Chapter 6
Conclusions and Future Work
The explosion of mobile data traffic raises the problem of spectrum scarcity, and it
is one of primary driving force for the next generation cellular networks. There are
various solutions proposed to increase the efficiency of existing cellular networks.
This thesis investigated two promising solutions to alleviate the above problem:
cooperative spectrum sharing and device to device communication.
The research work presented in the thesis are categorized into four parts. In
the first part, an OFDM based subcarrier sharing scheme is proposed in which the
secondary system attains OSS by helping the primary system to achieve its target
rate of communication. The proposed scheme is classified into two categories: (i)
non-adaptive CSS with MRC diversity (ii) adaptive CSS with BER-SC diversity.
As a primary contribution, for both categories, closed-form expressions of outage
probability for the primary and secondary systems are derived. Analytical expres-
sion of critical distance which determines the region of validity of the proposed
scheme is also derived. The proposed scheme alleviates the problem of secondary
signal interference at the primary system as occurs in single carrier based CSS
protocols. Moreover, incorporation of adaptive modulation at ST increases the
throughput of the primary system while satisfying its BER constraint. Results
show that, depending on the distance between PT and ST, it is possible to find
the exact number of subcarriers that can be forwarded by ST to PR for both adap-
tive and nonadaptive CSS protocols. Thus, remaining subcarriers can be used for
secondary system transmission.
In the second part, we presented the OFDMA based angle constraint D2D
124
communication. Rate and outage probability trade-offs of three D2D frameworks:
underlay, overlay, and C-D2D are computed. The closed-form expressions of the
outage probability for all three frameworks are derived. Specifically, for underlay
framework, the optimal distance between CU-BS and the minimum value of angle θ
(an angle between cellular link and D2D interference link) is also derived. By using
above parameters, BS can select a CU which can facilitate D2D communication
while satisfying its QoS. For overlay and C-D2D frameworks, an optimal subcarrier
sharing scheme is proposed which leads to performance gains for the cellular and
D2D users. There is good agreement between the analytical and simulation results.
The results also showed that for high cellular outage constraint, C-D2D framework
outperforms the other two frameworks.
In the third part, we augment the development of C-D2D, by proposing a best
D2D user selection protocol for the C-D2D framework wherein multiple D2D users
co-exist with the cellular users in the same cell. Specifically, among M available
D2D pairs, selected D2D user (based on received SNR) helps the CU by relaying
few subcarriers for uplink cellular transmission. In return, M D2D UE can share
the spectrum of single CU while satisfying the target QoS of CU. In addition, for
fair distribution of remaining subcarriers among M D2D pairs, a novel round robin
method with priority cube is also proposed. The rate and outage probability of
the proposed framework are derived for two cases:
• CU-DT-BS link (without MRC at BS)
• CU-DT-BS with cellular link (with MRC at BS).
For given distance, power and target rate, an optimal value of M is also derived.
By using M , BS maps the optimal number of D2D pairs with each CU while
satisfying the QoS of both cellular and D2D users. Simulation results show that
the proposed best D2D user selection scheme outperforms the conventional C-D2D
framework.
In the fourth part, we have designed and implemented an experimental testbed
for the C-D2D framework on NI-USRP 2922 platform. Further, measurement
results for OFDMA based C-D2D framework are also obtained. In an adaptive
C-D2D framework, we employ adaptive modulation and coding (AMC) at DT to
enhance the throughput of both cellular and D2D users. The measurement results
125
are obtained in terms of SER and received SNR. The spectral efficiency of the
proposed scheme is also computed. It is shown that the proposed framework sig-
nificantly improves the SER and BER of both cellular and D2D users as compared
to conventional direct communication.
6.1 Future work
In this section, we present the exciting directions for future work.
• In this thesis, we consider a single cell scenario with an assumption of neg-
ligible inter-cell interference. A straightforward extension of our work is the
investigation of the behavior of proposed scheme in the multi-cell scenario or
multi tier heterogeneous network.
• In this thesis, we assume that all subcarriers carry equal power and wireless
channel is modeled as Rayleigh frequency flat. As a part of future work, the
proposed work can be extended for frequency selective channel with variable
subcarrier power. Apart from above, different level of adaptation such as bin
level, frame level, subcarrier level can be adopted at different wireless nodes.
• For the measurement results, only one D2D pair is considered. As a part of
future work, multiple D2D pairs can be used to scale the performance of the
proposed scheme with the increase in density of devices. In addition, practical
performance of the best D2D user selection scheme for C-D2D framework
proposed in Chapter 4 can be obtained in a realistic scenario. Furthermore,
the experiments have been performed in an indoor laboratory environment.
It would be interesting to see the performance assessment in challenging
outdoor environment.
• Finally, in the higher frequency band (for e.g. mmWave), integration of D2D
communication and massive MIMO would be interesting. The mmWave
system requires directivity gain to counter the severe channel attenuations.
This directionality can lead to performance gains for D2D communication
due to less multi user interference in mmWave network.
