Page 44 - ITU Journal Future and evolving technologies – Volume 2 (2021), Issue 2
P. 44

ITU Journal on Future and Evolving Technologies, Volume 2 (2021), Issue 2




          nated to initiate content discovery, which decreases colli‑  cache placement algorithms. In the advertisement based
          sions and improves of loading.                       scheme (ABD2D) Help Users (HUs) advertise cached con‑
                                                               tent and Ordinary Users (OUs) look for content over a lim‑
              6.5 6  ABD2D-BM                                  ited number of dedicated resource blocks (RBs). On the
             Number of Beacons for Unit Offloading  4.5 5 4    In both cases, HUs have a clustered structure, which im‑
                                                               other hand, in the discovery based scheme (DBD2D), Or‑
                    ABD2D
                                                               dinary Users (OUs) send discovery messages in order to
                    DBD2D-BM
                    DBD2D
              5.5
                                                               express to the HUs their need for content.
                                                               proves both cache placement and content delivery. For
                                                               both procedures, we make comprehensive simulations
                                                               for various performance metrics and against varying pa‑
              3.5
                                                               rameters. Simulation results show that ABD2D along

                                                               tency, and improves of loading performance even for
              2.5
                                                               compelling circumstances, such as, high user density, less
               2 3                                             with clustered HUs, provides less overhead, reduced la‑
                                                               number of dedicated RBs, and high SINR requirement.
               -10   -5     0     5     10    15    20
                           SINR Threshold (dB)
                                                               In the scope of this paper, we have assumed that the
          Fig. 13 – Mean Number of Beacons for Unit Of loading for Varying SINR  content popularity among users is time‑invariant, well
          Threshold Value                                      known and not user dependent. In a more realistic sce‑
                                                               nario, stationary and totally identical user preferences
                                                               are not expected. Therefore, HU‑side time‑varying pop‑
             400
                                                               ularity and its estimation is planned to be investigated in
                                                               future work. Furthermore, the attitude of the proposed
             350
                                                               procedures against mobility will be examined in detail
                     ABD2D-BM                                  later on in conjunction with cluster lifetime estimation
             Mean Offloading  250  DBD2D                       riod. Also the proposed model and methods for in‑band
             300
                     ABD2D
                                                               and adaptive clustering and cache placement renewal pe‑
                     DBD2D-BM
                     DBCA+HCA
                     MPCC
                                                               D2D communication must be applicable to existing and
                                                               future technologies. The LTE Direct framework is consid‑
             200
                                                               ered while deriving a proposed resource allocation model
                                                               and access methodology. On the other hand, technical re‑
             150
                                                               ports and speci ication documents prepared by the 3GPP
                                                               that cover 5G proximity services are not available yet. Ac‑
             100
               -10   -5     0     5     10    15    20         cordingly, adaptation of the network model and proposed
                           SINR Threshold (dB)                 methodologies to 5G technology is prioritized for upcom‑
                                                               ing research.
             Fig. 14 – Mean Of loading for Varying SINR Threshold Value
          Lastly, Figs. 13 and 14 present the performance indica‑  REFERENCES
          tors against varying numbers of SINR threshold. Of load‑
          ing performance of the proposed ABD2D procedure tends  [1] A. Osseiran, J.F. Monserrat, P. Marsch, O. Queseth,
                                                                                                          ̈
          to furcate from the upper bound with the increase in the   H. Tullberg, M. Fallgren, K. Kusume, A. Hoglund,
          threshold value. But it is still less than 5% close to the  H. Droste, I. Silva, P. Rost, M. Boldi, J. Sachs, P.
          DBCA+HCA and signi icantly better than the ABD2D‑BM,       Popovski, D. Gozalvez‑Serrano, P. Fertl, Z. Li, F.
          DBD2D, and DBD2D‑BM. Since the number of users does        Sanchez Moya, G. Fodor, and J. Lianghai. 5G Mo‑
          not change, a decrease in the of loading also causes an in‑  bile and Wireless Communications Technology. June
          crease in the number of beacons. These results indicate    2016. ISBN: 9781107130098.
          that the ABD2D procedure is more convenient even for  [2] 3GPP. Universal Mobile Telecommunications Sys‑
          higher‑order modulation techniques that require higher     tem (UMTS); LTE; Proximity‑based services (ProSe);
          SINR values.                                               Technical Speci ication (TS) 23.303. Version 12.2.0.
                                                                     3rd Generation Partnership Project (3GPP), Sept.
          8.  CONCLUSIONS                                            2014. URL: https://www.etsi.org/deliver/
                                                                     etsi_ts/123300_123399/123303/12.02.00_
          In this work, we proposed procedures for Device‑to‑        60/ts_123303v120200p.pdf.
          Device (D2D) cache management, discovery, advertise‑
          ment, and content delivery.  We considered two dis‑   [3] W. Shuanshuan. “Application of D2D in 5G Net‑
          tinct procedures for device and content discovery and      works”. In: May 2015. URL: https :/ /www . zte.
          apply them with previously proposed clustering and





          30                                 © International Telecommunication Union, 2021
   39   40   41   42   43   44   45   46   47   48   49