Typical Handover Scenarios in Relay Enhanced LTE Networks

Typical Handover Scenarios in Relay Enhanced LTE Networks

Chintan I. Patel



Relays in networks are considered as an advantage as well as a compulsion in next generation wireless and wired networks. Certain procedures such as handovers, a radio resource management procedure plays a significant role in order to enable relaying in LTE networks. This paper reports about the various handover scenarios for Release 8 (LTE) mechanisms and a typical stepwise explanation of the handover process in each scenario.


1. Introduction


By introducing relays in next generation networks, an increase in capacity and coverage provided by LTE networks can be seen. Also, a significant increase in throughput and reduction in the overall power consumption of cellular networks can be observed. Improvement of coverage in indoors and hotspots can be effectively increased, reduction of cast of deploying cellular networks and extension of battery life of the User Equipments (UEs) is achieved by the use of relays.



Figure 1 depicts an example scenario that shows a basic type of architecture that can be deployed in order to connected different Relay Nodes (RNs) with eNBs (Base Station in LTE known as Enhanced Node B) through Relay links, each RN and eNB may be connected to UEs via Access links.


Proceedings of this article: Since the focus of the report is handover in LTE networks, handover procedures for LTE Release 8, Relay Enhanced LTE supporting centralized and distributed/decentralized relay structures, and Relay Nodes will be seen.



2. Typical Handover Scenarios


2.1 LTE Release 8 handover



There are three stages involved in the handover process in LTE Release 8, namely, initiation, execution and termination. The decisions are taken by the source eNB to handover a UE to another RN or a target eNB. This handover process is shown in fig.2.  


During initiation, the source eNB decides the handover for UE depending upon the measurement reports obtained from the UE. The next stage is execution in which the source eNB communicates with the target eNB. If the target eNB can allow the UE, depending upon the traffic it can handle, the message is passed on to the source eNB and thereafter to the UE for its handover. The downlink data at the source eNB to the UE is forwarded to the target eNB. The final stage is the completion/termination, in which the UE and the target eNB are synchronized and handover is confirmed.


When the handover is complete, the notification is sent to the MME and the update is sent to the serving gateway. After an acknowledgement is received from the MME, the target eNB asks the source eNB to release all the resources of the UE, which then communicates directly with the target eNB.


2.2 Relay Enhanced LTE handover


As seen in fig.1, the basic architecture of an LTE network also consists of Relay Nodes which play a significant part in the LTE model. With the introduction of RNs in the network, the handover behavior also changes, which will be discussed in the following sub-sections. Two types of processes exist, centralized and decentralized. In the present scenario, there is a need for an architecture that can support both centralized and distributed models.


2.2.1 Centralized Relaying



As shown in fig.3, the elements marked “New” are added in addition to the elements we observed in fig.2. The eNBr suggests that the eNode B’s support relaying. The procedure for the centralized process is the same except the introduction of Relay Nodes as compared to the handover process seen for Release 8. The handover behavior changes accordingly with the addition of RNs.


The handover is initiated by the source eNBr. The source eNBr controls the source RN and similarly, the target eNBr controls the target RN. The UE is connected to the source eNBr via the source RN. The measurement reports are sent to the source eNBr in order for it to make a handover decision. The handover decision depends upon the UE’s measurements whether a handover must be done or not. The handover request is then sent over to the target eNBr, which checks whether an admission of the UE is made with the target RN. An acknowledgement is sent to the source eNBr and subsequently to the source RN. Any buffered downlink packets must be allocated to the target RN via the target eNBr by the source eNBr. Hence, synchronization between the UE and the target RN is established and information regarding Timing Advance for the UE is sent on the uplink. The handover is confirmed by the UE to the target RN and the target RN confirms the handover to target eNBr which in turn passes the information to the Gateway via the MME.


The Serving Gateway confirms the handover complete and sends an acknowledgement to the target eNBr which further asks the source eNBr and subsequently the source RN to release the resources of the UE, so that all the data can be sent directly to the target RN. The centralized relaying suggests that the overall process is controlled by the eNBr’s (both source and target).


2.2.2 Distributed/Decentralized Relaying



Figure 4: Decentralized Relaying for handover in LTE

Element marked “New” are additional functions as seen previously in fig.3 and the ones marked “modified” have been changed with respect to the type of handover process. In decentralized handover, the source RN initiates the handover. Both the RN and eNBr’s work in collaboration to successfully conduct the handover.


The measurement reports form the UE are collected by the source RN and passed over to the source eNBr. The handover decision is taken by the source eNBr and a request has been sent to the target eNBr. The target eNBr performs the admission control on the backhaul link. The target RN receives the handover request from the target eNBr which then performs the admission control for the relay link.


The handover request is acknowledged and the handover command is sent to the UE. The downlink data is sent to the target RN through the target eNBr. All downlink packets are buffered at the target RN and synchronization is established with the UE. Timing Advance for the UE is performed at the uplink and the handover confirmation message is given to the target RN by the UE. The target RN sends the confirmation message to the target eNBr which then sends the handover complete message to the Gateway via the MME.


The target eNBr receives an acknowledgement from the Serving Gateway. The target eNBr asks the source eNBr to release resources which in turn sends a command to the source RN, which initiated the handover to release resources of the UE. The data transfer now takes place between the target RN and the UE after the handover has taken place.



3. Conclusion


Three handover scenarios were seen in this article depicting the different handover processes in each of them. Introduction of Relay Nodes indeed increase the coverage, throughput and capacity, to name a few. The Relay architectures are designed to provide convenient access to future technologies such as LTE-Advanced by which the handover of UEs and RNs would be made easier and effective. 


"Handover Framework for Relay Enhanced LTE Networks", Oumer Teyeb, Vinh Van Phan, Bernhaard Raaf, Simone Redana