{"id":11200,"date":"2024-10-30T19:48:19","date_gmt":"2024-10-30T16:48:19","guid":{"rendered":"https:\/\/vasexperts.com\/?p=11200"},"modified":"2025-08-11T14:13:31","modified_gmt":"2025-08-11T11:13:31","slug":"from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic","status":"publish","type":"post","link":"https:\/\/vasexperts.com\/br\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/","title":{"rendered":"Dos m\u00e9todos cl\u00e1ssicos \u00e0s redes neurais: Explorando o potencial da aprendizagem profunda na identifica\u00e7\u00e3o de tr\u00e1fego ofuscado"},"content":{"rendered":"<h2>M\u00e9todos tradicionais de an\u00e1lise de tr\u00e1fego de rede<\/h2>\r\nA classifica\u00e7\u00e3o do tr\u00e1fego de rede usando m\u00e9todos tradicionais envolve v\u00e1rias abordagens, cada uma com seus pr\u00f3prios pontos fortes e fracos. Vamos examinar os principais m\u00e9todos e suas limita\u00e7\u00f5es ao lidar com tr\u00e1fego ofuscado e criptografado.\r\n<h3>M\u00e9todo de indica\u00e7\u00e3o de nome de servidor (SNI)<\/h3>\r\nO m\u00e9todo SNI baseia-se na an\u00e1lise das informa\u00e7\u00f5es de dom\u00ednio que uma conex\u00e3o criptografada transmite em texto simples ao estabelecer uma sess\u00e3o TLS. Como o nome de dom\u00ednio \u00e9 especificado no cabe\u00e7alho \u201cServer Name\u201d (Nome do servidor) durante o handshake TLS, esse m\u00e9todo permite a identifica\u00e7\u00e3o de servidores e servi\u00e7os, mesmo que o tr\u00e1fego subsequente seja criptografado.\r\n\r\n      <strong>  Limita\u00e7\u00f5es do m\u00e9todo SNI:    <\/strong>  \r\n<ul>\r\n \t<li>   <strong>  Precis\u00e3o insuficiente com ofusca\u00e7\u00e3o de porta e tradu\u00e7\u00e3o de endere\u00e7o:   <\/strong>   Quando os endere\u00e7os IP e as portas s\u00e3o modificados ou ofuscados, a precis\u00e3o diminui porque o v\u00ednculo entre o SNI e um aplicativo espec\u00edfico pode ser interrompido.<\/li>\r\n \t<li>   <strong>  Incapacidade de identifica\u00e7\u00e3o ao usar VPNs:   <\/strong>   O cabe\u00e7alho SNI fica indispon\u00edvel para an\u00e1lise se o tr\u00e1fego passar por uma VPN, pois fica oculto pela criptografia do t\u00fanel.<\/li>\r\n \t<li>   <strong>  Falta de dados para todos os protocolos:   <\/strong>   Nem todos os protocolos e aplicativos transmitem dados por TLS, o que torna a an\u00e1lise baseada em SNI inaplic\u00e1vel a eles.<\/li>\r\n<\/ul>\r\n<h3>Inspe\u00e7\u00e3o de carga \u00fatil<\/h3>\r\nA inspe\u00e7\u00e3o de carga \u00fatil envolve uma an\u00e1lise detalhada do conte\u00fado do pacote para identificar padr\u00f5es e caracter\u00edsticas espec\u00edficas de um protocolo ou aplicativo. Esse m\u00e9todo oferece alta precis\u00e3o na determina\u00e7\u00e3o dos tipos de dados e na classifica\u00e7\u00e3o deles com base no conte\u00fado.\r\n\r\n      <strong>  Limitations of Payload Inspection:   <\/strong>  \r\n<ul>\r\n \t<li>   <strong>  Custos de recursos computacionais:   <\/strong>   A inspe\u00e7\u00e3o de carga \u00fatil requer recursos significativos devido \u00e0 necessidade de examinar o conte\u00fado de cada pacote.<\/li>\r\n \t<li>   <strong>  Problemas de privacidade:   <\/strong>   O acesso total aos dados do pacote gera preocupa\u00e7\u00f5es com a privacidade, especialmente quando se trabalha com dados pessoais ou corporativos.<\/li>\r\n \t<li>   <strong>  Incapacidade de analisar o tr\u00e1fego criptografado:   <\/strong>   A criptografia do tr\u00e1fego (TLS ou VPN) impossibilita a inspe\u00e7\u00e3o de carga \u00fatil, reduzindo a efic\u00e1cia desse m\u00e9todo em ambientes modernos em que uma parte significativa do tr\u00e1fego \u00e9 criptografada.<\/li>\r\n<\/ul>\r\n<h3>M\u00e9todos estat\u00edsticos de aprendizado de m\u00e1quina<\/h3>\r\nOs m\u00e9todos estat\u00edsticos de aprendizado de m\u00e1quina classificam o tr\u00e1fego com base em v\u00e1rias m\u00e9tricas e caracter\u00edsticas (como tamanhos de pacotes, frequ\u00eancia e intervalos de tempo). Os modelos podem ser treinados em dados estat\u00edsticos, permitindo a identifica\u00e7\u00e3o eficaz de determinados tipos de tr\u00e1fego em alguns casos.\r\n\r\n      <strong>  Limita\u00e7\u00f5es dos m\u00e9todos estat\u00edsticos de aprendizado de m\u00e1quina:   <\/strong>  \r\n<ul>\r\n \t<li>   <strong>  Necessidade de dados limpos e rotulados:   <\/strong>   Para uma opera\u00e7\u00e3o bem-sucedida, os modelos de aprendizado estat\u00edstico exigem dados rotulados de alta qualidade, o que \u00e9 um desafio para a coleta, especialmente para protocolos menos comuns.<\/li>\r\n \t<li>   <strong>  Uso intensivo de recursos:   <\/strong>   Esse m\u00e9todo requer recursos computacionais significativos, tornando a an\u00e1lise mais lenta em casos de grandes volumes de dados.<\/li>\r\n \t<li>   <strong>  Baixa efic\u00e1cia na presen\u00e7a de ofusca\u00e7\u00e3o de tr\u00e1fego:   <\/strong>   Protocolos que mascaram seus metadados ou mudam continuamente os padr\u00f5es de tr\u00e1fego podem complicar a an\u00e1lise, levando a uma baixa precis\u00e3o dos modelos estat\u00edsticos.<\/li>\r\n<\/ul>\r\nComo resultado, embora os m\u00e9todos tradicionais possam apresentar alta precis\u00e3o em alguns casos, eles enfrentam v\u00e1rias limita\u00e7\u00f5es, o que torna dif\u00edcil classificar os tipos de tr\u00e1fego modernos.\r\n\r\n<hr>\r\n\r\n<h2>Abordagem de rede neural para identificar o tr\u00e1fego de rede ofuscado<\/h2>\r\nNossa pesquisa explora a aprendizagem profunda como uma alternativa mais precisa e flex\u00edvel aos m\u00e9todos tradicionais. Implementamos modelos baseados em redes neurais convolucionais (CNN) e na arquitetura ResNet, adaptando-os para classifica\u00e7\u00e3o de alta precis\u00e3o de VPN criptografada e tr\u00e1fego de proxy.\r\n<h3>Dados<\/h3>\r\nUm conjunto de dados Netflow 10 (IPFIX) foi usado para classifica\u00e7\u00e3o de tr\u00e1fego, projetado para padronizar a transmiss\u00e3o de informa\u00e7\u00f5es de IP do exportador para o coletor, com o suporte de fabricantes como Cisco, Solera, VMware e Citrix. As especifica\u00e7\u00f5es do IPFIX s\u00e3o fornecidas nas RFCs 7011-7015 e RFC 5103.\r\n<h3>Coleta de dados<\/h3>\r\nOs dados foram coletados usando um dispositivo com um sistema de inspe\u00e7\u00e3o profunda de pacotes (DPI) conectado a outros dispositivos que geram tr\u00e1fego em v\u00e1rias VPNs. Essa abordagem capturou IPs e portas exclusivos gerados por VPNs com atribui\u00e7\u00f5es din\u00e2micas sob restri\u00e7\u00f5es, resultando em uma rica variedade de combina\u00e7\u00f5es exclusivas de IP e porta para treinar o modelo de rede neural.\r\n\r\n[product id=\u201d 19\u2033 type=\u201ddark\u201d]\r\n\r\nOs dados coletados inclu\u00edam os seguintes par\u00e2metros:\r\n<table width=\"555\" height=\"740\">\r\n<tbody>\r\n<tr>\r\n<th><b>Tipo de dados<\/b><\/th>\r\n<th><b>Descri\u00e7\u00e3o<\/b><\/th>\r\n<\/tr>\r\n<tr>\r\n<td>octet_delta_count<\/td>\r\n<td>Contador de entrada de comprimento N x 8 bits para o n\u00famero de bytes associados ao fluxo IP.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>contagem_delta_de_pacotes<\/td>\r\n<td>Contador de pacotes de entrada de comprimento N x 8 bits para o n\u00famero de pacotes associados ao fluxo de IP.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>identificador_de_protocolo<\/td>\r\n<td>Byte de protocolo IP.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>ip_class_of_service<\/td>\r\n<td>Classe de IP ou servi\u00e7o.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>porta_de_origem<\/td>\r\n<td>Porta do remetente.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>source_ipv4<\/td>\r\n<td>IPv4 do remetente.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>porta_destino<\/td>\r\n<td>Porta do destinat\u00e1rio.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>destination_ipv4<\/td>\r\n<td>IPv4 do destinat\u00e1rio.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>bgp_source_as_number<\/td>\r\n<td>N\u00famero do sistema aut\u00f4nomo BGP de origem (N pode ser 2 ou 4).<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>bgp_destination_as_number<\/td>\r\n<td>N\u00famero do sistema aut\u00f4nomo BGP de destino (N pode ser 2 ou 4).<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>input_snmp<\/td>\r\n<td>Identificador de LAN virtual associado \u00e0 interface de entrada.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>output_snmp<\/td>\r\n<td>Identificador de LAN virtual associado \u00e0 interface de sa\u00edda.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>ip_version<\/td>\r\n<td>Vers\u00e3o do protocolo IPv4 ou IPv6.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>post_nat_source_ipv4<\/td>\r\n<td>Fonte NAT IPv4.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>post_nat_source_port<\/td>\r\n<td>Porta NAT de origem.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>frgmt_delta_packs<\/td>\r\n<td>Delta de pacotes fragmentados.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>repeat_delta_pack<\/td>\r\n<td>Delta de retransmiss\u00f5es.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>tempo_de_entrega_do_pacote<\/td>\r\n<td>Atraso (RTT\/2), ms.<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>c\u00f3digo_do_protocolo<\/td>\r\n<td>C\u00f3digo do protocolo usando a classe de sistema aut\u00f4nomo para a rede neural.<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<h3>Processamento de dados antes do treinamento<\/h3>\r\n<span style=\"font-weight: 400;\">Os dados foram divididos em conjuntos de treinamento (80%) e teste (20%). Ajustes de equil\u00edbrio de classe e rotulagem de dados IPFIX foram aplicados para destacar classes espec\u00edficas.\r\n<h3>Treinamento<\/h3>\r\nAs redes neurais foram treinadas usando duas arquiteturas com ajuste de hiperpar\u00e2metro. As propor\u00e7\u00f5es de classe de protocolo na amostra de treinamento foram:\r\n<table>\r\n<tbody>\r\n<tr>\r\n<td><b>Protocolo<\/b><\/td>\r\n<td><b>Raz\u00e3o<\/b><\/td>\r\n<\/tr>\r\n<tr>\r\n<td>DNS<\/td>\r\n<td>18,67%<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>HTTP<\/td>\r\n<td>1,38%<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>HTTPS<\/td>\r\n<td>16,27%<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>DoH<\/td>\r\n<td>2,66%<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>ICMP<\/td>\r\n<td>4,83%<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>Bittorrent<\/td>\r\n<td>24,73%<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>AdGuard VPN<\/td>\r\n<td>2,34%<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>VPN Unlimited<\/td>\r\n<td>12,18%<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>Psiphon 3<\/td>\r\n<td>12,41%<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>Lanterna<\/td>\r\n<td>4,53%<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<h3>Teste<\/h3>\r\nOs modelos foram avaliados no conjunto de teste usando as m\u00e9tricas de precis\u00e3o, recupera\u00e7\u00e3o e pontua\u00e7\u00e3o F1:\r\n\r\n<span class=\"katex-display\"><span class=\"katex\"><span class=\"katex-mathml\"><\/span><\/span><\/span>\r\n\r\n<math xmlns=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" display=\"block\"><sem><mrow><mtext>Recall<\/mtext><mo>=<\/mo><mfrac><mtext>TP<\/mtext><mrow><mtext>TP<\/mtext><mo>+<\/mo><mtext>FN<\/mtext><\/mrow><\/mfrac><\/mrow><annotation encoding=\"application\/x-tex\">\\text{Recall} = \\frac{\\text{TP}}}{\\text{TP} \\text{FN}}<\/annotation><\/sem><\/math>\r\n\r\n<span class=\"katex-html\" aria-hidden=\"true\"><span class=\"base\"><span class=\"strut\"><\/span><span class=\"mord text\"><span class=\"mord\">Recall<\/span><\/span><span class=\"mspace\"><\/span><span class=\"mrel\">=<\/span><span class=\"mspace\"><\/span><\/span><span class=\"base\"><span class=\"strut\"><\/span><span class=\"mord\"><span class=\"mopen nulldelimiter\"><\/span><span class=\"mfrac\"><span class=\"vlist-t vlist-t2\"><span class=\"vlist-r\"><span class=\"vlist\"><span><span class=\"pstrut\"><\/span><span class=\"mord text\">TP<\/span><span class=\"mspace\"><\/span><span class=\"mbin\">+<\/span><span class=\"mspace\"><\/span><span class=\"mord text\">FN<\/span><\/span><span><span><span class=\"pstrut\"><\/span><span class=\"frac-line\"><\/span><\/span><span><span class=\"pstrut\"><\/span><span class=\"mord text\">TP<\/span><\/span><\/span><\/span><span class=\"vlist-s\"><\/span><\/span><span class=\"vlist-r\"><span class=\"vlist\"><span><\/span><\/span><\/span><\/span><\/span><span class=\"mclose nulldelimiter\"><\/span><\/span><\/span><\/span><\/span> <span class=\"katex-display\"><span class=\"katex\"><span class=\"katex-mathml\"><\/span><\/span><\/span>\r\n\r\n<math xmlns=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" display=\"block\"><semantics><mrow><mtext>Precision<\/mtext><mo>=<\/mo><mfrac><mtext>TP<\/mtext><mrow><mtext>TP<\/mtext><mo>+<\/mo><mtext>FP<\/mtext><\/mrow><\/mfrac><\/mrow><annotation encoding=\"application\/x-tex\">\\text{Precision} = \\frac{\\text{TP}}{\\text{TP} \\text{FP}}<\/annotation><\/semantics><\/math>\r\n\r\n<span class=\"katex-html\" aria-hidden=\"true\"><span class=\"base\"><span class=\"strut\"><\/span><span class=\"mord text\"><span class=\"mord\">Precis\u00e3o<\/span><\/span><span class=\"mspace\"><\/span><span class=\"mrel\">=<\/span><span class=\"mspace\"><\/span><\/span><span class=\"base\"><span class=\"strut\"><\/span><span class=\"mord\"><span class=\"mopen nulldelimiter\"><\/span><span class=\"mfrac\"><span class=\"vlist-t vlist-t2\"><span class=\"vlist-r\"><span class=\"vlist\"><span><span class=\"pstrut\"><\/span><span class=\"mord text\">TP<\/span><span class=\"mspace\"><\/span><span class=\"mbin\">+<\/span><span class=\"mspace\"><\/span><span class=\"mord text\">FP<\/span><\/span><span><span class=\"pstrut\"><\/span><span class=\"frac-line\"><\/span><\/span><span><span class=\"pstrut\"><\/span><span class=\"mord text\">TP<\/span><\/span><\/span><\/span><span class=\"vlist-s\"><\/span><\/span><span class=\"vlist-r\"><span class=\"vlist\"><span><\/span><\/span><\/span><\/span><\/span><span class=\"mclose nulldelimiter\"><\/span><\/span><\/span> <span class=\"katex-display\"><span class=\"katex\"><span class=\"katex-mathml\"><\/span><\/span><\/span>\r\n\r\n<math xmlns=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" display=\"block\"><sem><mrow><mtext>F1 Pontua\u00e7\u00e3o<\/mtext><mo>=<\/mo><mfrac><mrow><mn>2<\/mn><mo>\u00d7<\/mo><mtext>Recall<\/mtext><mo>\u00d7<\/mo><mtext>Precision<\/mtext><\/mrow><mrow><mrow><mtext>Recall<\/mtext><mo>+<\/mo><mtext>Precision<\/mtext><\/mrow><\/mrow><\/mfrac><\/mrow><annotation encoding=\"application\/x-tex\">\\text{F1 Score} = \\frac{2 \\times \\text{Recall} \\times \\text{Precision}}{\\text{Recall} + \\text{Precision}}<\/annotation><\/sem><\/math>\r\n\r\n<span class=\"katex-html\" aria-hidden=\"true\"><span class=\"base\"><span class=\"strut\"><\/span><span class=\"mord text\"><span class=\"mord\">F1 Score<\/span><\/span><span class=\"mspace\"><\/span><span class=\"mrel\">=<\/span><span class=\"mspace\"><\/span><\/span><span class=\"base\"><span class=\"strut\"><\/span><span class=\"mord\"><span class=\"mopen nulldelimiter\"><\/span><span class=\"mfrac\"><span class=\"vlist-t vlist-t2\"><span class=\"vlist-r\"><span class=\"vlist\"><span><span class=\"pstrut\"><\/span><span class=\"mord text\">Recall<\/span><span class=\"mspace\"><\/span><span class=\"mbin\">+<\/span><span class=\"mspace\"><\/span><span class=\"mord text\">Precis\u00e3o<\/span><\/span><span><span class=\"pstrut\"><\/span><span class=\"frac-line\"><\/span><\/span><span><span class=\"pstrut\"><\/span>2<span class=\"mspace\"><\/span><span class=\"mbin\">\u00d7<\/span><span class=\"mspace\"><\/span><span class=\"mord text\">Recall<\/span><span class=\"mspace\"><\/span><span class=\"mbin\">\u00d7<\/span><span class=\"mspace\"><\/span><span class=\"mord text\">Precision<\/span><\/span><\/span><\/span><span class=\"vlist-s\"><\/span><\/span><\/span><span class=\"vlist-r\"><span class=\"vlist\"><span><\/span><\/span><\/span><\/span><span class=\"mclose nulldelimiter\"><\/span><\/span><\/span>\r\n\r\nem que TP denota verdadeiros positivos, FN falsos negativos e FP falsos positivos.\r\n\r\nO experimento foi realizado em VPNs com um amplo intervalo de IPs para aumentar a objetividade dos resultados. O modelo de arquitetura ResNet demonstrou maior precis\u00e3o na classifica\u00e7\u00e3o dos protocolos de VPN.\r\n<h3>Resultados<\/h3>\r\n<h4>Rede Neural Convolucional Cl\u00e1ssica<\/h4>\r\n<table>\r\n<tbody>\r\n<tr>\r\n<td>Protocolo<\/td>\r\n<td>TP<\/td>\r\n<td>FP<\/td>\r\n<td>FN<\/td>\r\n<td>Score F1<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>AdGuard VPN<\/td>\r\n<td>28<\/td>\r\n<td>9<\/td>\r\n<td>50<\/td>\r\n<td>0,49<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>VPN Ilimitada<\/td>\r\n<td>3<\/td>\r\n<td>3<\/td>\r\n<td>22<\/td>\r\n<td>0,21<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>Psif\u00e3o 3<\/td>\r\n<td>8455<\/td>\r\n<td>160<\/td>\r\n<td>399<\/td>\r\n<td>0,97<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<h4>Arquitetura da ResNet<\/h4>\r\n<table>\r\n<tbody>\r\n<tr>\r\n<td>Protocolo<\/td>\r\n<td>TP<\/td>\r\n<td>FP<\/td>\r\n<td>FN<\/td>\r\n<td>Score F1<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>AdGuard VPN<\/td>\r\n<td>60<\/td>\r\n<td>5<\/td>\r\n<td>18<\/td>\r\n<td>0,84<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>VPN Ilimitada<\/td>\r\n<td>5<\/td>\r\n<td>9<\/td>\r\n<td>20<\/td>\r\n<td>0,26<\/td>\r\n<\/tr>\r\n<tr>\r\n<td>Psif\u00e3o 3<\/td>\r\n<td>8847<\/td>\r\n<td>1030<\/td>\r\n<td>7<\/td>\r\n<td>0,95<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\nA arquitetura ResNet mostrou maior efici\u00eancia na identifica\u00e7\u00e3o do tr\u00e1fego VPN e pode servir como uma base confi\u00e1vel para tarefas de classifica\u00e7\u00e3o de tr\u00e1fego criptografado.\r\n<h2>Conclus\u00e3o<\/h2>\r\nNeste artigo, examinamos m\u00e9todos de identifica\u00e7\u00e3o de tr\u00e1fego ofuscado, abrangendo abordagens cl\u00e1ssicas e de redes neurais. Embora os m\u00e9todos tradicionais ofere\u00e7am recursos b\u00e1sicos, eles t\u00eam limita\u00e7\u00f5es em ambientes din\u00e2micos de tr\u00e1fego e criptografia. As redes neurais modernas oferecem maior precis\u00e3o e flexibilidade, identificando com efic\u00e1cia o tr\u00e1fego ofuscado mesmo quando os m\u00e9todos tradicionais se mostram ineficazes. Assim, a mudan\u00e7a para abordagens de redes neurais marca um avan\u00e7o significativo na seguran\u00e7a da rede.","protected":false},"excerpt":{"rendered":"<p>A an\u00e1lise e a classifica\u00e7\u00e3o do tr\u00e1fego de rede tornaram-se essenciais para manter a resili\u00eancia e a seguran\u00e7a das redes de computadores contempor\u00e2neas. Com o r\u00e1pido aumento do volume de dados e a crescente complexidade dos m\u00e9todos de criptografia, a necessidade de uma classifica\u00e7\u00e3o eficaz do fluxo de rede continua a crescer. Ao identificar, categorizar e analisar o tr\u00e1fego de rede com precis\u00e3o, as organiza\u00e7\u00f5es podem detectar poss\u00edveis amea\u00e7as, otimizar o desempenho da rede e garantir a conformidade com os protocolos de seguran\u00e7a.<\/p>\n","protected":false},"author":24,"featured_media":11203,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[48],"tags":[],"class_list":["post-11200","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-telecom"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v23.1 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>(English) VASExperts<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/\",\"url\":\"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/\",\"name\":\"[:br]Dos m\u00e9todos cl\u00e1ssicos \u00e0s redes neurais: Explorando o potencial da aprendizagem profunda na identifica\u00e7\u00e3o de tr\u00e1fego ofuscado[:fr]Des m\u00e9thodes classiques aux r\u00e9seaux neuronaux : Exploration du potentiel de l'apprentissage profond dans l'identification du trafic obscurci[:es]De los m\u00e9todos cl\u00e1sicos a las redes neuronales: Explorando el potencial del Deep Learning en la identificaci\u00f3n de tr\u00e1fico ofuscado[:en]From classical methods to neural networks: Exploring the potential of Deep Learning in identifying obfuscated traffic[:] - (English) VASExperts\",\"isPartOf\":{\"@id\":\"https:\/\/vasexperts.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#primaryimage\"},\"thumbnailUrl\":\"\/wp-content\/uploads\/2024\/10\/frame-1321314805-1.jpg\",\"datePublished\":\"2024-10-30T16:48:19+00:00\",\"dateModified\":\"2025-08-11T11:13:31+00:00\",\"author\":{\"@id\":\"https:\/\/vasexperts.com\/#\/schema\/person\/f27674e15c2c6c09c36db4e68f9dd198\"},\"description\":\"[:en]Explore the transition from classical methods to deep learning in identifying obfuscated traffic. Learn how neural networks can enhance network performance and security in telecommunications.[:es]Explora la transici\u00f3n de los m\u00e9todos cl\u00e1sicos al aprendizaje profundo en la identificaci\u00f3n del tr\u00e1fico ofuscado. Descubre c\u00f3mo las redes neuronales pueden mejorar el rendimiento y la seguridad de las telecomunicaciones.[:fr]Explorez la transition des m\u00e9thodes classiques vers l\u2019apprentissage profond pour identifier le trafic obscurci. D\u00e9couvrez comment les r\u00e9seaux neuronaux peuvent am\u00e9liorer les performances et la s\u00e9curit\u00e9 des t\u00e9l\u00e9communications.[:br]Explore a transi\u00e7\u00e3o dos m\u00e9todos cl\u00e1ssicos para o deep learning na identifica\u00e7\u00e3o de tr\u00e1fego ofuscado. Saiba como as redes neurais podem melhorar o desempenho e a seguran\u00e7a nas telecomunica\u00e7\u00f5es.\",\"breadcrumb\":{\"@id\":\"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#breadcrumb\"},\"inLanguage\":\"br-PT\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"br-PT\",\"@id\":\"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#primaryimage\",\"url\":\"\/wp-content\/uploads\/2024\/10\/frame-1321314805-1.jpg\",\"contentUrl\":\"\/wp-content\/uploads\/2024\/10\/frame-1321314805-1.jpg\",\"width\":1000,\"height\":460},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u0413\u043b\u0430\u0432\u043d\u0430\u044f \u0441\u0442\u0440\u0430\u043d\u0438\u0446\u0430\",\"item\":\"https:\/\/vasexperts.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"From classical methods to neural networks: Exploring the potential of Deep Learning in identifying obfuscated traffic\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/vasexperts.com\/#website\",\"url\":\"https:\/\/vasexperts.com\/\",\"name\":\"ITGLOBAL.COM\",\"description\":\"(English) VASExperts\",\"inLanguage\":\"br-PT\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/vasexperts.com\/#\/schema\/person\/f27674e15c2c6c09c36db4e68f9dd198\",\"name\":\"Darya Aulova\",\"url\":\"https:\/\/vasexperts.com\/br\/blog\/author\/darya-aulova\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"(English) VASExperts","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/","url":"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/","name":"[:br]Dos m\u00e9todos cl\u00e1ssicos \u00e0s redes neurais: Explorando o potencial da aprendizagem profunda na identifica\u00e7\u00e3o de tr\u00e1fego ofuscado[:fr]Des m\u00e9thodes classiques aux r\u00e9seaux neuronaux : Exploration du potentiel de l'apprentissage profond dans l'identification du trafic obscurci[:es]De los m\u00e9todos cl\u00e1sicos a las redes neuronales: Explorando el potencial del Deep Learning en la identificaci\u00f3n de tr\u00e1fico ofuscado[:en]From classical methods to neural networks: Exploring the potential of Deep Learning in identifying obfuscated traffic[:] - (English) VASExperts","isPartOf":{"@id":"https:\/\/vasexperts.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#primaryimage"},"image":{"@id":"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#primaryimage"},"thumbnailUrl":"\/wp-content\/uploads\/2024\/10\/frame-1321314805-1.jpg","datePublished":"2024-10-30T16:48:19+00:00","dateModified":"2025-08-11T11:13:31+00:00","author":{"@id":"https:\/\/vasexperts.com\/#\/schema\/person\/f27674e15c2c6c09c36db4e68f9dd198"},"description":"[:en]Explore the transition from classical methods to deep learning in identifying obfuscated traffic. Learn how neural networks can enhance network performance and security in telecommunications.[:es]Explora la transici\u00f3n de los m\u00e9todos cl\u00e1sicos al aprendizaje profundo en la identificaci\u00f3n del tr\u00e1fico ofuscado. Descubre c\u00f3mo las redes neuronales pueden mejorar el rendimiento y la seguridad de las telecomunicaciones.[:fr]Explorez la transition des m\u00e9thodes classiques vers l\u2019apprentissage profond pour identifier le trafic obscurci. D\u00e9couvrez comment les r\u00e9seaux neuronaux peuvent am\u00e9liorer les performances et la s\u00e9curit\u00e9 des t\u00e9l\u00e9communications.[:br]Explore a transi\u00e7\u00e3o dos m\u00e9todos cl\u00e1ssicos para o deep learning na identifica\u00e7\u00e3o de tr\u00e1fego ofuscado. Saiba como as redes neurais podem melhorar o desempenho e a seguran\u00e7a nas telecomunica\u00e7\u00f5es.","breadcrumb":{"@id":"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#breadcrumb"},"inLanguage":"br-PT","potentialAction":[{"@type":"ReadAction","target":["https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/"]}]},{"@type":"ImageObject","inLanguage":"br-PT","@id":"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#primaryimage","url":"\/wp-content\/uploads\/2024\/10\/frame-1321314805-1.jpg","contentUrl":"\/wp-content\/uploads\/2024\/10\/frame-1321314805-1.jpg","width":1000,"height":460},{"@type":"BreadcrumbList","@id":"https:\/\/vasexperts.com\/blog\/telecom\/from-classical-methods-to-neural-networks-exploring-the-potential-of-deep-learning-in-identifying-obfuscated-traffic\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u0413\u043b\u0430\u0432\u043d\u0430\u044f \u0441\u0442\u0440\u0430\u043d\u0438\u0446\u0430","item":"https:\/\/vasexperts.com\/"},{"@type":"ListItem","position":2,"name":"From classical methods to neural networks: Exploring the potential of Deep Learning in identifying obfuscated traffic"}]},{"@type":"WebSite","@id":"https:\/\/vasexperts.com\/#website","url":"https:\/\/vasexperts.com\/","name":"ITGLOBAL.COM","description":"(English) VASExperts","inLanguage":"br-PT"},{"@type":"Person","@id":"https:\/\/vasexperts.com\/#\/schema\/person\/f27674e15c2c6c09c36db4e68f9dd198","name":"Darya Aulova","url":"https:\/\/vasexperts.com\/br\/blog\/author\/darya-aulova\/"}]}},"_links":{"self":[{"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/posts\/11200"}],"collection":[{"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/users\/24"}],"replies":[{"embeddable":true,"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/comments?post=11200"}],"version-history":[{"count":11,"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/posts\/11200\/revisions"}],"predecessor-version":[{"id":12763,"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/posts\/11200\/revisions\/12763"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/media\/11203"}],"wp:attachment":[{"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/media?parent=11200"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/categories?post=11200"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/vasexperts.com\/br\/wp-json\/wp\/v2\/tags?post=11200"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}