Mobile-to-mobile Cooperation

Project Description and goals:

Figure 1
Figure 1

Mobile-to-mobile (M2M) communication is a topic of interest in future 4G LTE-advanced networks. As opposed to normal M2M communication in LTE advanced in which the two mobiles only communicate between themselves, in this project, we consider M2M cooperation in the uplink transmission. This idea is illustrated in Figure 1 where two mobiles cooperate to increase their transmission rate to the base station.

We study the performance of the proposed scheme in terms of the achievable rate region, the outer bound, and the optimal resource allocation to maximize the transmission rates and outage performance.

 

 

 

 

 

 

Transmission Scheme

In order to meet the half-duplex constraint in wireless communication, we divide the transmission block into 3 phases with variable durations as shown in Figure 2. Each mobile splits its information into a cooperative and private part. The two mobiles partially exchange their information in the first two phases, then cooperatively transmit to the base station in the third phase. The base station decodes directly without any block delay, at the end of each block using joint decoding of the received signals over the 3 phases.

JSAC2
Figure 2

 

 

 

 

Results

Achievable rate region

Figure 3 shows the achievable rate region of the cooperative scheme, non-cooperative classical multiple access channel (MAC), and the outer bound. Results show that M2M cooperation can significantly enlarge the achievable rate region and becomes close to the outer bound as the cooperative links becomes stronger.

FIG5JSAC
Figure 3

 

 

 

 

 

 

 

Optimal resource allocations

In Figure 5, the two mobiles (U1 and U2) are set in fixed locations  while the base station location is varied over the entire x-y plan. The link strengths are affected by the distances between nodes as defined in the path loss model. Figure 5 shows the optimal transmission scheme for each base station location in order to maximize the sum rate. Results show that the optimal transmission scheme moves from no cooperation to partial decode-forward, and then to full decode forward as the cooperative links move from weaker to slightly stronger and then significantly stronger than the direct links, respectively.

regionsum
Figure 5

 

 

 

 

 

 

 

 

 

Outage Probability

Figures 6 and 7 show the outage performance of the proposed scheme compared with the non-cooperative scheme over Rayleigh fading channels. In Figure 6, given a specific target rate (R1=R2), the outage of the proposed cooperative scheme outperforms the non-cooperative scheme especially at high SNR. Moreover, the proposed scheme achieves a full diversity order of 2.

Figure 7 considers the outage rate region – the region at which the transmission rates can be received without outage. Result show that the proposed scheme can significantly enlarge the outage rate region.

Figure 7
Figure 7
Finovs3
Figure 6

 

 

 

 

 

 

 

 

 

 

 

 

Publications:

  1. “Spectral Efficiency and Outage Performance for Device-to-Device Cooperation in Uplink Cellular Communication,”
    A. Abu Al Haija and M. Vu, submitted to IEEE JSAC Special Issue on 5G Wireless Communication Systems, Dec. 2013.
  2. “Outage Analysis for Uplink Mobile-to-Mobile Cooperation,”
    A. Abu Al Haija and M. Vu, IEEE Conf. on Global Comm. (Globecom), Dec 2013.
  3. “Rate Maximization for Half-Duplex Multiple Access with Cooperating Transmitters,”
    A. Abu Al Haija and Mai Vu, IEEE Trans. on Comm., vol. 61, no. 9, pp. 3620 – 3634, Sept. 2013.
  4. “A Half-Duplex Cooperative Scheme with Partial Decode-Forward Relaying,”
    A. Abu Al Haija and M. Vu, IEEE Int’l Symposium on Info. Theory (ISIT), July 2011.
  5. “Throughput-Optimal Half-Duplex Cooperative Scheme with Partial Decode-Forward Relaying”,
    A. Abu Al Haija and M. Vu, IEEE Int’l Conf. on Communications (ICC), June 2011.
  6. “Joint Typicality Analysis for Half-Duplex Cooperative Communication,”
    A. Abu Al Haija and M. Vu, The 12th Canadian Workshop on Information Theory (CWIT), May 2011.
  7. “Can Half-Duplex Be Simply Derived from Full-Duplex Communications?”,
    A. Abu Al Haija and M. Vu, Workshop on Information Theory and Applications, UCSD, San Diego, Feb 2011.