Index

Introduction

Security Properties

For more see at Cryptography

OSI Security Architecture

5D of Perimeter Security

Security Mechanisms

X.800

Security Policy

Security Attack

Jamming

Sniffing

Spoofing

TCP/IP Architecture

Physical Layer Data Link Layer Network Layer Transport Layer Application Layer
Wired/Wireless MAC IPv4/IPv6 TCP/UDP/… HTTP/…

Physical Interception

Passive Interception: promiscuous mode (The network must broadcast all packages such as wireless) Active Interception: port mirroring, network tapping

Traffic Analysis

Data Acquisition

Data acquisition is the process of understanding the meaning of the captured traffic

BPF (Berkeley Packet Filter)

Problem:

BPF solution:

Traffic Analysis

Types of traffic analysis

Network Attacks

ARP Spoofing

ARP

Attack

Application

  1. MitM In cryptography and computer security, a man-in-the-middle attack is a cyberattack where the attacker secretly relays and possibly alters the communications between two parties who believe that they are directly communicating with each other.
  1. MAC Flooding
    • Switches have an internal CAM (Content Addressable Memory) table, which maps switch ports to MAC addresses.
    • When a frame arrives at switches, this lookup table is checked to know which port to forward each specific packet out of where the receiver is located
    • In a MAC flooding attack, a switch is fed many Ethernet frames, each containing different spoofed source MAC addresses to consume the limited memory in the switch

DHCP Starvation

DHCP Protocol

Attack

  1. If the legitimate DHCP server in the network starts responding to all bogus DHCP_REQUEST messages with random generated MAC address, available IP addresses in the DHCP server scope will be depleted within a very short span of time.

  2. Once the available number of IP addresses in the DHCP server is depleted, network attackers could set up a bogus DHCP server and respond to new DHCP DISCOVERY message from network DHCP clients

  3. The rogue server starts distributing IP addresses and other TCP/IP configuration settings including default gateway and DNS server IP addresses, which can now point to an IP address controlled by the attacker. Facilitates man-in-the-middle attack and sniffing attacks

ICMP Smurfing

Smurfing: amplification
Small amount of traffic are converted into large amounts of traffic, in order to attack the target

Spoofed ECHO_REQUEST

  1. Smurf malware is used to generate a fake Echo request, containing a spoofed source IP, which is actually the target server address
  2. The request is sent to an intermediate IP broadcast network
  3. The request is transmitted to all of the network hosts on the network
  4. Each host sends a ping response (ICMP ECHO_REPLY) to the spoofed source address
  5. With enough ICMP ECHO_REPLY forwarded, the target server is brought down which causes denial of service (DoS)

NTP (Network Time Protocol) Amplification DDoS Attack

  1. The attacker uses a botnet to prepare NTP request packets with spoofed addresses to NTP servers, which have their monlist command enabled. The spoofed IP address is the victim’s IP address
  2. The botnet sends the NTP request (a UDP packet) to NTP servers using its monlist command, resulting in a large response
  3. The server responds to the spoofed address with the resulting data, normally of much larger size than the original NTP request
  4. The victim receives the response and the surrounding network infrastructure becomes overwhelmed with the amplification of traffic, resulting in a DoS attack.

TCP SYN Flooding

SYN is the initial package that establishing the connection

For TCP see at Computer Network (Chinese Version)

  1. The attacker, who is a botmaster here, instructs the bots to send a high volume of SYN packets to the targeted server, often with spoofed IP addresses.

  2. The server then responds to each one of SYN packets and leaves an open port ready to receive the response

  3. While the server waits for the final ACK packet, which never arrives, the attacker continues to send more SYN packets. The arrival of each new SYN packet causes the server to temporarily maintain a new open port connection for a certain length of time, and once all the available resources have been utilised the server is unable to function normally.

TCP Session Hijacking

The attacker could also pretend to be the client to ask data from server.

Off-path Hijack

Guess the initial seq num:

What the attacker does

  1. Swamp the Client C with ICMP flooding, taking it out of the picture (because the attacker wants to spoof as C)
  2. Create a real connection to port 80 on the web Server, and record the sequence number returned by the web Server.
  3. Close the connection with the Server.
  4. Create a raw IP socket, change its protocol to that of TCP, and change its source IP to that of the Client (by writing in the kernel)
  5. Send a SYN packet (supposedly from the Client) to port 80 on the web server.
  6. The server then sends an SYN+ACK to the Client C, which is silently ignored because C is under ICMP flooding attack
  7. Send an ACK packet to the server with the acknowledgement number equal to the sequence number previously recorded plus N+1

After connection is established and hijacked

BGP Route Hijacking

Attack

  1. BGP prefix hijacking
    • When an AS announces route to network prefixes that it does not actually control
    • Such false information is added to routing tables in BGP routers across the Internet
  1. BGP sub-prefix hijacking
    • The attacker lies about a subset of the prefix rather than the whole prefix belonging to another AS
    • The false route is chosen because BGP prefers longest prefix matching

DNS Cache Poisoning

  1. Query QID Attack
    • Every DNS query has a QID
    • DNS use connectionless UDP
    • If the attacker responds to query with the right QID, then it wins, over the real nameserver’s response!
  2. RRSet attack
    • DNS response contains different Resource Record Sets, or RRSets.
    • In particular, an “additional” section, where name server can give additional info that may be “useful” for future lookups
    • In an iterative query, the .com nameserver says you can ask ns.example.com for the IP address of example.com. To help next request, an additional record might give IP for ns.example.com.
    • Darth abuses this feature and adds a record that says the IP address of victim.com is 88.88.88.88
    • FIX: Bailiwick checking (when asking for www.google.com, do not allow www.victim.com to be accepted as an additional record)

Firewall

Objectives

Main task: Access Control

Techniques for firewalls to control access

Firewall Types

Limitations

Canot protect

Packet Filtering Firewall

Filter Rules

Default = discard

Default = forward

Pros and Cons

Pros

Cons

attacks and countermeasures

IP address spoofing

Source routing attack

Tiny fragment attacks

Intrusion Detection and Prevention Systems

Terms

Detection Rate

See at Evaluation

$Sensitivity = True Positive Rate = Detection Rate = \frac{TP}{TP + TN}$

$Specificity = True Negative Rate = \frac{TN}{TN + FP}$

Intrusion Detection

IDS classification

How an IDS works

Honeypots

Classification

Honeypots Workflow

Phase: data control

Honeypots control and contain the activities of an attacker by interacting with the attacker through network protocols

Phase: data capture

-Honeypots monitor and log all of the activities of an attacker within the honeypot

Phase: data analysis

Honeytoken

IPsec

Security Association

Authentication Header (AH)

Header:

Two Modes:

Encapsulating Security Payload (ESP)

Header:

Transport Mode:

Tunnel Mode:

Internet Key Exchange (IKE)

DH Key Exchange

Perfect Forward Secrecy (PFS)

A property of key-agreement protocols ensuring that a session key derived from a set of long term keys cannot be compromised if one of the long-term keys is compromised in the future. The key used to protect transmission of data must not be used to derive any additional keys, and if the key used to protect transmission of data is derived from some other keying material, then that material must not be used to derive any more keys. In this way, compromise of a single key permits access only to data protected by that single key.

The trick to achieving Perfect Forward Secrecy is to generate a temporary session key, not derivable from the information stored at the node and forgotten after the session concludes.

SSL/TLS

For vulnerability see ssl vulnerabilities

User Interface Vulnerability

Privacy and Anonymity

Anonymity

Attacks

Requirement: anonymise the sender and/or the receiver

Mix Networks

Single Mix

For sending message M to agent at address B

$\{r_1, \{r_0, M\}_{pk_b}, B\}_{pk_{mix}} \rightarrow \{r_0, M\}_{pk_b}, B$

Foiling Traffic Analysis Requirements

  1. Agents/mixes work with uniformly sized items i.e., messages split (or padded) into fixed size blocks
  2. Order of arrival hidden by outputting items in batches
    • Can use fix ordering (e.g., lexicographic) or random ordering
  3. Repeated information must be blocked
    • So mixes must filter duplicates, cross-checking across batches
    • Or string r include, e.g., a time stamp
  4. Sufficient traffic from a large anonymity set is required
    • Few clients sending entails weak anonymity
    • Solution involves clients regularly sending (and receiving dummy messages).

Untraceable Return Addresses

Generalized Format

  1. Sender includes “return address”:

    $\{r_1, \{r_2, …, \{r_n, A_x\}_{pk_{mix_n}}\}_{pk_{mix_2}}\}_{pk_{mix_1}}, pk_x$

  2. Receiver sends to the “response” Mix:

    $\{r_1, \{r_2, …, \{r_n, A_x\}_{pk_{mix_n}}\}_{pk_{mix_2}}\}_{pk_{mix_1}}, \{r_0, M’\}_{pk_x}$

  3. Result of 1^{st} Return Mix:

    $\{r_2, …, \{r_n, A_x\}_{pk_{mix_n}}\}_{pk_{mix_2}}, \{\{r_0, M’\}_{pk_x}\}_{r_1}$

  4. Final result:

    $A_x, \{\{\{\{r_0, M’\}_{pk_x}\}_{r_1}\}_{r_2} …\}_{r_n}$

Attacks

Dining cryptographers (DC)

  1. Each of the three cryptographers $C_1, C_2$ and $C_3$ flips an unbiased coin keeping the result $b_i$ secret ($i \in \{1,2,3\}$).
  2. Each cryptographer whispers the result $b_i$ in the ear of the person to their immediate left.
  3. Each cryptographer computes $d_i = b_i \oplus b_{i-1}$, where § where $b_i$ is the cryptographer’s own coin flip and § $b_{i-1}$ is the coin flip of the person on the right § Note that $d_0 = d_3$ and $b_0 = b_3$
  4. A cryptographer that did not pay for the meal announces her own $d_i$ A cryptographer that did pay for the meal lies by announcing the negation of $d_i$ i.e., $d_i \oplus 1$.

Wireless and IoT security

Wireless security risk factors

802.11i Operation

IoT Security

IoT Architecture

IoT vulnerabilities by layers