Relay Technologies for WiMAX and LTE-Advanced Mobile Systems

Relay Technologies for WiMAX and LTE-Advanced Mobile Systems


Chintan I. Patel 

 

1. Introduction

 

The present scenario suggests tremendous growth in the mobile and fixed broadband worlds. With the establishment of 3G technologies, the authorized organizations are looking to deploy 4G network worldwide. Fourth Generation calls for very high data rates with a robust relay and backhaul architecture able to support these overwhelming peak data rates such as 100 Mbps and 1Gbps for mobile and fixed environments respectively.

 

In the paper [1], relay technology has been discussed, throwing light upon different types, transmission techniques, pairing schemes for WiMAX and LTE-Advanced technologies. The main objective is to build a strong interface for radio access and smooth communication among the base stations (eNB) and relay stations with the User Equipments (UE). Improvements in service coverage and system throughput are major attributes to be looked while installing the relay stations and examining its communications with eNBs and UE [1].

 

International Mobile Telecommunications-Advanced (IMT-Advanced) standards are defined by the International Telecommunications Union (ITU) for the standardization of 4G technologies like LTE-Advanced and mobile WiMAX (IEEE 802.16m). Relay technology is an important aspect taken into consideration during the standardization process. Relay transmission technique involves forwarding of UE/mobile units information to the local eNBs via relay stations installed at different locations to carry out this process. Service coverage and the overall throughput can be enhanced by using this technique [1].

 

 

2. Types of Relay technologies and their transmission schemes

 

Relay based multi-hop techniques were developed under the standard IEE 802.16j and IEEE 802.16m standards are in process by the Working Group to meet the IMT-Advanced standards.

 

2.1 Types of Relay

 

Following are the two types of relays defined in the LTE-Advanced and IEEE 802.16j standards [1].

 

 

 

 2.1.1 Type-I or Non-transparency

 

 As seen in Fig.1, Type-I relays are used to establish communication between local eNB/BS and UEs located far away from the eNB/BS. A relay link is established between UE and RS and an access link exists between enB/BS and RS. This access by the UE of the eNB takes place via the Relay Station(s) (RS). The Type-I RS transmits common reference signals and control information of eNB/BS to the UE in order to communicate with the UE and connect the same to the eNB/BS. This process allows to increase capacity and improve the service coverage, as the remote UE units are connected to the eNB/BS through the Type-I RS [1].

 

2.1.2 Type-II or Transparency

 

 If the UE is within the vicinity if the eNB, then Type-II relays are used. In this case, the RS does not transmit common reference signal or control information to the UE [1]. The local UE service quality and the link capacity can be improved by connecting the UE to the eNB via the RS in presence of a direct communication link between the eNB and UE [1]. “The overall capacity of the system are improved by achieving multipath diversity and transmission gains of the local UE” [1].

 

 

2.1.3 System performance of Type-I Relay

 

 

Table 1 shows the performance of the overall system for the Type-I relay. Consider that eNB is located at the center among randomly deployed UEs. Selective Decode and Forwarding (DCF) relaying technique is used in the process [1]. From the table 1, it is observed that throughout is increased significantly for the remote UEs and they increase as the number of RSs are increased. Thus, the system capacity may also increase. The throughout at the edge of a cell is said to be less generally. By using the relay stations, the performance at the edge are enhanced for the UEs.  

 

2.2 Transmission schemes


Generally two hop technique is applied between the eNB and the UE units through the RS.

 

2.2.1 Amplify and Forward (AF)

 

The RS receives the signal from eNB (or UE). This signal is then processed and undergoes amplification and is then passed over to the UE (or eNB).  This process consists of two phases as discussed. The AF scheme also amplifies noise, due to which it is not used for relay transmission. The advantage of this scheme is that it has a simple operation and very short delay [1] as no error checking or Cyclic Redundancy Check (CRC) techniques are performed as will be seen in other transmission schemes.

 

2.2.2 Selective Decode and Forward (DCF)

 

Channel decoding is done at the RS of the signal received from the eNB (or UE). If the decoded signal is correct, then CRC is performed to check for errors at the RS and the signal is forwarded to the UE (or eNB) [1]. The advantage of the DCF scheme is that it avoids error propagation, but the long processing delay is the disadvantage of using this technique.

 

2.2.3 Demodulation and Forward (DMF)

 

 In the DMF scheme, the RS demodulates the signal received from the eNB (or UE), without undergoing the decoding process. The signal is now modulated and forwarded to the UE (or eNB). The overall operation is simple and the processing delay is low. The propagation errors cannot be avoided [1].

 

 

2.3 Relay technologies in WiMAX and LTE-Advanced systems

 

 A RS may act as an eNB/BS provided it transmits its own synchronization channels, control channels, reference symbols and has a unique physical cell identifier [1]. Only traffic backhauling cannot be supported by such a RS, in order to appear as a eNB/BS to its UE units.

 

 

 

Table 2 [1] depicts a comparison between Type-I and non-transparency RSs. Type-I RSs are used in 3GPP LTE-Advanced systems and non-transparency RSs are used for 802.18j. The non-transparency RSs support multiple hops whereas Type-I supports up to 2 hops in the relay system. The overall transmission latency or delay in relay transmission is longer for IEEE 802.16j than in 3GPP LTE-Advanced technology.

 

For the legacy UEs to access the 3GPP LTE-Advanced networks, a fake Multicast Broadcast Single Frequency Network (MBSFN) technique is used to access the more advanced E-UTRAN [1].

 

 

3. Pairing schemes for Selection of Relay

 

Fig.1 shows multiple UEs and RSs in each cell. Pairing of these equipments and stations must be done in order to achieve relay gain. Pairing is mandatory to connect multiple UE units to the RSs and hence the eNB/BS in the area of communication.

 

 

3.1 Centralized Pairing Scheme

 

 In the centralized pairing scheme, the eNB collects information from all the RSs and UE units present in its vicinity. Here the eNB acts as a control node which is required to make pairing decisions for RSs with UE units [1]. Each RS prepares a service set by including the UEs it can serve within its vicinity and checks the service quality or channel conditions on links between RS and eNB, and RS and UEs. This is important information for pairing and is frequently updated at the local eNB for the process to take place.

 

Timely updates are received from all the RSs by the eNB. The eNB then generates a matrix by which the rows and columns correspond to the UE ID and RS ID. It is assumed that each RS is paired with just one UE unit, in order to calculate overall throughput of the system. If the RS serves its intended UE, then it is represented by ‘1’ in the matrix. For the rest UEs, a ‘0’ is put into the matrix since the RS communicates with just one UE (as assumed). This matrix is filled up by multiple RSs and the output matrix is obtained. Each column will have one non-zero number which represents the UE corresponding to its RS. By combining all the columns, the overall throughput can be obtained as a result of the combination of all the UEs corresponding to their RSs in a cell [1].

 

 

3.2 Distributed Pairing Scheme

 

 In the distributed pairing scheme, contention-based MAC mechanism and local channel information are used by the RSs to select the respective UEs [1]. A common slotted communication channel is shared by all the RSs. ‘N’ slots are grouped into ‘M’ pairing sections. In the distributed pairing process, all the RSs with one UE select a slot from N slots to broadcast the UE associated with it (RSs).

 

If more than one RSs select the same slot, a collision occurs and the RSs are required to try once again in the next pairing section. All the pairing is recorded in the service set at the eNB. If there are multiple RSs associated to multiple UEs, then they are allowed to enter the pairing process once the single UE are served the purpose. If the multiple RSs are associated to any of the previously served UE, then according to the updated service set, the RSs must remove the particular UE (s) that has already been served previously.

 

By prioritizing RSs associated to single UE, collisions can be reduced and the overall system capacity can be improved comparable to multiple-RS-multiple-UE scenario [1]. Signal overheads as observed in centralized pairing scheme is reduced in the distributed pairing scheme.  

 

Pairing probability of UE units in a cell

 

 


As seen in fig.2, the pairing probability of distributed scheme is more than centralized as the number of relay stations increase.

 

 

4. Impact of Relay technology on Next Generation Networks (NGN)

 

 We know that NGN are in great demand and therefore require more capacity and throughout. Relay technology truly helps in providing good service quality, enhanced throughputs at edges of the cell and remote locations and hence more capacity in terms of number of users. Centralized pairing scheme for relays do have signaling overheads, but they provide better performance gains than their counterpart distributed pairing scheme [1].

 

 

NGN consists of multi-user environments, and the pairing schemes proposed in the paper [1] achieve maximum number for served UEs and enhanced throughput characteristics. From fig.3 it is seen that centralized pairing scheme provides better throughput than distributed pairing scheme, therefore providing UEs with high data rates and efficient operation.

 

 

5. Conclusion

 

 It is concluded that relays provide flexibility along with enhanced throughput capability and increase in service quality and capacity. Since NGN are complex, DCF transmission mechanism provides better performance than AF and DMF relay techniques [1]. Centralized and distributed pairing schemes achieve maximum performance with increase in capacity and throughputs. Relay mechanisms help in connecting remote UEs to eNBs/BSs and provide enhanced performance at the edge of the cell also.

 

 

 

Reference

 

[1] Relay technologies for WiMax and LTE-advanced mobile systems”, Yang Yang; Honglin Hu; Jing Xu; Guoqiang Mao; 
Communications Magazine, IEEE 
Volume: 47 , Issue: 10 
Digital Object Identifier: 10.1109/MCOM.2009.5273815 
Publication Year: 2009

 

 

 

 

Comments