User-Assisted Relaying in Cellular Networks

User-Assisted Relaying in Cellular Networks

Figure 1. System model of uplink user-assisted relaying.

Project Overview:

Relay-aided cooperative communication techniques represent a promising technology that improves performance in poor coverage areas by enabling ubiquitous coverage even for users in the most unfavorable channel conditions. In this project, we exploit stochastic geometry to study the potential of using user-assisted relaying in future cellular networks, propose geometric based cooperation policies, and study the effect of the additional transmission of the relaying nodes on interference and the performance of user-assisted relaying when deployed system-wide in a cellular network.


The latest release of the LTE standard allows the deployment of fixed wireless relays to help cell-edge mobiles. Yet, other advanced cellular relaying modes are expected in 5G systems to improve the topology and robustness of a cellular network. User-equipment based (user-assisted) relaying will be enabled by Device-to-Device (D2D) communications and is expected to provide more flexibility than fixed relaying in expanding the base station (BS) coverage into obscured areas, especially where there is high density of idle UEs. To the best of our knowledge, this project is the first that considers the analysis of user-assisted relaying in a network-wide cellular context. Since some idle users are now transmitting by relaying information of other users, the amount of interference generated to the network will increase. In this project, we use stochastic geometry as a tool to model and analyze this interference as well as the cooperation policy which governs how to select the idle user to act as a relay.

Transmission Scheme and Interference Model:

Figure 2. Gamma distribution fitting of interference power at the destination BS and the relay in the 1st phase (SNR = 15 dB).

In this work, we consider a transmission scheme in which the active UE (source) has the option to decide between using direct transmission or half-duplex partial decode-and-forward transmission according to a cooperation policy. This cooperation policy will ideally depend on the actual quality of the active UE -to- relaying UE and the active UE -to- BS links which depend on the distance between nodes (pathloss), fading, and interference. It is sometimes difficult to obtain information about the channel fading especially in a fast fading channel, hence, we propose two practical cooperation policies: a pure geometric policy that only takes pathloss into account and depends on the locations of the nodes, and a hybrid policy that takes into account pathloss and fading of the channel.

We model the interference to a cell of fixed radius as a Gamma distributed random variable that depends on the field of interferers location, their transmit powers, the transmission strategy they deploy, and the fading of their channels to the cell under study. Our analysis shows that Gamma distribution is a good fit of the actual interference distribution.

Discussion and Results

Rate Gain versus Relay Location

Results in Fig. 3 show that performance of the system using the pure geometric cooperation policy gets worse than simulation using the ideal cooperation policy when the source-to-relay and source-to-destination distances ratio, r2/r1, is above 65%, i.e. when the relay is closer to the destination than the source. This difference is mainly due to the lack of the small scale fading information at the transmitting node in deciding whether to perform cooperation or not. An improvement to the performance is observed when we make use of the small scale fading knowledge as in the hybrid cooperation policy which exhibits a close match with simulation results.

Since cell sizes can vary slightly in practical cellular networks deployment, we examine in Fig. 4 the effect of cell size on performance under the hybrid policy by changing the radius of the cell under consideration both below and above the typical average value given the BSs density.

Both Fig. 3 and Fig. 4 suggest that we can reduce both cell size and transmit power without affecting the performance of user-assisted relaying and that it is usually more beneficial to have the relaying node closer to the active user than to the destination BS especially when we lack to the knowledge of the small scale fading of the channel. The maximum gain is achieved when the relay user is approximately midway (about 0.4 of the distance) between the active user and the BS.

Figure 3. Average transmission rate versus ratio between cooperation and direct distances for a cell edge user (r1 = 260m, Rc = 300m).
Figure 4. Average transmission rate versus ratio between cooperation and direct distances for different cell sizes (r1 = 120m, Pmax = 26 dBm).

Rate Rate Gain Averaged over all Idle User Locations

Fig. 5 shows the rate gain averaged over all possible locations of the relaying user versus the distance from the active user to the destination BS and compare it to maximum gain of the ideal case where the active user always finds a relaying node at exactly 0.4 distance between the active user and the destination BS, as suggested by results in Figs. 3. We note that uplink user-assisted relaying poses significant gains for near the cell-edge users while can be irrelevant for active users close to the destination BS.

Fig. 6 provides a quantitative evaluation of the average rate gain obtained for active users occupying the farthest 1/3 and 1/2 of the cell radius, averaged over all the possible locations of both the active and relay users. The gain increases with higher density of idle users, suggesting that user-assisted relaying is suitable for crowded population areas. These results show that when applying our scheme to the users towards the cell edge, we can achieve higher percentage gain.

Figure 5. Average transmission rate gain versus direct distance from the active user to the BS (Rc = 300m, Pmax = 23 dBm).
Figure 6. Average transmission rate gain versus users density ratio (Pmax=23 dBm).


  1. “Interference and Throughput Analysis of Uplink User-Assisted Relaying in Cellular Networks,”
    H. ElKotby and M. Vu, accepted to IEEE 25th Int’l Symp. on Personal, Indoor and Mobile Radio Commun. (PIMRC), Washington DC, Sept. 2014.
  2. “Uplink User-Assisted Relaying Deployment in Cellular Networks,”
    H. E. Elkotby and M. Vu, submitted to IEEE Trans. on Wireless Communications, August 2014.