Carrier Aggregation in LTE- A (LTE Advanced)

Carrier Aggregation Introduction:

Carrier Aggregation is a feature which came into 3GPP Release 10. It is part of the feature of LTE- Advanced which is a common term used to describe the improvements in LTE. The focus of Release 10 was to advance LTE towards providing better user experience by providing higher data rates in cost-efficient way and in the same way maintaining backward compatibility.

The two main functionalities that are brought to achieve this criterion for LTE-A are Carrier Aggregation and Multi Antenna techniques.

What is Carrier Aggregation?

Carrier Aggregation is a feature in LTE for providing very high data volumes and data rates by the process of combining contiguous and non- contiguous spectrum bands. This can be explained in a simple way. For Ex: A Mobile operator can have 10MHZ in L1800 Band and 10 MHZ in L900 band. So, by using the functionality of CA we can combine these two bands to provide 20MHZ of Carrier Channel BW.

Few Pointers about CA:

  1. CA is a key feature which allows operators to create virtual carrier bandwidth
  2. It can help operators to increase individual carrier bandwidths for different layers by combining them by implementing the CA functionality
  3. CA enables combinations of up to 5 Carrier components. Multiple LTE carriers each with Bandwidth of 20MHZ can be transmitted in parallel to and from from the same UE.
  4. As of now CA can provides bandwidth up to 100 MHZ (20 MHZ * 5 Component carrier)
  5. Cross Component Carrier Scheduling is supported by CA, where we can use the control channel of one carrier (PCC) to allocate resources (scheduling functionality) of another carrier (SCC)

How the operator gains from Carrier Aggregation?

  1. The operator can provide increased download speed across the entire coverage area
  2. Utilization of higher capacity and spectral efficiency
  3. The Spectrum can be better utilized by pairing the high band downlink carrier frequency with a low band uplink carrier frequency
  4. There can be a use of supplementary downlink only band.
  5. The TDD spectrum can be effectively used by using FDD uplink frequency (We have a process of FDD- TDD carrier aggregation, will be explained in detail in coming sections)

   

fig : 1                                                                                                           fig : 2

Source: www.qualcom.com

Fig 1 explains a scenario where 5 carriers of 20 MHZ has been configured with Carrier Aggregation functionality to create a virtual bandwidth of 100 MHZ.  Fig 2 explains a scenario where a Pre REL 10 UE behavior without CA and when CA is implemented to a UE with 2CC (2 component carrier)

Types of Carrier Aggregation: There are two types of Carrier aggregation

  1. Intraband
  2. Interband

We will now see in detail about each of the CA types.

1.Intraband Contiguous: This Configuration refers to contiguous carriers aggregated in the same operating band. In this type of CA single Band is used (For Ex: L1800 or L900). The spacing between the center frequencies of the component carriers (continuously aggregated) is multiple of 300KHZ. In this case only one trans-receiver is required at UE end and the UE should be compatible to operate on a larger Bandwidth.

  fig: 3

 

2.Intraband Non-Contiguous: Here Carriers from the same band is used, but they are not continuous. So, the multi carrier signal can not be considered as a single signal. For this reason, the design and implementation are a bit complex especially from the UE perspective. Unlike Intraband Contagious where only a single Transceiver is needed, noncontiguous requires two transceivers.

fig: 4

  1. Interband Noncontiguous: Here different band of frequency is used. The UE is this case needs to have Transceivers of multiple bands. This method is important in ensuring mobility integrity using the propagation characteristics of multiple bands.

Deployment Scenarios of Carrier Aggregation:     

Carrier Aggregation can be deployed in multiple scenarios as shown in the Fig 4.

Scenario 1:  Here the Enb antennas are Collocated and same radiation patterns for both the Component Carrier (CC1 and CC2). Here both the CC1 and CC2 provides same coverage pattern and are within the same band.

Scenario2: Here the Enb has Component Carriers from different bands and there is a gap between the two carriers. Here both the CCs have different coverage and here we can use CA to get higher throughput at the areas where the coverage is overlapping. Here FreqCC1<FreqCC2.  Here F1 provides  coverage F2 is used to increase throughput

Scenario3: Here CC1 and CC2 are collocated and are of different bands. But the radiation patters of the antenna of each carrier are intentionally changed to improve cell edge coverage.

Scenario4: Here the CC1 provides the macro coverage but CC2 provides throughput at traffic hotspots, mostly by small cells.

 

 

Band Combinations and UE Class:

CA_X: Denotes intra-Band Contiguous CA

Ex: CA_5 here 5 denotes a carrier of Band 5

CA _X_X- Denotes Intra-Band Non-Contiguous CA

Ex: CA_5_5: Here 5_5 denotes two carrier

CA_X_Y – Denotes Inter Band CA

EX: CA_5_8: Here 5 denotes a carrier of Band 5 and 8 denotes a carrier of Band 8

Carrier aggregation is supported from UE from Category 6 as per below table:

Source:www.cablewireless.com

The below figure explains the CA bandwidth Class

Ex:  Let us understand the CA terminologies by referring the above table. If the UE sends an information to the network mention CA_3C, it means it supports CA bandwidth class C (it will cater to 2 Contiguous component carrier) and it supports Band 3.

Next if UE transmits this information to Network as CA_25A_25A, it means the UE supports one carrier of Band 25 and an another noncontiguous carrier of Band 25.

We will check another Scenario CA_1A_8A, this means that the UE supports 1 carrier of Band A and another carrier of Band 8. These carriers can take any BW of LTE starting from 1.4 to 20 MHZ so there can be many combinations.

 

Source:www.5gamericas.com

 

Component Carriers:

Each Carrier in CA are known as Component Carriers. CC are two types:

  1. Primary Component Carrier: This is the main carrier in CA and there is a primary downlink CC and a corresponding uplink CC. The Major functions of PCC are
  • Random Access Procedure takes place by the PCC
  • PCC handles RRC and NAS functionalities
  • Measurement and Mobility aspects are taken care by PCC
  1. Secondary Component Carrier: The other carriers in CA are called as Secondary Component Carrier. The Major functions of SCC are
  • Activated only when the UE is in connected mode and helps in to increase data throughout
  • Can be dynamically activated and deactivated based on data Buffer in Eb or UE

 

PCC carries both Signaling messages and user data whereas SCC carries only user data. In FDD the number of CC in UL and DL can be different, but please note the no of UL CC is always lesser than or equal to DL CC. Each CC can be of different BW also.

Fig 5 (Source: www.3gpp.org)

Cross Carrier Scheduling: To understand the concept of cross carrier scheduling, lets discuss a bit about few LTE channels. PDCCH provides information on resource allocation, MCS, HARQ which enables data transmission to happen in PDSCH and PUSCH. There can be a possibility that for each CC, PDCCH carries the resource assignment and scheduling information for PDSCH and PUSCH. But there is an interesting feature that comes in CA, that is called Cross Carrier Scheduling. By Cross Carrier scheduling a PDCCH on PCC can ne configured to enable scheduling of PDSCH and PUSCH transmissions on another component carrier. This is done by a three-bit carrier indicator field (CIF) inserted at the beginning of PDCCH messages. This functionality helps to support ICIC (Inter Cell Interference Coordination) for PDCCH in HETNET.   Important point is Cross Carrier Scheduling is not applied to P-CELL, it is always scheduled by its dedicated PDCCH. The Enode B can enable or disable Cross-Carrier Scheduling for each Carrier on a UE.

 

Cell Types: There are three cell types in Carrier Aggregation:

Primary Cell (PCell): The Cell which is operating on Primary Frequency is called the Primary Cell. In this cell the UE performs the Initial Access or RRC reestablishment process and Handover Functionalities. It carries both signaling and data throughput

Secondary Cell (SCell): This cell operates on secondary frequency and can be configured only after RRC connection is established. It carries only User data and provides additional radio resources.

Serving Cell: For UE which are not CA activated and the UE is only RRC_Connected, there is only one serving cell. But for UE which are in RRC_Connected and CA is activated there can be more than one Serving cell.

 

SWOT ANALYSIS of CARRIER AGGREGATION

Carrier Aggregation States and Transitions:

The carrier that is for PCell in Downlink is the Downlink Primary Component Carrier (DL PCC) and the carrier that is for SCell in downlink is the Downlink Secondary Component Carrier (DL SCC). At CA and when RRC connected, a UE will have at least one PCell and 0 or more SCells.

RRC performs the addition, deletion and reconfiguration of SCell. During HO all the SCells that are attached to the UE will be removed along with the PCell and new set of SCells will be configured when HO is completed and new PCell is attached to the UE.

                                                                                        fig.7

Pointers:

  1. During attach, RRC reconfiguration and HO, the eNB checks the CA license , the CA neighbor cell configuration and UE capability. If all the conditions are fulfilled, the CA UE is configured with a downlink SCell in the same RRC connection Reconfiguration Message
  2. CA feature supports SCell dynamic activation and deactivation which helps in UE battery savings.
  3. MAC layer is responsible for CA activation and De activation.
  4. All HARQ related transmissions are sent on the uplink via the PCell.
  5. Downlink transmissions on SCell are only possible if the SCell is in activated state.
  6. If the SCell is configured but deactivated, there is no data transmission.
  7. UEs doesn’t monitor deactivated component carriers. This helps in battery savings
  8. Dynamic activation and deactivation of SCell ensures that it is only activated when there is data demand that could benefit from transmitting on more than one carrier.

 

Carrier Aggregation SCell Configuration:

In CA, the RRC layer is responsible for CA configuration and deconfiguration. CA can only be applied to UEs which are in RRC connected mode and it happens only with the PCell. In CA, there is only one RRC connection for an UE. MAC layer is responsible for the split in data transmission over different CC.

 

Fig 8: SCell Configuration Process

RRC connection setup process is similar to CA capable UEs and non- CA capable UEs. SCell addition happens only when the RRC Connection Setup Complete message is received by the eNB by dedicated signaling in RRC Connection reconfiguration message. Addition, removal and reconfiguration of SCell is performed using RRC Signaling.

Below we will discuss a scenario where S-Cell is being configured as part of intra-frequency HO

fig 9

The above figure illustrates a scenario to depict the signaling process involved in intra frequency HO with a UE configured with a SCell. Reconfiguration message is used to change PCI of the Cell from 125 to 225.

We will explain the RRC signaling process for SCell configuration and reconfiguration process in Intra Freq HO.

  1. Event A3 measurement reports are sent by the UE to indicate that the neighboring cell is better than the serving cell.
  2. In response eNB sends RRC connection reconfiguration message to the UE to complete the intra freq HO. This will indicate the UE to remove the existing SCell and information to attach to another SCell after HO is complete
  3. UE now goes over with the HO process and sends a RRC connection reconfiguration Complete message to the new PCell.
  4. UE now starts reading the SIB messages on the new PCell.
  5. The UE receives another set of RRC connection reconfiguration message which contains the measurement report information to allow the UE to operate on the PCell. The UE continues to read SIB messages on the new PCell
  6. At this time since the UE is CA enabled it starts sending A1 measurement reports to the eNB indicating that the signal level is better than threshold.
  1. The RBS sends a RRC Connection Reconfiguration message to remove measurement id 4, which is the Event A1 measurement for confirmation of the coverage of the SCell
  2. The UE sends an RRC Connection Reconfiguration Complete message to acknowledge the receipt of the measurement information and confirm the changes to the measurement information.

 

The RRC information sent above is to facilitate the HO process from one PCell to target PCell and deconfigure existing SCell and configure new SCell.

In CA Configured state the UE is ready to be activated but SCell is not used, but in CA activated state the UE is ready to receive data on SCell.

Below are the detailed pointers of the activities performed by PCell and SCell

UE capability determines the total number of serving Cells that can be configured and this information is transmitted by the UE when UE is registering in the network. For CA capable UE, the SI for the PCELL is obtained from SIB messages. The UE doesn’t read the SIB messages for the SCell as it gets that information from the PCell which is provided by dedicated RRC signaling.

 

SCell Candidate Configuration:

Which Cell can be configured for SCell is decided by the operator and the SCell is always deconfigured at the time of an incoming HO. When PCell is changed due to intra LTE HO, the old SCell is removed and new SCell is configured, if the SCell is the old SCell, then also it is de- configured and then configured again. SCell can be deconfigured in poor coverage area and when there is interference from other cells.

 

Carrier Aggregation SCell Activation:

SCell can be activated / Deactivated based on the below three scenarios

  1. Need Based: SCell can be activated and deactivated based on the amount of data transmission present in RLC buffer of the UE
  2. Coverage Based: If the SCell has poor radio coverage it is deactivated
  3. Prohibit Timer: Activation and Deactivation is not allowed when the prohibit timer is running following previous transition.

fig .10

Carrier Aggregation and SCell deactivation:

  1. The amount of data in the RBS buffer for a UE is less than the threshold value.
  2. The channel quality of the SCell is less than the threshold value.
  3. The prohibit timer is not running.

 

Now for dynamic SCell Selection process we need to consider the following measurement reports. This feature is implemented to enable the UE to determine the right SCell among all the probable candidates for a particular PCell.

Event A1: Serving cell is better than threshold. This explains if the coverage of the current SCell is better than the threshold.

Event A2: Serving cell is worse than the threshold. This report indicates whether the coverage of the current SCell is poor, if yes then the SCell is deactivated.

Event A6: SCell neighbor becomes offset better than the SCell. This helps to determine the best candidate among SCells on the same frequency band.

Summary of Handover Events for quick reference:

Uplink Carrier Aggregation:

Uplink carrier aggregation operates in conjugation with Downlink Carrier aggregation and it is always to be remembered that the no of CC on uplink is always less than or equal to DL CC , it cant exceed the DL CC. PUCCH is transmitted by the UE on the uplink PCell. An UL CA SCell is selected from the available downlink SCell , if the UE is uplink CA capable and if frequency is supported by UE. But Uplink CA comes with the below limitations.

  1. We can have maximum 1 SCell configured for UL CA
  2. UL CA Is not supported in Intra-Band noncontiguous CA
  3. FDD+TDD UL CA is not supported
  4. The UL CC and DL CC needs to be on the same band.

Few Important points:

1.When SCell is activated the UE monitors the PDCCH of the PCell and all the active Scells to determine whether there is a downlink transmission intended for the UE on that cell

2.When the SCell is activated all HARQ ACKs for all Cells, both PCell and SCell are transmitted to the RBS by PCell.

3.CSI reports of the SCell are sent to the RBS using the PCell.

FDD-TDD CA:

In earlier releases CA was possible only between FDD-FDD or between TDD-TDD. But now the process of CA has advanced to FDD-TD CA which offers the proposition where we can aggregate FDD and TDD Carriers. Point here to be noted is that for FDD-TDD CA, the PCell will always be FDD.

CA Future Enhancements:

CA started with release 10 where maximum of 5 Component carriers can be aggregated. But now CA can be extended till 7CC DL carrier aggregation. But these are all license dependent and comes with prerequisites.

 

Some Brainstorming now, with a cup of Cup

1.From which 3GPP release CA has come for implementation?

Ans: From Release 10

2.From which Category of UE CA is supported?

Ans: Category 6

3.How Network identifies that CA is supported by UE?

Ans: During initial registration UE send UE capability information to the eNB, from those messages the Network can know if the UE supports CA

4.In Release 10 how many CC can we have?

Ans: 1PCC and 4 SCC can be supported in release 10.

5.In Release 13 how many CC can we have?

Ans: Up to 32 CC

6.What is the role of the MAC Layer in CA?

Ans: 1. Activate and Deactivate Secondary cell.

2.Aggregates data from multiple cells.

7.Why MIMO when we have Carrier Aggregation?

Ans: CA is like decreasing the congestion in a highway by increasing the lanes. For example, here parallel data streams are transmitted on separate EARFCN. But consider the case when there is no possibility of extra BW or EARFCN, in that case MIMO comes in picture. MIMO is basically sending parallel data stream on the same frequency exploiting Antenna architecture.

8.What are the limitation for Carrier Aggregation with respect to Carrier Frequency?

Ans: Carrier aggregation can not be implemented for the same frequency band. Even if they are in contiguous band there should be a separation of 300KHZ between the EARFCNs.



Categories: techlteworld

Tags: , , , , , , , ,

Leave a Reply

%d bloggers like this: