Sun’iy Intellekt va Raqamli Iqtisodiyot (SIRI 2025)

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 5G TARMOG‘INI SEGMENTLASH KONTSEPSIYASI, TARIXI VA TAMOYILLARI Muxammedinov Kobeysin Kuanishovich Muhammad al-Xorazmiy nomidagi Toshkent axborot texnologiyalari universiteti tayanch doktoranti, kobiyaga@gmail.com Annotatsiya: Texnologiyaning rivolanishi va telekommunikatsiya tarmoqlariga ulanishlar va malumot almashish hajmining o‘sishi hozirgi telekommunikatsiya tarmog‘ining yetishmasligi muammosiga olib keladi. Bunga qo‘shimcha beshinchi avlod texnologiyalarining rivojlanishi, yangi funksiya va xizmatlarga moslashuvchan tarmoq texnologiyasi sifatida dasturiy konfiguratsiyalanadigan tarmoq tamoili rivojlanib bormoqda. Bu sohada dunyo tajribalarini o‘rganishda 5G tarmoqlarida segmentlarga ajratib xizmat ko‘rsatish tamoilidagi tarmoqni segmentlash tushunchasini chuqur o‘rganish ushbu ishning maqsadi hisoblanadi. Keyingi ishlarimizda tarmoqni segmentlashni dasturiy konfiguratsiyalanadigan tarmoqlar uchun o‘rganib chiqamiz. Kalit so‘zlar: 5G tarmoq, Dasturiy konfiguratsiyalanadigan tarmoq, Tarmoqni segmentlash, Tarmoq funktsiyalarini virtualizatsiyalash Kirish 1960-yillardan boshlab tarmoqni segmentlash(Network Slicing)[1] tushunchasi vaqt almashish(time-sharing) va virtual xotirani(virtual memory) qoʻllab-quvvatlovchi birinchi IBM operatsion tizimi (CP-40) dizaynidan soʻng virtualizatsiya kontsepsiyalariga tayangan[2]. Bunday dizayn bir vaqtning o‘zida o‘n beshtagacha foydalanuvchini joylashtirish imkoniyatiga ega bo‘lgan tizimni taqdim etdi va shaxsga ham apparat, ham dasturiy ta’minotning alohida to‘plamida mustaqil ishlashga ruxsat berilishi mumkin edi[3]. o‘shandan beri jismoniy qurilmadan virtual obyekt yaratilishi mumkin bo‘lgan tarmoqni virtualizatsiya qilish g‘oyasi shakllandi. Network Slicing inglizchadan, Slice atamasi o‘zbek tiliga segment sifatida tarjima qilinganligi sababli, matnda tarmoq segmentlash atamasi qo‘llaniladi. Tarmoq segmentatsiyasi jismoniy tarmoqni maxsus tarmoq talablaridan kelib chiqqan holda talabga binoan o‘rnatiladigan bir nechta dasturga xos virtual tarmoqlarga bo‘lishni o‘z ichiga oladi. Tarmoq segmentatsiyasi tarmoq funksiyasini virtualizatsiya texnologiyasi va DKT asosida amalga oshiriladi. Ko‘rish(Vizion) virtual tizimlarni turli tarmoq resurslari, hisoblash infratuzilmalari va saqlash qurilmalari bo‘ylab qamrab olish edi[2]. 1970-1980 yillarda tarmoq virtualizatsiyasi maʼlumotlar markazlarida keng qoʻllanila boshlandi, bu yerda masofaviy saytlar Internet orqali xavfsiz va boshqariladigan ishlash bilan bogʻlangan. 1980-yillarning oxirida umumiy jismoniy infratuzilma orqali ishlaydigan virtual tarmoqni yaratish uchun tarmoq tugunlari mantiqiy havolalar orqali ulangan overlay tarmoqlar taklif qilindi. Overlay tarmoqlari tarmoqni segmentlash kontsepsiyasining dastlabki shaklidir, chunki u turli xil ma’muriy domenlarda turli resurslarni birlashtiradi va yakuniy foydalanuvchilarga xizmat sifati(QoS)ni kafolatlaydi. Ustki(Overlay) tarmoqlar moslashuvchan bo‘lsa da, ular tarmoq boshqaruvida avtomatlashtirish va dasturlash 261

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot xususiyatlariga ega emas. 1990-yillar davomida va 2000-yillarning boshlarida tugun(node) operatsion tizimi resurslarni boshqarish tizimlarini ta’minlay oladigan faol va dasturlashtiriladigan tarmoq taklif qilindi. Shundan so‘ng, yangi tarmoq protokollarini tekshirish va baholash mumkin bo‘lgan turli platformalar hamda mustaqil tarmoq(federative) sinov muhitlari tashkil etildi. 2008 yilda AQSH Milliy Fanlar Jamg‘armasi loyihasi tarmoqni virtualizatsiya kontsepsiyalariga asoslangan GENI sinov maydonchasini taqdim etdi[4]. Maqsad, federatsiyalangan resurslar va mobil tarmoq muhitini hisobga olgan holda toza tarmoq bo‘yicha tadqiqotlarni ilgari surish edi. Beshinchi avlod tarmog‘ini segmentlash misoli 1-rasmda keltirilgan. 1-rasm. Beshinchi avlod tarmog‘ini segmentlash misoli GENI umumiy tarmoq sinov maydonchasi boʻlib, bir vaqtning oʻzida bir nechta tajribalarni oʻtkazishi mumkin. Ushbu tendensiyadan so‘ng, 2009-yilda DKT tadqiqotchilarga ochiq interfeyslar orqali dasturlash imkoniyatlaridan foydalaniladigan kampus tarmog‘ining tarmoq segmentida tajriba o‘tkazish imkonini berdi [5]. Beshinchi avlod tarmog‘ini segmentlash ixtiro qilingan va birinchi navbatda Keyingi avlod mobil tarmog‘i (NGMN - Next Generation Mobile Network) tomonidan taqdim etilgan[6]. NGMN tomonidan ta’riflanganidek, tarmoq segmenti - bu umumiy (jismoniy yoki virtual) infratuzilmada ishlaydigan, o‘zaro izolyatsiya qilingan, mustaqil nazorat va boshqaruvga ega bo‘lgan E2E mantiqiy tarmog‘i/bulut bo‘lib, talabga binoan yaratilishi mumkin. Tarmoq segmenti bir xil yoki turli ma’muriyatlardagi alohida domenlardagi o‘zaro faoliyat domen komponentlaridan yoki kirish tarmog‘iga, transport tarmog‘iga, asosiy tarmoqqa va chekka tarmoqlarga tegishli komponentlardan iborat bo‘lishi mumkin. Shunday qilib, tarmoq segmentlari ko‘p xizmat va ko‘p xizmatni qo‘llab-quvvatlash uchun mustaqil, o‘zaro izolyatsiya qilingan, boshqarilishi va dasturlashtirilishi mumkin. Tarmoqni segmentlash kontsepsiyasi umumiy jismoniy infratuzilma platformasi ustidagi bir nechta mantiqiy va mustaqil tarmoqlarni osonlashtirishi mumkin[7]. o‘shandan beri turli standartlashtirish organlari tarmoqni segmentlashning ta’rifini boshqa nuqtai nazardan o‘rganib chiqdilar. ITU (International Telecommunication Union – Xalqaro telekommunikatsiya ittifoqi) tarmoqni segmentlashni tarmoq dasturiy ta’minotining asosiy tushunchasi sifatida ko‘zda tutadi, bu bir nechta virtual resurslardan tashkil topgan, ajratilgan va dasturlashtiriladigan boshqaruv va ma’lumotlar tekisligi bilan jihozlangan Mantiqiy izolyatsiyalangan tarmoq segmentlarini osonlashtiradi[8]. 3GPP tarmoqni segmentlashni “operatorga turli xil talablarni qo‘yadigan turli bozor ssenariylari uchun optimallashtirilgan echimlarni taqdim etish uchun moslashtirilgan tarmoqlarni yaratish imkonini beruvchi texnologiya” sifatida belgilaydi[9]. 262

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 Biznes nuqtai nazaridan, segment muayyan biznes ishi yoki xizmatini bajarish uchun zarur bo‘lgan barcha tegishli tarmoq resurslari, funksiyalari va aktivlari, jumladan, OSS(Operational Support Systems - Operatsion qo‘llab-quvvatlash tizimlari), BSS(Business Support Systems - Biznes qo‘llab-quvvatlash tizimlari) va DevOps(Development Operations) yani dasturiy ta’minot ishlab chiqish va IT operatsiyalarni integratsiya qilish jarayonlarini o‘z ichiga oladi. Ikki xil segment mavjud: 1) provayderning ichki xizmatlari uchun foydalaniladigan, ularni to‘liq nazorat qilish va boshqarishni o‘zida saqlaydigan ichki qismlar; 2) mijozlarga xizmat ko‘rsatadigan, mijozga ajratilgan tarmoqlar/bulutlar/ma’lumotlar markazlari sifatida ko‘rinadigan tashqi bo‘limlar. Tarmoqni segmentlash ilova provayderlariga yoki jismoniy tarmoq infratuzilmasi bo‘lmagan turli vertikal segmentlarga radio, bulut va tarmoq resurslarini taklif qilishi mumkin. Shunday qilib, u xizmat turidan kelib chiqib, mijozlar talablarini qondirish uchun tarmoq ishini moslashtirish orqali xizmatni farqlashni ta’minlaydi[10]. Beshinchi avlod dasturiy konfiguratsiyalanadigan tarmoqlarda tarmoqlarni segmentlash va ular bilan bog‘liq ishlashni o‘z ichiga olgan asosiy tamoyillar quyidagilar[7]: Tarmoq ishini avtomatlashtirish: tarmoq segmentlarini avtomatlashtirish (masalan, joylashtirish, oʻzgartirish, oʻchirish), tarmoq resurslarini optimallashtirish (avtomatik masshtablash/migratsiya/avtomatik tiklash), shuningdek boshqaruv va maʼlumotlar tekisliklari oʻrtasidagi dinamik oʻzaro taʼsirni dinamik ravishda boshqarish imkonini beradi. • Yuqori ishonchlilik, masshtablilik va izolyatsiya: Bular 5G tarmog‘ini segmentlashning asosiy xususiyatlari bo‘lib, turli ishlash talablari bo‘lgan xizmatlar uchun nosozliklarni darhol aniqlash mexanizmlaridan foydalangan holda har bir ijarachi uchun ishlash kafolatlari va xavfsizligini ta’minlaydi. • Dasturlash imkoniyati: Dasturlash xizmatlarni taqdim etishni, tarmoqlarni boshqarishni va integratsiya va operatsion muammolarni, ayniqsa aloqa xizmatlarini qo‘llabquvvatlashni soddalashtiradi. Masalan, u uchinchi tomonlarga ajratilgan boʻlak resurslarini (masalan, tarmoq imkoniyatlarini ochib beruvchi ochiq API’lardan foydalangan holda tarmoq va bulutli resurslarni) boshqarish imkonini beradi. Bu, oʻz navbatida, 5G dasturiy taʼminoti va virtualizatsiya qilingan tarmoqlarda talab boʻyicha xizmatga yoʻnaltirilgan moslashtirish va resurs egiluvchanligini osonlashtiradi. • Ierarxik abstraksiya: Tarmoqni segmentlash mantiqiy ravishda yaratish orqali qo‘shimcha abstraksiya qatlamini yoki tarmoq resurslarining jismoniy alohida guruhlari va virtual tarmoq funktsiyalari konfiguratsiyasi kiritadi. Ushbu virtuallik avvalgisidan yuqori bo‘lgan tarmoq segmenti xizmatidan xizmat ko‘rsatishni osonlashtiradi. Masalan, tarmoq operatorlari va internet provayderlari boshqa sanoat kompaniyalariga tarmoqlardan oʻz xizmatlarining bir qismi sifatida foydalanishlari uchun tarmoqni segmentlashdan foydalanishlari mumkin (masalan, vertikal oʻyinchilar yuqori ishonchli tarmoqqa ulangan avtomobil, oʻta past kechikish bilan onlayn oʻyin, kafolatlangan tarmoqli kengligi bilan video oqim va h.k.). • Segment moslashuvi: Segment moslashuvi ma’lumotlar va boshqaruv tekisligini ajratuvchi DKT yordamida virtual tarmoq topologiyasining barcha qatlamlarida amalga oshiriladi. Ma’lumotlar tekisligida NFV(Network Functions Virtualization-Tarmoq • 263

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot funktsiyalarini virtualizatsiyalash) imkoniyatlari xizmatga moslashtirilgan tarmoq funktsiyalari va ma’lumotlarni uzatish mexanizmlarini taqdim etadi, bu erda qo‘shimcha xizmatlarni Sun’iy intellekt (SI) yordamida yoqish mumkin. Shuni ta’kidlash kerakki, xususiylashtirish ma’lum bir 5G ijarachisiga ajratilgan tarmoq resurslaridan ma’lum bir xizmat talablariga javob berish uchun samarali foydalanishni ta’minlaydi. Tarmoq resurslarining egiluvchanligi: Tarmoq resurslarining egiluvchanligi samarali va buzilmaydigan qayta ta’minlash mexanizmi orqali amalga oshiriladi, bunda ajratilgan resurslar oshiriladi/pasaytiriladi. Shunday qilib, elastiklik geografik joylashuvidan qat’i nazar, foydalanuvchilarning istalgan SLA/ELA (Service Level Agreement - Xizmat darajasi kelishuvi va Enterprise License Agreement - Korporativ litsenziya kelishuvi)lariga erishishni ta’minlaydi. • DKT Kontroller 2-ijarachi VM 3 1-ijarachi VM 1 VM 2 3-ijarachi VM 4 Hisoblash klasteri OF kommutataor Virtual L2 1-segment Hisoblash klasteri OF kommutator Virtual L2 2-segment 2-ijarachi VM 3 1-ijarachi VM 1 VM 2 3-ijarachi VM 4 2-ijarachi VM 11 1-ijarachi VM 5 VM 6 Hisoblash klasteri OF kommutator Asosiy tarmoq 2-rasm: Tarmoqni segmentlashdan foydalanish holati [11] 2-rasmda tarmoqni segmentlashning amaliy ssenariysi ko‘rsatilgan. Ushbu kontekstda ijarachi bir yoki bir nechta Virtual Infratuzilma (VI) yoki tarmoq xizmatlariga egalik qiluvchi va ulardan foydalanadigan mantiqiy shaxs sifatida aniqlanadi. Shunday qilib, u o‘zining substrat tarmog‘i orqali VI-larni ajratishi va turli ijarachilarga xizmatlarni taklif qilish uchun bir nechta L2 tarmoq segmentlarini taqdim etishi mumkin. Mobil virtual tarmoq operatori kabi har bir ijarachi tarmoq segmentiga egalik qiladi va boshqaradi. Shu nuqtai nazardan, virtual L2 1-segment 1-ijarachi va 2-ijarachiga tegishli. Shuni ta’kidlash muhimki, boshqa ijarachilar har qanday ijarachining infratuzilmasidan foydalanishi mumkin. Adabiyotlar 1. S. Nanda and T. Chiueh, “A Survey on Virtualization Technologies.” https://courses.cs.vt.edu/~cs5204/fall07-kafura/Papers/Virtualization/SurveyVirtualization-Technologies-TR.pdf 2. R. Goldberg, “Survey of Virtual Machine Research,” IEEE Computer , vol. 7, pp. 34– 45, 1974. 3. A. Lindquist, R. Seeber, and L. Comeau, “A Time-Sharing System using an Associative Memory,” IEEE Proceedings , vol. 54, pp. 1774–1779, Dec 1966. 264

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 4. B. Mark, J. S. Chase, O. Max, R. D, R. Robert, and S. I, “GENI: A Federated Testbed for Innovative Network Experiments,” Computer Networks , vol. 61, no. 1, pp. 5–23, 2014. 5. A. Lara, A. Kolasani, and B. Ramamurthy, “Network innovation using OpenFlow: A survey,” IEEE Communication Surveys Tutorials , vol. 16, no. 1, pp. 493–512, 2014. 6. “NGMN Alliance, 5G WHITE PAPER,” Feb 2015. 7. I. Afolabi, A. Ksentini, M. Bagaa, T. Taleb, M. Corici, and A. Nakao, “Towards 5G Network Slicing over Multiple-Domains,” in IEICE Transactions on Communications,pp.1992–2006, 2017. 8. “ITU-T Y.3011, Framework of Network Virtualization for Future Networks,” January 2012. 9. X. de Foy and A. Rahman, “Network Slicing -3GPP Use Case draft-defoy-netslices3GPP-network-slicing-02,” InterDigital Communications, LLC, Dec 2017. 10. J. Menglan, C. Massimo, and T. Mahmoodi, “Network slicing management & prioritization in 5G mobile systems,” European Wireless 2016; 22th European Wireless Conference, May 2016. 11. A. Hakiri, A. Gokhal, P. Berthou, D. C. Schmidt, and T. Gayraud, “Software-Defined Networking: Challenges and research opportunities for Future Internet,” Computer Networks , vol. 75, pp. 453–471, Dec 2014. 265

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot TA’LIM JARAYONLARINI TAHLIL QILISH UCHUN SUN’IY INTELLEKT MODELLARINI ISHLAB CHIQISH Melibayeva Xilolaxon Numonjon qizi1, Xolmatov Numonjon Meliboyevich2, Umarov Muhriddin Abduxalil o‘g‘li1 1“University of management and future technologies” universiteti, Kommunikatsiya va raqamli texnologiyalar kafedrasi, Toshkent, 100208, O‘zbekiston; numanjanovnaxilolaxon@gmail.com, muhriddin.umarov.1992@gmail.com 2Axborot Kommunikatsiya Texnologiyalari va Aloqa Harbiy Instituti dotsent, numonjon5058@gmail.com Annotatsiya. Ushbu maqola ta’lim jarayonlarini tahlil qilish uchun sun’iy intellekt (SI) modellarining evolyutsiyasini o‘rganadi. U taʼlim sohasida maʼlumotlarni toʻplash, baholash va qarorlar qabul qilishni yaxshilash uchun sunʼiy intellektga asoslangan tizimlarning imkoniyatlarini tahlil qiladi. Tadqiqot asosiy yondashuvlar, to‘siqlar va o‘qitish va ta’lim samaradorligini oshirishda SI ning istiqbolli afzalliklariga urg‘u beradi. Natijalar SIni ta’lim sektoriga kiritish bo‘yicha joriy tadqiqotni yanada kengaytiradi. Kalit so‘zlar: Sun’iy intellekt, ta’limni baholash, SI modellari, ma’lumotlarga asoslangan ta’lim, intellektual tizimlar. Kirish So‘nggi yillarda sun’iy intellekt (SI) ko‘plab sohalarni, jumladan, ta’limni ham o‘zgartirdi. Katta ma’lumotlar va mashinani o‘rganish metodologiyalarining tobora ortib borayotgani ta’lim jarayonlarini tahlil qilish va yaxshilashga qodir bo‘lgan SI modellarini yaratishga yordam berdi. Ushbu modellar o‘qituvchilarga o‘quvchilar faoliyatini baholash, ta’lim natijalarini prognozlash va real vaqt ma’lumotlari asosida o‘qitish strategiyalarini sozlashda yordam beradi. Taʼlimga sunʼiy intellekt asosidagi tahliliy integratsiya bir qator afzalliklarni beradi, jumladan, qarorlar qabul qilishning yaxshilanishi, oʻquv muammolarini erta aniqlash va faollikni oshirish. Shunga qaramay, ma’lumotlarning maxfiyligi, algoritmik tarafkashlik va sun’iy intellekt asosidagi ta’lim bo‘yicha malakali mutaxassislarga bo‘lgan talab kabi muammolar jiddiy muammolar bo‘lib qolmoqda. Ushbu maqola ta’lim jarayonlarini tahlil qilish uchun mo‘ljallangan sun’iy intellekt modellarining evolyutsiyasini, shu jumladan ularning usullari, qo‘llanilishi va zamonaviy ta’limga ta’sirini o‘rganishdan iborat. 1-rasm. Intellekt uchligi. 266

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 Ayni paytda mamlakatimizda ta’lim sohasiga katta e’tibor qaratilmoqda. Ushbu tizimni takomillashtirish borasidagi say-harakatlar ham katta. Hozirgi texnologiyalarni ushbu tizimga integratsiyalash, bu orqali ta’lim tizimini zamonaviy standartlar va talablarga moslashtirish borasidagi say-harakatlar bu g‘oyalarning dalilidir. o‘rta va oliy ta’limda sun’iy intellekt (SI) ilovalarini ishlab chiqish va ulardan foydalanish o‘quvchilarning o‘rganish va o‘qituvchilar tomonidan qo‘llaniladigan o‘qitish usullarini sezilarli darajada o‘zgartirishi mumkin. SI ilovalari talabalar uchun moslashtirilgan o‘rganish tajribasini taqdim etishi va ta’lim sohasidagi muammolarga yangi yechimlarni taqdim etishi mumkin [1]. Sun’iy intellektning asosiy xususiyati uning muayyan maqsadga erishish ehtimolini maksimal darajada oshiradigan faoliyatni baholash va amalga oshirish qobiliyatidir. Sun’iy intellekt va kompyuter texnologiyalarining ta’limga integratsiyalashuvi zamonaviylashayotgan dunyoda tobora muhim ahamiyat kasb etmoqda. Taʼlimni raqamlashtirish ushbu soha rivojiga xizmat qilishi shubhasiz. Adabiyotlar tahlili va metodologiya. Ayni paytda mamlakatimizda ta’lim sohasiga katta e’tibor qaratilmoqda. Ta’lim sohasida ham muhim islohotlar amalga oshirilmoqda. Ta’lim tizimi samaradorligini oshirish va bu borada xalqaro hamkorlikni yo‘lga qo‘yish bo‘yicha ishlar olib borilmoqda. Hozirgi vaqtda o‘quv jarayoniga interfaol usullar, innovatsion texnologiyalar, pedagogik va axborot texnologiyalarini integratsiyalashuviga qiziqish ortib bormoqda. Ayni paytda ta’lim jarayonini takomillashtirish va bu sohani zamon talablari asosida standartlashtirish bo‘yicha jiddiy ishlar olib borilmoqda. Ushbu tartib-qoidalarni modernizatsiya qilish bizga global bilimlardan foydalanish va zamonaviy yutuqlarga moslashish imkonini beradi. Sun’iy intellekt: Zamonaviy yondashuv, Styuart Jonatan Rassell va Piter Norvig mualliflari bo‘lgan sun’iy intellekt bo‘yicha yuqori baholangan darslik dastlab 1995-yilda nashr etilgan, uning to‘rtinchi nashri 2020-yilda chiqarilgan. Sun’iy intellekt aqlli shaxslarni tushunishga va yangi intellektlarni yaratishga intiladi. “Kelajakni aniq bashorat qilib bo‘lmasa-da, inson darajasidagi (yoki undan yuqori) aql-zakovatga ega bo‘lgan kompyuterlar bizning kundalik hayotimizga va sivilizatsiya taraqqiyotiga sezilarli ta’sir ko‘rsatishi aniq”, — deb ta’kidlaydilar ular [4]. Ayni paytda ta’limni rivojlantirish bo‘yicha xalqaro tajribalardan ilhom olish, ularni amaliyotda ijodiy qo‘llash va zamon talablariga moslashtirish bo‘yicha tashabbuslar ilgari surilmoqda. Hujjatlar O‘zbekiston Respublikasi Prezidentining 2016-yil 29-dekabrdagi PQ2707-sonli “2017-2021-yillarda maktabgacha ta’lim tizimini yanada takomillashtirish chora-tadbirlari to‘g‘risida”gi, 2017-yil 9-sentabrdagi PQ-3261-sonli “Ta’lim tizimini tubdan takomillashtirish va dekreatsiya qilish chora-tadbirlari to‘g‘risida”gi qarorlaridan iborat. 2017-yil 30-sentabrdagi “Maktabgacha ta’lim tizimini boshqarishni tubdan takomillashtirish chora-tadbirlari to‘g‘risida”gi PF-5198-son. Natijalar va muhokama. Ushbu omillarni hisobga olgan holda, turli sohalarda sun’iy intellekt texnologiyalaridan foydalangan holda innovatsion biznes modellari, mahsulotlar va xizmatlar ko‘rsatish usullarini rivojlantirish uchun qulay ekotizim yaratish va ularni belgilangan ustuvor sektor va sohalarda amalga oshirish bo‘yicha tashabbuslar boshlandi. Buning mantiqiy sababi shundaki, sun’iy intellekt texnologiyalari bizning zamonaviy istiqbolimiz orqali yuqori natijalarga erishishga imkon beradi. Hozirgi ulkan o‘sish davrida 267

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot kompyuterlar va boshqa ilg‘or texnologiyalarning ahamiyati beqiyos. Hozirgi vaqtda barcha mamlakatlarda zamonaviy texnologiyalar va sun’iy intellekt sohasini rivojlantirishga, shuningdek, ushbu texnologiyalarni tadqiqot va ta’limga integratsiyalashuviga katta e’tibor qaratilmoqda. Ushbu yo‘nalish ushbu sohalarning rivojlanishini shubhasiz qo‘llabquvvatlaydi. Axborot texnologiyalarining ta’lim tizimiga jadal integratsiyalashuvi zamonaviy texnologiyalarni qo‘llash ko‘lamini kengaytirmoqda. Shu bilan birga, ta’limda zamonaviy axborot texnologiyalarining rivojlanish tendensiyalarini ta’kidlash mumkin. Bu jarayonlar bizga samarali natijalar va yutuqlarga erishish imkonini beradi. Pedagogik texnologiyalarning rivojlanishi ularning nazariy asoslarini tanqidiy tahlil qilish va ta’lim jarayoniga samarali yondashuvni aniqlashga bog‘liq. Muayyan vaqt oralig‘ida pedagogik vositalarning nazariy asoslarini o‘zgartirish orqali ularning ,,muammo sohasini “ o‘zgartirish maqsadga muvofiq bo‘ladi. o‘quv jarayonini texnologik usul bilan qurish muayyan pedagogik vazifalarni ijobiy hal etishni taqozo etadi [5]. Ushbu strategiyaning ta’lim tizimiga tatbiq etilishi ijobiy natijalar berdi. Yangi ixtirochilik dasturi doirasida o‘qitish zamonaviy ta’limning asosiy strategiyasi sifatida e’tirof etilmoqda. o‘quv dasturiy ta’minotining o‘quv quroli, o‘rganish asosi va axborotni qayta ishlash usuli sifatidagi funktsiyalarini taqdim etish. o‘quvuslubiy majmualarni ishlab chiqishda hisoblash vositalari bilan bir qatorda o‘quv va ko‘rgazmali apparatlarning funksiyalarini kiritish. Bunday komplekslardan foydalanish o‘quvchiga o‘rganilayotgan jarayonga oid ma’lumotlarni to‘plash va saqlashga yordam beradi, shuning uchun bu jarayonlarning qonuniyatlari va mohiyatini ochib beradi. Bu esa o‘quvchining mustaqil ma’lumotlarga ega bo‘lishi bilan birga intellektual va ijodiy qobiliyatlarini rivojlantirish imkonini beradi. Multimedia tizimlarini ishlab chiqishda kompyuterlar va audio-vizual uzatish texnologiyalarining funksiyalarini integratsiyalash. Dasturiy-texnik vositalar va qurilmalar to‘plamini o‘z ichiga olgan ushbu tizimlar ma’lumotlarning bir nechta shakllarini (matn, grafik, tovush, rasmlar) birlashtiradi va foydalanuvchi bilan interaktiv aloqani ta’minlaydi [6]. Buning oqibatlari haqiqatan ham aniq. Hozirgi vaqtda ko‘plab olimlar sun’iy intellekt texnologiyasini takomillashtirish orqali modernizatsiya qilish tarafdori. ,,Sun’iy intellekt” bo‘yicha o‘rganish natijasi - bu miya funksiyalarini tushunishga, inson psixikasining sirlarini ochishga va insonga o‘xshash intellektni namoyish etuvchi mashinalarni ishlab chiqishga intilish. Intellektual jarayonlarni modellashtirishning asosiy potensiali aniq semantika va cheklangan lug‘atga ega tilda ifodalangan har qanday miya funksiyasini yoki aqliy faoliyatini elektron raqamli kompyuterga o‘tkazish qobiliyatidadir. Ushbu yondashuv texnologiya va insoniyatni birlashtiradi, texnologiya vazifalarni bajaradi va inson mehnatini kamaytiradi. SI ilovalari talabalar uchun individual ta’lim tajribasini osonlashtirishi mumkin. Talabalar ma’lumotlarini, shu jumladan test natijalari, uy vazifalari va o‘rganish afzalliklarini tahlil qilish orqali SI o‘quvchilarning maksimal salohiyatiga erishishda yordam berish uchun moslashtirilgan fikr-mulohazalar va ta’lim resurslarini taqdim etishi mumkin .[2] Moslashuvchan o‘rganishni osonlashtirish: SI ilovalari o‘quvchilarning individual ishlashiga qarab qiyinchilik darajasi va o‘rganish tezligini o‘zgartiradigan moslashuvchan ta’lim tizimlarini ishlab chiqish uchun ishlatilishi mumkin. Bu o‘quvchilarning faolligi va motivatsiyasini saqlab qolishga yordam beradi, ularning o‘z mahoratiga mos keladigan darajada o‘rganishini ta’minlaydi [7]. Ushbu ta’limda amalga oshirilgan harakatlar tizimli ravishda tartibga solinadi va rejalashtirilgan. o‘quv jarayonida talabalarga keng ko‘lamli 268

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 yordam: SI ilovalari o‘quvchilarga turli yondashuvlar, jumladan repetitorlik va maslahat xizmatlari orqali yordam berishi mumkin. Sun’iy intellektga asoslangan chatbotlar o‘quv ishlarida yordamga muhtoj yoki ruhiy salomatlik bilan bog‘liq muammolarga duch kelgan talabalarga kechayu kunduz yordam ko‘rsatishi mumkin. Bu yerda ushbu tizimning foydali xususiyatlari paydo bo‘ladi. U barcha nuqtai nazarlarni hisobga olgan holda ishlaydi. AIning afzalliklari hozirda shubhasizdir. Ushbu tizimni ishlab chiqish va amaliyotga keng joriy etish talab etiladi. Ma’muriy samaradorlikni oshirish: SI dan baholash va davomat monitoringi kabi ma’muriy funksiyalarni soddalashtirish uchun foydalanish mumkin. Bu o‘qituvchilarning o‘qitish va talabalarga yo‘l-yo‘riq berishga bag‘ishlangan vaqtini tejashga yordam beradi. Vaqt masalasi bugungi kunda juda muhim masala. Ushbu muammoni SI yordami orqali hal qilish mumkin. Zamonaviy ta’limda ta’lim metodologiyalarida, shu jumladan onlayn ta’limda sun’iy intellektdan foydalanish shubhasizdir. Ushbu vositalar tizimli o‘rganish imkonini beradi va yuqori sifatli, tez va samarali ta'lim tizimini osonlashtiradi. Bashoratli tahlillardan foydalanish: SI dasturlari talabalar natijalarini prognoz qilishi va qo‘shimcha yordamga muhtoj bo‘lgan xavf ostidagi o‘quvchilarni aniq belgilashi mumkin. Talabalarning o‘rganish ma’lumotlarini tahlil qilish orqali SI akademik qiyinchiliklarni aniqlay oladi va ularning muvaffaqiyatini osonlashtirish uchun zarur tadbirlarni amalga oshiradi. Ushbu tizimdagi yuqorida aytib o‘tilgan vazifalarga qo‘shimcha ravishda, biz ushbu tizimdagi ma’lumotlarni tartibga solish zaruratini ham hal qilishimiz mumkin. Ushbu sohada sun’iy intellektning beqiyos ahamiyati yuqorida aytib o‘tilgan fikrlar bilan namoyon bo‘ladi. Maktab tizimini takomillashtirish muhim yutuqlarimizdan biridir. Har bir yosh shaxs zamonaviy bilim olish bilan birga tez rivojlanayotgan davrga moslashadi. Ushbu sohani keng tarqalgan amaliyotga integratsiyalash yangi yutuqlar sari sezilarli muvaffaqiyat bo‘ladi. Harakatlar strategiyasi va undan keyingi rivojlanish strategiyasi orqali mamlakatimizda keyingi yillarda salmoqli yutuqlarga erishildi. Ta'lim sohasida sezilarli o‘zgarishlar yuz berdi. Sanoati rivojlangan davlatlar singari biz ham ta’lim tizimini izchil takomillashtirdik, o‘quv jarayoniga texnologiyani kiritdik, mukammal ta’limni ta’minladik va malakali kadrlar yetishtirdik. Bu ushbu sohani rivojlantirish uchun ko‘plab yaxshi tuzilgan tashabbuslar va o‘zgarishlarni o‘z ichiga oladi. Zamonaviy ta'lim tizimini yaratish va ushbu sohada texnologiya integratsiyasi. Ta’lim tizimini modernizatsiya qilishda sun’iy intellektning o‘rni katta ekanligini e’tirof etish zarur. Xulosa. Ta’lim jarayonlarini tahlil qilish uchun sun’iy intellekt modellarini yaratish zamonaviy ta’limda sezilarli taraqqiyotni anglatadi. Ushbu modellar pedagogik amaliyotlarni, talabalarning ishtirokini va institutsional qarorlarni qabul qilishni kuchaytiradigan ma’lumotlarga asoslangan tushunchalarni beradi. Axloq, tarafkashlik va ma’lumotlar xavfsizligi bilan bog‘liq qiyinchiliklarni hal qilish, mumkin bo‘lgan afzalliklarga qaramay, SIning ta’limga adolatli va muvaffaqiyatli kiritilishini kafolatlash uchun juda muhimdir. Keyingi tadqiqotlar SI algoritmlarini optimallashtirish, talqin qilish qobiliyatini yaxshilash va ta’lim sharoitida SIning afzalliklarini maksimal darajada oshirish uchun SI mutaxassislari va o‘qituvchilari o‘rtasidagi hamkorlikni rag‘batlantirishga qaratilishi kerak. Xulosa qilib aytganda, raqamli texnologiyalarni ta’lim sohasiga keng miqyosda integratsiya qilish va sektorni raqamli platformalarga o‘tkazish zarur. Har tomonlama barkamol shaxsni 269

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot shakllantirish, mustaqil kasb tanlash, kasbiy o‘zini-o‘zi shakllantirish, malaka oshirishda axborot texnologiyalarining o‘rni va ta’siri shubhasiz kengaymoqda. Ayni paytda mamlakatimizda ta’lim tizimiga katta e’tibor qaratilmoqda. Bu yoshlarni ta’lim tizimiga jalb qilish va ularning vaqtini samarali tashkil etish bo‘yicha say-harakatlarni o‘z ichiga oladi. Ayni paytda ta’lim sohasida ko‘plab ma’naviy-ma’rifiy tashabbuslar amalga oshirilmoqda. Ayni vaqtda yoshlar bilan davra suhbatlari o‘tkazilmoqda. Bugungi kunda yoshlarga ta’lim-tarbiya berish jarayonlari nafaqat oilalar, maktablar, balki keng miqyosda ham kundalik tashvishga aylangan. Biz hozirda ilm-fan taraqqiyoti bilan ajralib turadigan jamiyatda yashayapmiz. Yoshlarimiz hamon zamonaviy o‘zgarishlarga moslashgan. Ko‘plab yoshlar mazmunli o‘yinkulgi uchun turli xil bo‘sh tarmoqlardan foydalanmoqda. Shunday qilib, bu zamonaviy bolalarning ta’limiga salbiy ta’sir qiladi. Biz yoshlarni ta’limga yo‘naltirishimiz kerak. Biz ushbu sohani rivojlantirishimiz va ta’lim tizimini zamonaviy talablarga moslashtirgan holda sun’iy intellektni ta’limga keng ko‘lamli integratsiya qilishimiz kerak. Adabiyotlar 1. Sabirjonov R.A., Ayupov R.H. va boshqalar. Zamonaviy raqamli texnologiyalarning rivojlanish istiqbollari. “Media” nashriyoti, 2022 yil. 202 bet. 2. Sharof Rashidov nomidagi Samarqand davlat universiteti I.A. FILIPOVA “Sun’iy intellektni huquqiy tartibga solish” (ma’ruza kursi) Samarqand-2022 b-7 3. Ya.X.G‘afforov. Tarix o‘qitish metodikasi. Darslik. -Toshkent:-211b 4. Russell SJ, Norvig P. Sun’iy intellekt: zamonaviy yondashuv. Prentice Hall. Englewood Cliffs, Nyu-Jersi. 1995 yil. 3-bet 5. G‘afforov Ya “Mahsus fanlarni o‘qitishda yangi pedagogic tehnalogiyanardan foydalanish usullari” Toshkent ‘Shafoat Nur Fayz’ nashriyoti, 2021-yil 4-bet 6. N.A. Shermuxamedova, Ilmiy-tadqiqot metodologiyasi. T. “Fan va texnologiya” nash, 2014, 403-407 betlar 7. Ayupov R.H., Tursunov S.Q. Raqamli texnologiyalar. o‘quv qo‘llanma. Toshkent: “Lesson Press” nashriyoti, 2023 yil. 464 bet. 270

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 INFOKOMMUNIKATSIYA TIZIMLARIDA ENERGIYA SAMARADORLIGINI OSHIRISHNING IOTGA ASOSLANGAN TEXNOLOGIK YONDASHUVLARINI TADQIQ QILISH. Shodiyev Shuxrat Hayit o‘g‘li1, Abdumalikov Akmaljon Abduxoliq o‘g‘li2 1Kommunikatsiya va raqamli texnologiyalar kafedrasi, “University of management and future technologies” universiteti, Toshkent, 100208, O‘zbekiston; sh.shodiyev222@gmail.com 2o‘ZMU Jizzax filiali Ilmiy tadqiqotlar, inovatsiyalar va ilmiy pedagog kadrlar tayyorlash bo‘limi boshlig‘i Toshkent, 100208, O‘zbekiston; a.abdumalikov@gmail.com Annotatsiya: Ushbu maqolada infokommunikatsiya tizimlarida energiya samaradorligini oshirish uchun IoT (Internet of Things – Narsalar Interneti) ga asoslangan texnologik yondashuvlar tadqiq qilinadi. Zamonaviy telekommunikatsiya infratuzilmalari va axborot tarmoqlarida energiya sarfini optimallashtirishning dolzarb muammolari ko‘rib chiqiladi. Shuningdek, ushbu yondashuvning afzalliklari, cheklovlari va kelajakdagi rivojlanish istiqbollari muhokama qilinadi. Natijalar shuni ko‘rsatadiki, IoT texnologiyalari infokommunikatsiya tizimlarida barqarorlik va samaradorlikni oshirish uchun muhim vosita bo‘lib xizmat qiladi. Kalit so‘zlar: Infokommunikatsiya tizimlari, energiya samaradorligi, energiya tejash, avtomatlashtirish, barqarorlik. Kirish Zamonaviy dunyoda infokommunikatsiya tizimlari iqtisodiyot, ta’lim, sog‘liqni saqlash va kundalik hayotning ajralmas qismiga aylandi. Ushbu tizimlarning uzluksiz ishlashi va kengayishi energiya resurslariga bo‘lgan talabning oshishiga olib keldi, bu esa ekologik muammolar va operatsion xarajatlarni oshirish kabi dolzarb masalalarni keltirib chiqarmoqda. Ushbu maqolada infokommunikatsiya tizimlarida energiya samaradorligini oshirish uchun IoT ga asoslangan texnologik yondashuvlarning imkoniyatlari, amaliy qo‘llanilishi va kelajakdagi rivojlanish yo‘nalishlari atroflicha tadqiq qilinadi. Tadqiqotning maqsadi – ushbu texnologiyalarning samaradorlikni oshirishdagi o‘rni va ularni joriy etishning potensial afzalliklarini aniqlashdan iborat. Tadqiqotning dolzarbligi Bugungi kunda infokommunikatsiya tizimlari global raqamli transformatsiyaning asosiy poydevori sifatida tobora muhim ahamiyat kasb etmoqda. Shu bilan birga, energiya resurslarining cheklanganligi va iqtisodiy samaradorlikka bo‘lgan talab bu sohada innovatsion yechimlarni izlashni zarur qilmoqda. IoT texnologiyalari real vaqtda energiya sarfini monitoring qilish, tarmoq resurslarini aqlli boshqarish va samarasiz jarayonlarni minimallashtirish imkonini berib, ushbu muammolarni hal qilishda muhim vosita sifatida ko‘rilmoqda. Tadqiqotning dolzarbligi infokommunikatsiya tizimlarida energiya samaradorligini oshirish zarurati, IoT asosidagi yondashuvlarning amaliy ahamiyati va ularning kelajakdagi infratuzilmalarni barqaror rivojlantirishga qo‘shadigan hissasi bilan 271

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot bog‘liq. Ushbu ish nafaqat texnologik jihatdan, balki ekologik va iqtisodiy nuqtai nazardan ham dolzarb bo‘lib, sohada samarali yechimlar ishlab chiqishga xizmat qiladi. Tadqiqot metodologiyasi Ushbu tadqiqot infokommunikatsiya tizimlarida energiya samaradorligini oshirish uchun IoT ga asoslangan texnologik yondashuvlarni o‘rganishda aralash metodologiyadan foydalanadi, bu esa miqdoriy va sifatli tahlillarni birlashtirish imkonini beradi. Tadqiqot jarayoni bir necha bosqichlardan iborat bo‘lib, har bir bosqich muayyan maqsadlarni amalga oshirishga xizmat qiladi. Birinchi bosqichda, mavjud infokommunikatsiya tizimlarining energiya sarfi bo‘yicha ma’lumotlar yig‘iladi. Ushbu tajribada har xil yuk sharoitlarida (masalan, yuqori trafik va past trafik davrlari) energiya sarfini aniqlash uchun simulyatsiya modellaridan foydalaniladi. Tahlil va natijalar muhokamasi Tadqiqot jarayonida infokommunikatsiya tizimlarida IoT ga asoslangan texnologik yondashuvlarning energiya samaradorligiga ta’siri tahlil qilindi va bir qator muhim natijalar qayd etildi. Tajriba davomida IoT sensorlari yordamida real vaqtda yig‘ilgan ma’lumotlar shuni ko‘rsatdiki, energiya sarfini monitoring qilish va aqlli boshqaruv tizimlari yordamida umumiy iste’mol o‘rtacha 18-25% ga qisqardi. Xususan, yuqori trafik davrlarida tarmoq resurslarini dinamik taqsimlash algoritmlari qo‘llanilganda, energiya sarfi 30% gacha kamaydi, bu esa avvalgi tadqiqotlarda Smit (2021) tomonidan keltirilgan natijalar bilan mos keladi. Simulyatsiya modellaridan olingan ma’lumotlar shuni tasdiqladiki, IoT asosidagi avtomatlashtirish tizimlari samarasiz quvvat sarfini aniqlash va uni optimallashtirishda muhim rol o‘ynaydi. Energiya samaradorligini hisoblash uchun quyidagi formula ishlatildi: 𝑬𝑬 = (𝑬𝑬_𝒊𝒊𝒊𝒊 − 𝑬𝑬_𝒐𝒐𝒐𝒐𝒐𝒐) / 𝑬𝑬_𝒊𝒊𝒊𝒊 × 𝟏𝟏𝟏𝟏𝟏𝟏% (1) Bu yerda E – energiya samaradorligi foizda, E_in – tizimga kiruvchi umumiy energiya (kVt/soat), E_out – foydalanilmagan yoki isrof qilingan energiya (kVt/soat). IoT joriy etilgandan so‘ng samaradorlik o‘zgarishi quyidagicha hisoblanadi: 𝜟𝜟𝜟𝜟 = 𝑬𝑬_𝑰𝑰𝑰𝑰𝑰𝑰 − 𝑬𝑬_𝒃𝒃𝒃𝒃𝒃𝒃𝒃𝒃 (2) ΔE – samaradorlikdagi farq, E_IoT – IoT texnologiyasi qo‘llanilgandagi samaradorlik, E_base – IoT siz bazaviy samaradorlik. IoT qo‘llanilishi natijasida energiya sarfi o‘zgarishi (1-jadval). 1-jadval. Algoritmlarning aniqlik ko‘rsatkichlari Tarmoq turi IoT siz energiya IoT bilan energiya Samaradorlik o‘sishi sarfi (kVt/soat) sarfi (kVt/soat) (%) Ma’lumotlar markazi 500 425 15% 5G tarmoq 300 210 30% Kichik telekom tizimi 100 90 10% 272

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 Jadvaldan ko‘rinib turibdiki, 5G tarmoqlarda IoT texnologiyasi eng yuqori samaradorlikni (30%) ko‘rsatdi, kichik tizimlarda esa o‘sish minimal (10%) bo‘ldi. Xulosa va takliflar Ushbu tadqiqot infokommunikatsiya tizimlarida energiya samaradorligini oshirishda IoT ga asoslangan texnologik yondashuvlarning samaradorligini o‘rganishga bag‘ishlandi. Tahlil natijalari shuni ko‘rsatdiki, IoT texnologiyalari real vaqtda monitoring, aqlli resurslarni boshqarish va avtomatlashtirilgan tahlil vositalari orqali energiya sarfini sezilarli darajada kamaytirish imkonini beradi. Shu bilan birga, tadqiqot IoT qurilmalarining o‘ziga xos energiya talabi va infratuzilmaviy cheklovlar kabi muayyan kamchiliklarni ham aniqladi, bu esa ushbu texnologiyalarni joriy etishda ehtiyotkorlik bilan rejalashtirish zarurligini ko‘rsatadi. Xulosa sifatida, IoT ga asoslangan yondashuvlar infokommunikatsiya tizimlarining barqarorligi va iqtisodiy samaradorligini oshirishda muhim vosita sifatida o‘zini namoyon qildi. Biroq, ularning to‘liq salohiyatidan foydalanish uchun qo‘shimcha takomillashtirish talab etiladi. Taklif sifatida, IoT texnologiyalarini infokommunikatsiya sohasida yanada samarali qo‘llashga va global barqarorlik maqsadlariga erishishga xizmat qiladi. Tadqiqot natijalari va takliflar kelgusida ushbu sohada davom etadigan ishlar uchun mustahkam asos yaratadi. Adabiyotlar 1. Ahmedov, A. (2023) Cybersecurity Challenges in Digital Era, Central Asian Journal of Technology, vol. 7, no. 2, pp. 34-50, O‘zbekiston, Tashkent Tech Press. 2. Chen, L. va Zhang, Y. (2024) Real-Time Cyber Threat Detection with AI, International Journal of Cybersecurity, vol. 12, pp. 78-92, Xitoy, Science Press. 3. Garcia, M. (2022) Artificial Intelligence in Network Security, Journal of Future Technologies, vol. 5, no. 1, pp. 15-28, Ispaniya, Tech Innovate Publishing. 4. Ivanov, P. va Sokolov, D. (2023) Machine Learning for Phishing Detection, Cybersecurity Advances, vol. 9, pp. 55-70, Rossiya, Moscow Cyber Press. 5. Kim, H., Park, J. va Lee, S. (2024) Deep Learning for Threat Detection, IEEE Transactions on Cybersecurity, vol. 10, pp. 100-115, O‘zbekiston, IEEE Press. 6. Lee, S. (2022) Limitations of Rule-Based Systems, Cybersecurity Review, vol. 5, no. 3, pp. 12-25, Rossiya, CyberTech Publishing. 7. Patel, R. (2023) AI-Based Solutions for Cybercrime Prevention, Journal of Digital Security, vol. 6, pp. 40-55, Hindiston, Digital Press. 8. Smith, J. (2023) Machine Learning in Cybersecurity, Journal of Cyber Defense, vol. 8, pp. 45-60, AQSh, Defense Press. 9. Sobirov, B. (2024) Kiberxavfsizlikda sun‘iy intellekt, Ilm-Fan Jurnali, vol. 3, no. 1, pp. 20-35, O‘zbekiston, Fan va Texnologiya Nashriyoti. 273

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot AN ARTICLE FOCUSES ON REAL-TIME FARM SURVEILLANCE USING IOT AND AI BASED ON YOLOV MODELS, ENHANCING PREVIOUS VERSIONS FOR ANIMAL INTRUSION DETECTION Sadikov Mahmudjon Akmuratovich, Toshtemirov Temirbek Qodirjon o‘g‘li Muhammad al-Xorazmiy nomidagi Toshkent Axborot Texnologiyalari Universiteti, imaxmaxmud@gmail.com, temirbektoshtemirov@gmail.com Abstract: The article presents a novel method for preventing animal invasion in agricultural fields, utilizing advanced sensor technologies, image processing, and IoT integration, thereby reducing agricultural losses and safeguarding crops. The YoloV7 model, with exceptional accuracy and precision, has been used in an IoT device to provide real-time intrusion alarms on farms. The ESP32cam module was used to integrate this technology, transforming farm monitoring and providing farmers with timely knowledge to prevent threats and protect agricultural production. Keywords: Agricultural security, animal detection, crop protection, early warning system, IoT, intrusion detection, image processing, machine learning, sensor technology, smart farming, wildlife intrusion, Yolo V7. Introduction Farming, frequently referred to as the "lifeblood of sustenance," is essential to replenishing our food supply. Just producing these resources is insufficient, though, as destroying them before they can be used loses precious raw materials in addition to time and money. Protection of farmlands and agricultural assets is therefore essential, particularly in light of the growing problem of animal encroachment. This danger, which is posed by bears, macaques, deer, porcupines, wild boars, and monkeys, has not only been demonstrated to be quite dangerous but has occasionally even killed people. Due to strict wildlife regulations, local farmers are disproportionately affected by this issue, suffering losses of up to 50% of their crops. As a result, they are reluctant to take drastic actions [1]. The foundation of both providing essential raw materials to industry and supplying the population's dietary needs is agriculture. In contrast, agricultural industries suffer significant crop losses as a result of animal intervention. Recently, one of the main causes of concern has been the destruction of crops by roving wild animals. Among the most notable predations are those of wild boars, macaques, porcupines, deer, monkeys, and bears, which have occasionally led to human death. Basic commodities farmed in rural areas, such potatoes and wheat, are particularly affected negatively in terms of agricultural productivity; small-scale farmers bear the brunt of these losses, which can reach 40% to 50% of total yield. By skillfully utilizing state-of-the-art object detection algorithms, the Smart Farm Surveillance System aims to provide proactive, real-time monitoring of agricultural and critical regions. This new strategy promises quick detection and mitigation of potential threats by seamlessly fusing cloud computing with AI-powered camera systems. The necessity for robust security measures and the burgeoning agriculture sector 274

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 underscores the rising demand for contemporary monitoring systems. With traditional monitoring systems' inability to identify and react to hazards in real time, it is imperative that advanced technologies like the Smart Farm Surveillance System be adopted. By guaranteeing the early identification of intruders, animal presence, or unauthorized activity, this system successfully tackles the challenge of timely threat detection on farms. The Smart Farm Surveillance System is built on state-of-the-art object identification algorithms that employ deep learning and machine learning methods to identify a variety of objects, including specific animal species and human incursions. It offers proactive, realtime surveillance to prevent potential threats, setting off alarms for questionable activity like intruders or wildlife encroachment. Integration of cloud technologies allows for secure data storage for analysis and remote access to surveillance feeds. It provides data-driven insights for crop management, is flexible enough to accommodate different farm sizes, and promotes humane security measures. The system addresses the problem of wildlife interference in farming while preserving sustainability and profitability, making it globally relevant despite the initial investment due to its cost-effectiveness and adherence to wildlife regulations. There are numerous agricultural applications and advantages to the Smart Farm Surveillance System. Due to its ability to identify and notify farmers of the presence of animals including wild boars, macaques, porcupines, deer, monkeys, and bears, it is a powerful tool for wildlife deterrence. This enables farmers to take immediate action to prevent crop damage. Additionally, the system is excellent at detecting intruders, guarding against theft and unauthorized access. By providing information on growth, health, and environmental aspects, it supports agricultural monitoring and helps with decisions about pest control and irrigation. Gathering environmental data promotes precision farming, which maximizes the use of available resources. Certain versions are also capable of handling animals, offering security and care for them. A Raspberry Pi CPU and a variety of sensors are included in this study's innovative agricultural security system. By combining the powers of ultrasonic and passive infrared (PIR) sensors, it improves agricultural security. The ultrasonic sensor is essential for identifying the size and identity of the intruder, while the PIR sensors are programmed to activate whenever they sense any motion within their range. The system can identify intruders based on their physical characteristics and sound alerts in response thanks to the wide coverage provided by four PIR sensors positioned at various heights. A camera is triggered when an intrusion is detected, capturing a picture of the intruder and initiating an alert distribution system via the GSM module and email. To further deter and repel the invader, an aural alert or buzzer is turned on right away. This study offers a workable, affordable solution to the urgent security issues that are pervasive in agriculture, shielding farmers from the escalating risk of wildlife incursions. By providing an automated system that significantly reduces the need for human interaction, it gives farmers real-time notifications. Crucially, this method effectively classifies attackers according to their size and level of threat, enabling defenses to be customized for each distinct group. This paper outlines a novel animal identification and warning system that uses computer vision and artificial intelligence (AI) to safeguard agricultural grounds. To guarantee precise animal identification, cameras are positioned strategically to monitor possible entry points for wild 275

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot animals. The captured visual data is then carefully examined using OpenCV in combination with a retrained MobileNet SSD model. An auditory alarm is triggered as soon as an animal is spotted, discouraging the intruding species and alerting the farming community at the same time. In addition, the system maintains a thorough log of every known creature. The pervasive problem of agricultural output depletion in Indian agriculture due to uncontrolled wildlife disturbances is addressed in this study. It offers a surveillance system driven by AI that can identify and track animal behavior, enabling farmers to improve crop protection strategies and lower losses. The method significantly lessens the requirement for human interaction by using an automated and economical approach. Most importantly, it performs exceptionally well at precisely identifying and categorizing animals in agricultural settings, setting off alarms, and alerting farmers. This resourceful mechanism acts as a good bulwark, reducing crop susceptibility and animal invasion losses, as research has previously shown [2]. Table 1. Summary of Research Contribution. Authors Approach Contributions Dhake Sensor technology (PIR and ultrasound) Notifications in real time and for motion detection and intruder automated classification of classification. intruders. Vardhan Artificial intelligence and computer vision Building a system for the are used to identify animals. detection and alerting of animals in real time. Yadahalli system with many sensors ultrasound, and camera). Geetha PIR, LDR, and fire detection sensors are IoT-based hazard detection, examples of IoT technology. including forest fires. Andacarapu W-CoHOG, or computer vision. (PIR, Sort trespassers and provide real-time alerts. obtaining enhanced features for detecting animal behavior. Gogoi image processing (using SIFT to find Agriculture uses image objects). processing tools to identify animals. Lanza IoT deployments and architecture are Advice on the advantages, analyzed. difficulties, and future directions of IoT research in agriculture. Galyna developing artificial intelligence with the IoT is being used to support Internet of Things to enable smart smart agriculture by collecting agriculture. data in real time. Due to the wide coverage provided by four PIR sensors positioned at various heights, the system is able to classify intruders according to their physical characteristics and sound 276

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 alarms when necessary. When a camera detects an incursion, it takes a photo of the intruder and uses the GSM module and email to start an alert distribution system. Methodology and Implementation This innovative approach provides a significant improvement in addressing the issue of agricultural infiltration and safeguarding crops against potential animal damage or human theft. Compared to more conventional methods like creating wooden scarecrows to ward off animals, this is a far more successful solution. The proposed system leverages cloud computing for improved item detection and fast user warning. The suggested system's methodology is shown in Figure 1. First, microcontrollers are used to easily transmit the live video from the fields to the cloud servers. These cloud servers have two purposes: they contain machine learning algorithms that identify animals and potential intrusions, and they store real-time agricultural field imagery. Real-time intrusion detection is made possible by the deployment of machine learning algorithms on cloud servers. Figure 1. Suggested System Methodology 2.1. Intrusion Detection Both humans and animals are typically referred to as "intruders" in agricultural contexts [24]. There are numerous potential repercussions for these intrusions in this setting. Because they can eat or damage crops, animals—wild or domesticated—pose a serious threat to them. Conversely, human invaders, who are often driven by the unlawful practice of crop stealing, may have far from altruistic motives. This system's main objective is to offer a reliable and effective instrument for identifying invasions on agricultural land. The fundamental idea is to activate a prompt and responsive warning and alarm system upon detection of such incursions. This technique is essential for ensuring the prompt interception of unauthorized entities, including people and animals, averting potential harm and theft [3]. It is crucial to identify agricultural system invasions. By using advanced entry detection systems, it is now possible to lessen the risk of crop loss from animal exploitation and human aggression. 2.1.1 Collecting Data 277

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot The suggested system's data collection process is crucial since it forms the foundation for building and evaluating the deep learning models. To assure the system's dependability, a variety of data collection techniques were used. It is impossible to overstate the importance of utilizing diverse datasets, particularly when it comes to applications with a wide range of complexity. To accurately quantify the deep learning models' capabilities, it was essential to assess their performance in a variety of scenarios. Consequently, three distinct datasets were meticulously constructed. The 580 images in the first dataset are essential to the model training procedure. Images from two Kaggle datasets—one featuring people and the other featuring animals—are combined in this dataset. A complete dataset fit for testing, evaluation, and training is the end outcome of this consolidation. Each of the 30 distinct animal sets used in this fusion has 60 photographs. To provide a comprehensive training environment, 60 pictures of people and 60 pictures of open fields were also provided. 2.1.2. Selecting appropriate Model. Selecting a model for the intrusion detection part of the Farm Guard System is crucial [26]. Since the system's overall performance is determined by its capacity to consistently identify and classify intruders, whether they are people or animals, the model utilized for this task is essential to its efficacy. The system's primary objective of safeguarding agricultural land may be compromised by inaccurate or subpar models. The model selection process is shown in Fig. 2. Figure 2. Model Selections 2.1.3. Comparison of the models. The five main object identification models—Inception, Xception, AlexNet, VGG16, and YOLOv7—are thoroughly reviewed in table III. Important criteria such architectural designs, object detection speeds, accuracy, training dataset requirements, and hardware compatibility are used to evaluate each model. In our analysis, YOLOv7 emerges as a noteworthy choice. By combining object localization and classification into a single pass, YOLOv7 sets itself apart with its one-shot detection architecture. It is perfect for real-time applications because of this innovation and its quick detecting speed. Moreover, YOLOv7 is a great substitute for workloads requiring precision because it accomplishes this speed without 278

Sun’iy Intellekt va Raqamli Iqtisodiyot SIRI-2025 compromising accuracy. Despite its high detection accuracy, Inception, Xception, and VGG16 sometimes require large training datasets [3-6]. However, given its ability to maintain accuracy, YOLOv7's low dataset need is significant. Additionally, performance is enhanced by YOLOv7's support for GPU acceleration, particularly in real-time processing settings. Table 2. Comparison of Object detecting models Conclusion An inventive and effective approach to intrusion detection, the Farm Guard System offers farmers a reliable and secure solution. This intelligent surveillance system makes use of state-of-the-art technology and AWS features to enable farmers to proactively safeguard their crops and livestock. By providing speaker controls, quick alarms, and real-time object recognition, the technology ensures farm security. The versatility of the Smart Farm Surveillance System extends its usefulness beyond of the farming sector. From railway track monitoring to border security, wildlife protection, forest fire detection, disaster response, industrial safety, environmental monitoring, agricultural pest control, and oil and gas pipeline inspection, the integration of cloud-based object identification opens up a world of opportunities. By offering real-time monitoring and danger identification, this adaptable technology transforms surveillance techniques and improves security across numerous industries. The Farm Guard System serves as both a basis for upcoming developments and a remedy for current issues. The system's capacity to meet security requirements across a range of businesses will only get better as new options and technological advancements arise. In the surveillance and security sector, the Farm Guard System is a beacon of efficiency and innovation, continuously improving intrusion detection systems and expanding their scope of use. With its advanced features and adaptability, this smart monitoring system demonstrates the possibility of using technology to protect and preserve assets, whether they are vital infrastructure in various industries or crops on a farm. References 1. Anoop, N. R., Krishnan, S., & Ganesh, T. (2023). Elephants in the farm–changing temporal and seasonal patterns of human-elephant interactions in a forest-agriculture matrix in the Western Ghats, India. Frontiers in Conservation Science, 4, 1142325. 279

SIRI-2025 Sun’iy Intellekt va Raqamli Iqtisodiyot 2. Kommineni, M., Lavanya, M., & Vardhan, V. H. (2022). Agricultural farms utilizing computer vision (ai) and machine learning techniques for animal detection and alarm systems. Journal of Pharmaceutical Negative Results, 3292-3300. 3.Hong, S. J., Park, S., Lee, C. H., Kim, S., Roh, S. W., Nurhisna, N. I., & Kim, G. (2023). Application of X-ray imaging and convolutional neural networks in the prediction of tomato seed viability. IEEE Access. 4. A. V. Deshpande, “Design and implementation of an intelligent security system for farm protection from wild animals,” International Journal of Science and Research, ISSN (Online), pp. 2319–7064, 2016. 5. S. Pandey and S. B. Bajracharya, “Crop protection and its effectiveness against wildlife: A case study of two villages of Shivapuri National Park, Nepal,” Nepal Journal of Science and Technology, vol. 16, no. 1, pp. 1–10, 2015. 6.K. Rao, R. Maikhuri, S. Naut iyal, and K. G. Saxena, “ Crop damage and livestock depredation by wildlife: a case study fromNanda Devi biosphere reserve, India,” Journal of Environmental Management, vol. 66, no. 3, pp. 317–327, 2002. 280