PDCCH allcoates between 1 to 3 symbols. So PDSCH will have between 11 to 13 with normal CP and 9 to 11 for extended CP. The number of symbols allocated are given in PCFICH.
In the middle, around DC subcarrier, 6 RBs are used for some common signals and channels in some subframes. Synchronisation signals are present in first and 6th subframe. PSS is transmitted on 7th OFDM symbol and SSS on 6th OFDM symbol of those subframes.62 out of 72 subcarriers are used for Synchronisation Signals. PBCH is present in the first subframe of all the frames in these 72 subcarriers.
What are the modulation and multiplexing schemes in LTE?
Modulation from QPSK to 64 QAM
Coding is mainly turbo coding.
Describe the DL scheduling?
Scheduling information is encoded in Downlink Control Information which is then mapped to REGs of PDCCH. DCI format 0 is for UL Scheduling Grant. From 1 to 2A, it is for DL scheduling assignment .
Other information like HARQ FBI, MCS and Power Control Commands are also present.
DCI format 3/3A is for UL TPC and is used when there is no data transmission between UE and nw.
UE has to decode the DCCH every ms for getting the scheduling assignment information. DRX is defined for battery saving, short and long DRX cycles are defined.
Scheduling can be localised or distributed. Semi Persistent Scheduling, configured by RRC also is present.
Describe UL scheduling?
Scheduling grants in DCI0, UE has to check DCCH scrambled with its RNTI. UE UL scheduler considers CQI,BSR,ACK/NACK and SR
SPS scheduling is allowed.No distributed scheduling.
Describe UL Slot structure?
Frame 10 ms, 20 slots,10 subframes as in DL. No synchronisation signals, Timing Allignment is used for synchronisation. PUCCH is on the upper and lower edges of system bw, remaining for PUSCH.
PRACH-6 consecutive RBs, which are defined in SIB2.
UL reference signals for channel estimation,send over entire frequency range.Send at the Middle of every slot. Independant of UE transmission on PUSCH/PUCCH, nw can request to transmit SRS on entire range or part of the frequency, for freq selective scheduling.
Channels in LTE:
==============
Down Link
=======
Physical Channels:
1.PDSCH
2.PDCCH,PCFICH,PHICH
3.PBCCH
4.PMCH
Transport Channels:
DL-SH,
BCH,PCH
Logical Channels:
DTCH,DCCH,CCCH,PCCH,BCCH
UpLink:
=======
Physical Channels:-
PUSCH,PUCCH,PRACH
Transport:- UL-SCH,RACH
Logical:-DTCH,DCCH,CCCH
[FTA:Channel Mapping]
Describe the initial UE Access procedures?
When an LTE phone is switched on, it first needs to identify the presence of an LTE cell. It will scan the entire frequency bands and will search for PSS. PSS gives the physical cell id group(0,1,2).This also gives the slot timing (and frequency synchronisation).Now it will decode SSS which is broadcast 1 slot prior to PSS. This gives physical cell id and frame synchronisation..It identifies if the system is FDD or TDD and CP length.Now it will decode BCH, First MIBs (PBCH) to get the system bandwidth,SFN and PHICH configuration and then the SIBs. No
w the UE does RACH procedure for accessing the network.
RACH procedure in LTE
RACH procedure in LTE is of 2 types -contention based(initial access,rrc reestablishment,ul data transfer) and non contention based(used during handovers & dl data arrival when rach resource are allocated by nw)
Contention Based RA procedure:
========================
UE selects a random access preamble and transmits in time and frequency resources of PRACH with initial RACH power settings from SIB. Now UE will listen to the scheduled RAR with a specific RA-RARNTI,matching the RACH time slot in which preamble request is send.If it doesnt get a response, power ramping procedure will happen.RAR will contain UL grant,TA,Temporary CRNTI. First scheduled UL transmission (msg 3 eg. rrcconnectionrequest) and msg4 which is contention resolution message(rrcconnection complete)
For non contention based, nodeb allocates rach resources and no need for cotention resolution.
