Page 176 - ITU Journal Future and evolving technologies – Volume 2 (2021), Issue 2

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

-9 2 -5
Parameters:D=2.2 10 m /s, =20nm,T =6 10 s,L=6,K=4,J=1 sults of Fig. 6 show that the BER drops, as the value of
0 b
10 increases, provided that the SNR is suf iciently high. This
observation implies that MAI can be ef iciently mitigated
by deleting an appropriate number of entries from the de‑
-1
10
tection matrix. However, at the low SNR region, the EGC‑
IM with more elements removed from each row of the de‑
-2
10 tection matrix might not result in BER performance im‑
BER provement. In other words, for a given SNR and a given
value of , there is an optimum value of to attain the
10 -3
EGC-IM (M=8) best possible error performance.
EGC-IM (M=16)
EGC-IM (M=32)
10 -4 EGC (M=8)
EGC (M=16) 5. CONCLUSIONS
EGC (M=32)
In this paper a novel EGC‑IM detection scheme has been
-5
10
0 2 4 6 8 10 12 14 16 18 20 proposed for MTH‑MoSK DMC systems. We have in‑
SNR(dB) vestigated the BER achievable by the EGC‑IM and com‑
pared it with that of the conventional EGC‑aided detec‑
Fig. 5 – BER versus SNR performance of the MTH‑MoSK DMC systems
with the conventional EGC and proposed EGC‑IM, when the number of tion scheme. The results show that in the MTH‑MoSK
molecular types is = 8, 16 or 32. DMC systems supporting multiple nano‑machines, the
1
data rate . The reason behind this is that for a given value EGC‑IM scheme is capable of achieving much better BER
performance than the conventional EGC scheme within
of , MAI reduces as increases. This is because the −2
MAI entries are more sparsely distributed in the detec‑ the practical SNR region resulting in a BER below 10 .
Therefore, the proposed EGC‑IM scheme is ef icient for
tion matrix, as implied in Fig. 2, when becomes larger.
MAI mitigation. Furthermore, the proposed EGC‑IM has
Again, as shown by Fig. 5, the EGC‑IM detection scheme
outperforms the conventional EGC scheme and further‑ low complexity. In comparison with the conventional
EGC scheme, the extra operations required by the EGC‑IM
more, the performance gain at a given SNR increases with
scheme are only identifying some of the largest elements
the increase of .
in each row and deleting them. In addition, for given val‑
2
-5
-9
Parameters:D=2.2 10 m /s, =20nm,T =6 10 s,M=16,L=8,K=5 ues of , , and SNR, there is an optimum number of
10 0 b deleted elements from each row to achieve the best BER
EGC-IM (J=4)
performance. Our future research will consider the effect
EGC-IM (J=3) of the different distances from distributed nano‑machines
10 -1 EGC-IM (J=2)
EGC-IM (J=1) as well as the design of the extended IM schemes that are
EGC-IM (J=0)
ef icient for operation in this communication scenario.
10 -2
BER REFERENCES
-3
10
[1] G. K. Walia, D. K. K. Randhawa, and K. S. Malhi.
“A brief survey on molecular communications in
-4 nanonetworks”. In: 2016 International Conference
10
on Computational Techniques in Information and
Communication Technologies (ICCTICT). Mar. 2016,
-5
10
0 2 4 6 8 10 12 14 16 18 20 pp. 343–348.
SNR(dB)
[2] Y. Zamiri‑Jafarian, S. Gazor, and H. Zamiri‑Jafarian.
Fig. 6 – BER versus SNR performance of the MTH‑MoSK DMC systems “Molecular code division multiple access in nano
with the EGC‑IM, when different number of elements per row are re‑ communication systems”. In: 2016 IEEE Wireless
moved from the detection matrix.
Communications and Networking Conference. Apr.
Finally, Fig. 6 demonstrates the effect of the number of el‑ 2016, pp. 1–6.
ements removed from each of the rows in the detection [3] L. Wang and A. W. Eckford. “Nonnegative code divi‑
matrix , on the BER performance of the MTH‑MoSK sion multiple access techniques in molecular com‑
DMC systems with the proposed EGC‑IM. In the consid‑ munication”. In: 2017 15th Canadian Workshop on
ered MTH‑MoSK DMC systems, we assume that there are Information Theory (CWIT). June 2017, pp. 1–5.
= 5 nano‑machines supported. Note that, when = 0,
[4] S. Korte, M. Damrath, M. Damrath, and P. A. Hoeher.
the EGC‑IM is reduced to the conventional EGC. The re‑
“Multiple Channel Access Techniques for Diffusion‑
Based Molecular Communications”. In: SCC 2017;
1
When = 8, 16 and 32, the data rates are 3, 4, and 5 bits per symbol, 11th International ITG Conference on Systems, Com‑
respectively.
munications and Coding. Feb. 2017, pp. 1–6.

162 © International Telecommunication Union, 2021