Why TD-LTE ?

Operators are spending millions of dollars just on the license only. It is expected that TDD spectrum will be sold at lower price as compared to FDD spectrum and secondly the TDD spectrum is also available in many countries. So certainly there is emerging interest in TD-LTE. Some of the main advantages of TD-LTE over FDD-LTE are as follows

In FDD LTE (Frequency division duplex)

  • duplexing scheme requires paired spectrum which in other words means that the downlink and uplink transmission should be on different frequency spectrum
  • There is wastage of frequency resources in up-link if most UEs in the cells are using downlink spectrum most the time or vice versa (For example if most users in he cells are watching YouTube or downloading some files)
  • Since different range of frequencies are used in uplink and downlink, therefore channel characteristics are different in both directions. UE needs to report the downlink channel characteristics to the eNB on its uplink channels
  • Additional hardware required on UE and eNB side to separate uplink and downlink frequencies which adds extra cost to the terminals and base stations
Now in TD-LTE (Time division duplex) ? 

  • Paired spectrum is not required and both downlink / uplink communication occurs on single frequency channel
  • Channel characteristics are same for both uplink and downlink channels
  • Hardware cost is low, no need of diplexer in UE and eNB
  • Spectrum is efficiently utilized since UEs can use all the slots for downlink traffic if there is no uplink traffic

TD-LTE frame structure is shown in the figure below
Subframe 0 and 5 carries synchronization signals and system information blocks
Subframe 1 is special subframe that is used to carry information for switching between uplink and downlink for 10ms switching period
For 5 ms switching period both subframe 1 and 6 are used



ICIC and eICIC

Background

  • LTE is designed for frequency reuse 1 (To maximize spectrum efficiency), which means that all the neighbor cells are using same frequency channels and therefore there is no cell-planning to deal with the interference issues
  • There is a high probability that a resource block scheduled to cell edge user, is also being transmitted by neighbor cell, resulting in high interference, eventually low throughput or call drops (see figure) 
  • Traffic channel can sustain upto 10% of BLER in low SINR but control channels cannot. Neighbor interference can result in radio link failures at cell edge.
  • Heterogeneous networks require some sort of interference mitigation, since pico-cells/femto cells and macro-cells are overlapping in many scenarios


ICIC (Inter-cell interference coordination)

  • Inter-cell interference coordination is introduced in 3GPP release 8
  • ICIC is introduced to deal with interference issues at cell-edge
  • ICIC mitigates interference on traffic channels only
  • ICIC uses power and frequency domain to mitigate cell-edge interference from neighbor cells 
  • One scheme of ICIC is where neighbor eNBs use different sets of resource blocks through out the cell at given time i.e. no two neighbor eNBs will use same resource assignments for their UEs. This greatly improves cell-edge SINR. The disadvantage is decrease in throughput throughout the cell, since full resources blocks are not being utilized.
  • In the second scheme, all eNBs utilize complete range of resource blocks for centrally located users but for cell-edge users, no two neighbor eNBs uses the same set of resource blocks at give time
  • In the third scheme (probably the preferred scheme), all the neighbor eNBs use different power schemes across the spectrum while resource block assignment can be according to second scheme explained above. For example, eNB can use power boost for cell edge users with specific set of resources (not used by neighbors), while keeping low signal power for center users with availability of all resource blocks (see the figure)
  • X2 interface is used to share the information between the eNBs

eICIC (enhanced Inter-cell interference coordination)

  • eICIC introduced in 3GPP release 10
  • eICIC introduced to deal interference issues in Heterogeneous Networks (HetNet)
  • eICIC mitigates interference on traffic and control channels
  • eICIC uses power, frequency and also time domain to mitigate intra-frequency interference in heterogeneous networks
  • eICIC introduces concept of "Almost blank subframe" (ABS). ABS subframes do not send any traffic channels and are mostly control channel frames with very low power. If macro cell configure ABS subframes then UEs connected to pico/femto cells can send their data during such ABS frames and avoid interference from macro cell (see the figure)
  • ABS configuration is shared via OAM or x2 interface

Master Information Block (MIB) in LTE


The very first step for UE to gain initial access to the network after completing initial cell synchronization is to read the Master information block (MIB) on BCCH (Logical channel), BCH (Transport channel) and PBCH (Physical channel). Resource elements used by MIB are the first 4 OFDMA symbols of second slot of first subframe of a radio frame. On frequency domain it occupies 72 subcarriers. MIB carries very little but most important information for UE initial access. The content of MIB includes

·         Downlink channel bandwidth in term of resource blocks (RBs)
·         PHICH configuration (PHICH duration and PHICH resource)
·         System Frame Number

New MIB is broadcasted every radio frame for which SFN mod 4 = 0 (40ms repetition) while its copies are broadcasted in the middle 10ms radio frames as shown in the figure below



Why PHICH is carried by MIB? 
After decoding MIB, UE has to decode PDCCH to read other system information blocks (SIBs).  PDCCH, PHICH and PCFICH share the resources in the control region of a subframe. So to find the available resources for PDCCH, UE has to know the PHICH configuration only, as PCFICH resources are fixed and known

Semi persistent scheduling

Every VoIP packet is received / sent every 20ms when the user is talking whereas in silence period, discontinuous transmission (DTX) is used to reduce the transmission rate. Also, in order to sustain voice quality, silent insertion descriptor (SID) packet arrives every 160ms. The frequent arrival/transmission of VoIP packet means large control overhead for lower layers (L1/L2) in the radio protocol stack. To deal with this issue, semi persistent scheduling plays an important role.

Scheduling is a mechanism where UE requests eNB for the resource allocation during each transmission time interval (TTI). If UE has some data that it needs to transmit continuously, it will request eNB every TTI for the resource allocation. This scheduling type is dynamic scheduling. The advantage of dynamic scheduling is flexibility and diversity of resource allocation but as mentioned, this results in huge L1/L2 load which in turn means inefficient use of scarce radio resources.

In case of semi persistent scheduling, eNB can assign predefined chunk of radio resources for VoIP users with interval of 20ms. Therefore, UE is not required to request resources each TTI, saving control plan overhead. This scheduling is semi-persistent in the sense that eNB can change the resource allocation type or location if required for link adaptation or other factors.

Primary and secondary synchronization signals (PSS & SSS) in LTE


Cell synchronization is the very first step when UE wants to camp on any cell. From this, UE acquires physical cell id (PCI), time slot and frame synchronization, which will enable UE to read system information blocks from a particular network.

UE will tune it radio turn by turning to different frequency channels depending upon which bands it is 
supporting. Assuming that it is currently tuned to a specific band / channel, UE first finds the primary 
synchronization signal (PSS) which is located in the last OFDM symbol of first time slot of the first subframe (subframe 0) of radio frame as shown in figure (green squares). This enables UE to be synchronized on subframe level. The PSS is repeated in subframe 5 which means UE is synchronized on 5ms basis since each subframe is 1ms. From PSS, UE is also able to obtain physical layer identity (0 to 2).

In the next step UE finds the secondary synchronization signal (SSS). SSS symbols are also located in the same subframe of PSS but in the symbol before PSS as shown in the figure(yellow squares). From SSS, UE is able to obtain physical layer cell identity group number (0 to 167).

Using physical layer identity and cell identity group number, UE knows the PCI for this cell now. In LTE 504 physical layer cell identities (PCI) are allowed and are divided into unique 168 cell layer identity groups where each group consist of three physical layer identity. As mentioned earlier, UE detects physical layer identity from PSS and physical layer cell identity group from SSS. Assuming physical layer identity = 1 and cell identity group=2 then the PCI for given cell is

PCI = 3*(Physical layer cell identity group)+ physical layer identity = 3*2+1 = 7

Once UE knows the PCI for a given cell, it also knows the location of cell Reference signals as shown in figure (red and black squares). Reference signals are used in channel estimation, cell selection / reselection and handover procedures. 



LTE Resource Grid (Source: http://paul.wad.homepage.dk/LTE/lte_resource_grid.html)

After cell synchronization procedure, UE will proceed to read Master information and other System information blocks
Please see below sections
Master information block (MIB)
System information block 1 (SIB1)
System information block 2 (SIB2)


Robust Header Compression (RoHC)

For VoIP packets, the size of headers (IP/UDP/RTP) is usually larger than the data itself. Compressing these protocol headers over end to end (User to user) connection is not feasible since the headers play important role, but there is possibility to compress them over the air interface i.e. between UE and eNB/NB/BTS etc. Over the air interface, protocol headers do not play any role, so therefore there is considerable reduction in the data over head on radio interface. For IPv4, UDP and RTP, the amount of overhead due to headers is 40 bytes while RoHC can compress this to 2 or 3 bytes. RoHC is described in RFC 3095



In LTE, the role of Packet data convergence protocol layer (PDCP) is to implement RoHC as shown below (Only uplink traffic shown). 
RoHC is applied to user plane traffic only while this feature is transparent to control plane traffic.



Use of RoHC with conversational bearer (QCI 1) in LTE will result in efficient use of radio resources since it can reduce the user plane traffic by several bytes over air interface.