Tsarin Ƙirƙirar Taurari Mai Girma don Tsarin Sadarwar Haske Mai Bayyane Mai Launuka Da Yawa

1. Gabatarwa & Bayyani

Wannan takarda, "Ƙirƙirar Taurari don Sadarwar Haske Mai Bayyane Mai Launuka Da Yawa," ta gabatar da ci gaba mai mahimmanci a fagen Sadarwar Haske Mai Bayyane (VLC). Masu rubutu sun ba da shawarar CSK-Advanced, sabon tsarin ƙirƙirar taurari mai girma wanda aka tsara don tsarin da ke amfani da fitilun LED Ja/Kore/Shuɗi (RGB LEDs). Aikin yana magance matsalolin mahimmanci na al'adar Canjin Launi (CSK), kamar asarar inganci daga ƙayyadaddun ƙarfin jimla, yayin da ake haɗa mahimman buƙatun haske kamar Ma'anar Fitar Launi (CRI) da Ƙimar Ƙarfin Haske (LER) a matsayin ƙayyadaddun ingantawa.

2. Cikakken Fahimta: Tsarin CSK-Advanced

Babban nasarar takardar ita ce wucewa daga kallon tashoshin RGB a matsayin masu ɗaukar kaya kawai. CSK-Advanced yana ɗaukar sararin siginar a matsayin haɗin kai, taurari mai girma inda kowace alama ta zama vector da ke ayyana ƙayyadaddun ƙarfi don fitilun LED ja, kore, da shuɗi lokaci guda. Wannan tsarin gabaɗaya yana ba da damar haɗin gwiwar ingantaccen aikin sadarwa (Ƙimar Kuskuren Bit - BER) da ingancin haske a ƙarƙashin ƙayyadaddun duniya kamar Ƙimar Ƙarfin Kololuwa zuwa Matsakaici (PAPR) na kowane LED. Canji ne daga tsarin ƙirƙira na matakin sashi zuwa na tsarin tsarin, mai kama da canjin tsarin da ingantaccen ƙarshe-zuwa-ƙarshe ya kawo a cikin tsarin koyo mai zurfi, kamar yadda aka gani a cikin ayyuka kamar takardar CycleGAN ta asali wadda ta koyi ayyukan taswira tsakanin yankunan hotuna tare.

3. Tsarin Hankali: Daga Matsala zuwa Magani

Takardar ta gina hujjarta tare da bayyananniyar ci gaba ta hankali mai matakai uku.

3.1. Tsarin Tsarin & Ƙirƙirar Tashoshi Masu Kyau

An kafa tushe tare da tsarin $N_r$, $N_g$, $N_b$ LEDs. An tsara babbar matsalar ingantawa don rage Ƙimar Kuskuren Alama (SER) ta hanyar ƙara Mafi ƙarancin Nisa na Euclidean (MED) tsakanin maki taurari a cikin sararin ƙarfi na 3D $(I_r, I_g, I_b)$. Muhimmanci, ƙuntatawa ba abin da ake yi bayan haka ba ne amma an haɗa su cikin ma'anar matsalar: ƙayyadaddun matsakaicin ƙarfin haske, maƙasudai na daidaitattun launuka don haskakawa, da iyakokin PAPR na gani na kowane don sarrafa murdiya mara layi a cikin kowane tashar launin LED.

3.2. Sarrafa Tsangwama Tsakanin Tashoshi (CwC)

Daga nan an faɗaɗa ƙirar zuwa yanayin aiki na tsangwama tsakanin tashoshin launi, wanda aka ƙirƙira ta hanyar matrix tashar $\mathbf{H}$. Maimakon yin daidaitawa a mai karɓa (bayan daidaitawa), wanda zai iya ƙara ƙara hayaniya, masu rubutu sun ba da shawarar mai daidaitawa kafin aiki wanda ya dogara da Rarraba Maɗaukaki (SVD). An sake ƙirƙira taurarin a cikin sararin tashar da aka canza, wanda aka raba. An nuna wannan tsarin mai himma yana fi dacewa da tsare-tsaren daidaitawa bayan aiki kamar Zero-Forcing (ZF) ko Linear Minimum Mean Squared Error (LMMSE), musamman a cikin yanayi mai hayaniya.

3.3. Lakabin Taurari tare da BSA

Mataki na ƙarshe yana magance taswirar jerin bit zuwa alamomin taurari. Masu rubutu sun yi amfani da Algorithm na Sauyin Binary (BSA)—ana ba da rahoton cewa a karon farko a cikin lakabin taurari mai girma na VLC—don nemo mafi kyawun taswira mai kama da Gray wanda ke rage BER don wani yanayin lissafi na taurari, yana rufe madauki akan ingantaccen aikin ƙarshe-zuwa-ƙarshe.

4. Ƙarfi & Kurakurai: Ƙima Mai Muhimmanci

Ƙarfi:

• Haɗin Ƙuntatawa Gabaɗaya: Sarrafa sadarwa (MED, BER), haskakawa (CRI, LER, wurin launi), da kayan aiki (PAPR) lokaci gida yana da misali kuma yana da alaƙa da masana'antu.
• Rage Tsangwama Mai Himma: Daidaitawar da ta dogara da SVD kafin aiki magani ne mai wayo kuma mai inganci ga matsalar aiki ta yau da kullun.
• Sabon Algorithm: Yin amfani da BSA don lakabi a cikin wannan mahallin shine wayayyun haɗin gwiwa daga ka'idar sadarwa ta dijital.

Kurakurai & Rashin ciki:

• Rikicin Lissafi: Takardar ba ta yi magana game da farashin lissafi na warware matsalar ingantaccen MED mai ƙuntatawa don manyan girman taurari ba, wanda zai iya zama shinge ga daidaitawar ainihin lokaci.
• Zato na Muhalli Mai Ƙarfi: Ƙirar tana ɗaukan tashar tsaye. Tashoshin VLC na cikin gida suna fuskantar toshewa da inuwa mai ƙarfi; ƙarfin tsarin ga irin waɗannan bambance-bambancen ba a gwada shi ba.
• Rashin Kayan Aiki: Yayin da ake la'akari da PAPR, sauran rashin daidaituwa kamar rashin layi na LED (fiye da yankewa) da tasirin zafi ba a ƙirƙira su ba, wanda zai iya ƙara fa'idar aiki.

5. Fahimta Mai Aiki & Hanyoyin Gaba

Ga masu bincike da injiniyoyi, wannan takarda tana ba da cikakkiyar tsari:

1. Ɗauki Tsarin Ingantawa Haɗin gwiwa: Yi la'akari da ƙirar tsarin VLC a matsayin haɗin gwiwar ingantaccen sadarwa da haskakawa, ba matsaloli biyu daban ba.
2. Daidaitawa Kafin Aiki Fiye da Bayan Aiki: A cikin yanayin tsangwama, saka hannun jari a cikin ƙirar karkatarwa/daidaitawa kafin aiki don ƙarin ingantaccen aiki.
3. Bincika Taurari Masu Daidaitawa: Mataki na gaba na hankali shine haɓaka algorithms masu sauƙi waɗanda za su iya daidaita taurarin a ainihin lokaci bisa ga canje-canjen buƙatun haskakawa ko yanayin tashar, watakila ta amfani da koyon inji don saurin ingantawa.
4. Tura Ƙa'idodi: Aiki kamar wannan yakamata ya sanar da sake maimaitawa na ƙa'idodin VLC na gaba (fiye da IEEE 802.15.7) don haɗa da ƙarin ma'anoni na taurari masu sassauƙa da ci gaba.

6. Zurfin Fasaha

6.1. Tsarin Lissafi

Za a iya taƙaita babban ingantaccen aiki don tashar da ta dace kamar haka: $$\begin{aligned} \max_{\{\mathbf{s}_i\}} & \quad d_{\min} = \min_{i \neq j} \|\mathbf{s}_i - \mathbf{s}_j\| \\ \text{s.t.} & \quad \frac{1}{M}\sum_{i=1}^{M} \mathbf{s}_i = \mathbf{P}_{\text{avg}} \quad \text{(Matsakaicin Ƙarfi)} \\ & \quad \mathbf{C}(\mathbf{s}_i) = \mathbf{c}_{\text{target}} \quad \text{(Wurin Launi)} \\ & \quad \max(\mathbf{s}_i^{(k)}) / \text{avg}(\mathbf{s}_i^{(k)}) \leq \Gamma_{\text{PAPR}} \quad \forall k \in \{r,g,b\} \end{aligned}$$ inda $\mathbf{s}_i = [I_r, I_g, I_b]_i^T$ shine maƙasudin taurari, $M$ shine girman taurari, kuma $\mathbf{C}(\cdot)$ yana lissafin daidaitattun wurin launi.

6.2. Sakamakon Gwaji & Aiki

Takardar ta gabatar da sakamakon lamba da ke nuna fifikon CSK-Advanced:

• BER vs. SNR: Ƙarƙashin launukan haskakawa marasa daidaituwa (misali, ja mai rinjaye), CSK-Advanced yana samun BER mai ƙasa sosai idan aka kwatanta da tsare-tsaren PAM da aka raba na al'ada da CSK na asali, musamman a matsakaici zuwa babban SNR.
• Ƙarfin Tsangwama: Ƙirar daidaitawa kafin aiki wanda ya dogara da SVD yana nuna gibin aikin BER a sarari akan daidaitawar bayan aiki na ZF da LMMSE, musamman yayin da tsangwamar tsangwama ke ƙaruwa. Ana wakilta wannan a cikin tsarin zane na BER vs. ma'aunin tsangwama.
• Zane-zanen Taurari: Takardar mai yiwuwa ta haɗa da zane-zane masu watsawa na 3D waɗanda ke nuna maki taurari masu ingantaccen lissafi don CSK-Advanced, suna bambanta su da ƙarin grid na yau da kullun amma marasa inganci na tsare-tsare na al'ada. Waɗannan zane-zane suna nuna babban MED da aka samu ta hanyar ingantawa.

7. Tsarin Bincike & Misalin Hali

Hali: Ƙirƙirar tsarin VLC don ɗakin nunin gidan kayan tarihi.

1. Bukatu: Haskaka zane tare da ƙayyadaddun zafin jiki na musamman, wanda aka tsara (misali, 3000K dumi fari) don hana lalacewa, yayin samar da ɓoyayyen rafi na bayanan jagorar sauti.
2. Aiwatar da Tsarin CSK-Advanced:
   • Ma'anar Ƙuntatawa: Saita $\mathbf{c}_{\text{target}}$ zuwa daidaitattun launuka da ake buƙata. Ayyana iyakokin PAPR mai tsauri don tabbatar da tsawon rayuwar LED. Saita ƙuntatawar CRI mai girma don ingantaccen fitar da launi.
   • Ƙirar Tashar: Auna/kimanta matrix tsangwama 3x3 $\mathbf{H}$ don ƙayyadaddun kayan aikin LED RGB da na'urorin gano hoto da aka yi amfani da su.
   • Ingantawa: Gudanar da haɓaka MED tare da ƙuntatawa na sama kuma a daidaita ta amfani da SVD bisa $\mathbf{H}$.
   • Lakabi: Aiwatar da BSA ga sakamakon taurari na 3D don taswirar bit ɗin bayanan sauti don rage kurakuran sake kunnawa.
3. Sakamako: Tsarin haskakawa wanda ya cika ma'auni na ingantaccen haske na kiyayewa yayin da yake watsa bayanai cikin aminci, wanda ke da wahala tare da ƙirar da aka raba.

8. Duban Aikace-aikace & Bincike na Gaba

Aikace-aikace Nan da Nan: Haɗin kai masu sauri, masu tsaro a cikin wuraren da ke da haske mai mahimmanci: asibitoci (ɗakunan MRI), ɗakunan jiragen sama, saitunan masana'antu tare da ƙuntatawar EMI. Hanyoyin Bincike na Gaba:

• Koyon Injin don Ingantawa: Yi amfani da koyo mai zurfi ko koyo mai tushen gradient (wanda aka yi wahayi daga tsare-tsare kamar PyTorch/TensorFlow) don warware rikicin ƙuntatawa mai rikitarwa da sauri ko daidaitawa.
• Haɗin kai tare da Cibiyoyin Sadarwar LiFi: Ta yaya CSK-Advanced ke aiki a cikin cibiyoyin sadarwa masu amfani da yawa, sel da yawa na LiFi? Ana buƙatar bincike cikin rarraba albarkatu da sarrafa tsangwama.
• Bayan RGB: Faɗaɗa tsarin zuwa fitilun LED masu yawa (misali, RGB + Fari, ko Cyan) don maɗaukaki mafi girma da ƙimar bayanai.
• Haɗin kai tare da Fasahar Hoton Silicon: Bincika haɗin gwiwar ƙira tare da sabbin dandamali na micro-LED da hoton silicon don masu watsawa masu ƙarfi, masu sauri, kamar yadda ƙungiyoyin bincike kamar Cibiyar Ƙirƙirar Hoton Haɗin kai ta Amurka (AIM Photonics) suka ruwaito.

9. Nassoshi

1. Gao, Q., Gong, C., Wang, R., Xu, Z., & Hua, Y. (2014). Ƙirƙirar Taurari don Sadarwar Haske Mai Bayyane Mai Launuka Da Yawa. arXiv preprint arXiv:1410.5932.
2. Ma'auni na IEEE don Gida da Cibiyoyin Sadarwar Birane–Sashi na 15.7: Sadarwar Waya mara waya ta Gajeren zango ta amfani da Haske Mai Bayyane. (2011). IEEE Std 802.15.7-2011.
3. Zhu, J., Park, T., Isola, P., & Efros, A. A. (2017). Fassarar Hotuna-zuwa-Hotuna mara Haɗin gwiwa ta amfani da Cibiyoyin Adawa na Haɗin kai. Proceedings of the IEEE International Conference on Computer Vision (ICCV). (Nassoshi na CycleGAN don kwatancen ingantaccen haɗin gwiwa).
4. Kahn, J. M., & Barry, J. R. (1997). Sadarwar infrared mara waya. Proceedings of the IEEE, 85(2), 265-298.
5. AIM Photonics. (n.d.). Binciken Hoton Haɗin kai. An samo daga https://www.aimphotonics.com/ (Misalin dandamali na kayan aiki mai ci gaba).
6. Drost, R. J., & Sadler, B. M. (2014). Ƙirƙirar taurari don maɓalli mai canza launi ta amfani da algorithms na billiards. IEEE GLOBECOM Workshops.