π Lecture 38 β Security/Extensions of WSN Part V
π Overview: This lecture covers security primitives in TinySec, encryption schemes, and keying mechanisms for wireless sensor networks. It then introduces Wireless Multimedia Sensor Networks (WMSN) and Wireless Sensor and Actor Networks (WSAN), discussing their architectures, applications, advantages, and design challenges.
ποΈ Topics Covered
Security primitives in TinySec including message authentication codes and initialization vectors, TinySec design options for authentication encryption and authentication only, IV format design, symmetric key encryption schemes including stream ciphers and CBC mode, keying mechanisms from per-link to group and single network keys, then Wireless Multimedia Sensor Networks architecture, advantages, design considerations, and deployed examples, followed by Wireless Sensor and Actor Networks motivations, comparison with WSN, architecture types, and key issues.
π Lecture Summary
Security Primitives
Message Authentication code (MAC) is a cryptographic secure checksum for checking message integrity. Computing a MAC requires authorized senders and receivers to share a secret key, and this key is part of the input to a MAC computation. If an adversary alters a valid message or injects a bogus message, she cannot compute the corresponding MAC value.
π Definition β MAC: a cryptographic secure checksum for checking the message integrity, computed using a shared secret key between authorized sender and receiver.
Initialization vector (IV) is used to achieve semantic security β encrypting the same plaintext two times should give two different ciphertexts. A common technique for achieving semantic security is to use a unique initialization vector (IV) for each invocation of algorithm. The IV is a side input to the encryption algorithm.
π Definition β IV: a unique side input to the encryption algorithm that ensures the same plaintext produces different ciphertexts each time it is encrypted.
TINYSEC-DESIGN
TinySec offers 2 Security Options:
- Authentication Encryption (Tinysec-AE): TinySec encrypts the data payload and authenticates the packet with a MAC. The MAC is computed over the encrypted data and the packet header.
- Authentication only (Tinysec-Au): TinySec authenticates the entire packet with a MAC, but the data payload is not encrypted.
Encryption requires semantically secure design, which typically requires two design decisions: specifying the IV format and selecting an encryption scheme.
Tinysec IV format
If the IV is too long, it adds unnecessary bits to the packet. If too short, there is a risk of repetition. An n-bit IV will repeat after 2^n + 1 uses. If we use an n-bit counter, repetitions will not happen before that point. Pseudorandom IV would repeat with probability of 2^(n/2).
Encryption Schemes
Symmetric key encryption schemes fall into two classes:
Stream ciphers: A stream cipher typically uses a key K and IV as a seed and stretches it into a large pseudorandom keystream GK(IV). The keystream is then XORed against the message. Stream ciphers have a devastating failure mode: if the same IV is ever used to encrypt two different packets, then it is often possible to recover both plaintexts.
Modes of operation using block ciphers: A block cipher is a keyed pseudorandom permutation over small bit strings, typically 8 or 16 bytes. CBC (Cipher Block Chaining) is the most appropriate scheme for sensor networks because it works better with repeated IVs.
CBC
In CBC mode, the IV is XORed with the first data block before it is encrypted. The result of encryption is fed back into the encryption of the next block. The plaintext is XORed with the previous ciphertext block before it is encrypted. The encryption of each block depends on all the previous blocks. This requires that the decryption side processes all encrypted blocks sequentially.
An error in an encrypted block causes the block with the error to be completely garbled. The subsequent block will have bit errors at the same positions as the original erroneous block. The blocks following the second block will not be affected by the error. Hence, CBC is self-recovering.
Keying Mechanism
Several keying options exist:
- Per-link keying: Use a separate TinySec key for each pair of nodes wishing to communicate. Drawback: Key distribution becomes a challenge.
- Group keying: Allow a group of nodes to share a TinySec key rather than each pair. Group keying provides an intermediate level of resilience.
- Single network key: The simplest mechanism uses a single key for the entire network, preloaded before deployment. The adversary can compromise one node and get the key.
The appropriate keying mechanism for a particular network depends on several factors. A TinySec key consists of a pair of Skipjack keys β one for authentication, one for encryption.
Wireless Multimedia Sensor Networks
WMSN are networks of wirelessly interconnected devices that allow retrieving video and audio streams, still images, and scalar sensor data. They are able to store, process in real-time, correlate, and fuse multimedia data originated from heterogeneous sources.
Reference Architecture of WMSN
[Reference architecture diagram shown in lecture includes multimedia sensor nodes with cameras and microphones, interconnected through a wireless network, with a sink/base station connected to a central processing/storage system, and potentially connected to the internet or other networks.]
Advantages
- Enlarging the Views: Provide multiple disparate viewpoints to overcome occlusion effects.
- Enhancing the Views: Redundancy provides enhanced quality.
- Enabling Multi-resolution Views: Heterogeneous media streams with different granularity can be acquired from the same point of view.
Design Considerations
- Application-specific QoS requirements for snapshot and streaming multimedia.
- Flexible architecture to support heterogeneous applications.
- Multimedia source coding including intra-frame/inter-frame and distributed source coding.
- Multimedia in-network processing.
- Multimedia coverage model development.
- Power consumption.
Examples of Deployed WMSN
SensEye performs three tasks: object detection, recognition, and tracking. Its objective is to demonstrate that a camera sensor network containing heterogeneous elements provides numerous benefits over traditional homogeneous sensor networks.
New Applications include: Storage and Retrieval of Interesting Activities (e.g., IrisNet, 2004), Traffic congestion avoidance and enforcement systems, Smart parking advice system (2005), Automated Assistance for the elderly and family monitors (2005), Manufacturing process control for semiconductor chip, food or pharmaceutical products.
Application Layer
The services offered by the application layer include:
- Providing traffic management and admission control functionalities.
- Performing source coding according to application requirements and hardware constraints, using advanced multimedia encoding techniques.
- Developing flexible OS and Middleware to make functional abstractions and information gathered by the scalar and multimedia sensors available to higher layer applications.
Traffic Management and Admission Control
Tasks include: Prevent applications from establishing data flows when the network resources needed are not available; provide traffic classes for differentiated service between real-time and delay-tolerant applications, and loss-tolerant and loss-intolerant applications.
Related work: An application admission control algorithm (2003) whose objective is to maximize network lifetime subject to bandwidth and reliability constraints. Another method (2003) determines admissions based on added energy load and application rewards.
Transport Layer
Regarding TCP or UDP for WMSN: For real-time applications like streaming media, UDP seems preferred over TCP due to effect of dropping packets in UDP and support for traffic heterogeneity. However, TCP with appropriate modifications is preferable over UDP for WMSNs, if standardized protocols are to be used.
Non-Standard Protocols
Focusing on reliability:
- Reliable Multi-Segment Transport (RMST) (2004) or Pump Slowly Fetch Quickly (PSFQ) protocol (2005): Loss-intolerant packets are separated and ensured to be successfully transmitted. Loss-intolerant packets are buffered at intermediate nodes, allowing for faster retransmission in case of packet loss. Other packets are transmitted in UDP manner. No congestion avoidance.
- Event-to-Sink Reliable Transport (ESRT) protocol (2005): Not best-effort but reliable requirement-based rate control with congestion detection and avoidance.
Using Multiple Paths
- Regulating streaming through multiple TCP connections (2005): Sender sends desired streaming rate and allows throughput reduction to the receiver. Receiver measures actual throughput, controls the rate within allowed bounds by using multiple TCP connections and dynamically changing its TCP window size for each connection.
- Splitting a large burst of data into several smaller bursts using Multi-flow Real-time Transport Protocol (MRTP) (2006).
- Allows the sink to regulate multiple sources associated with a single event using Congestion Detection and Avoidance (CODA) protocol (2003).
WSAN: I. Motivations
Environmental Applications: Detecting and extinguishing forest fire. Distributed Robotics & Sensor Networks: Mobile robots dispersed throughout the field in sensor networks, e.g., mines detection and destruction. Structure health monitoring and control: Sensors to observe seismic excitation in bridges/buildings; actors to reduce deflections. Surveillance/Emergency handling: Immediate alerts of changes in patient status; relay data to hospital, correlate with patient records. Battlefield applications: Sensors detect explosive materials or weapons; actors annihilate them or function as tank. Microclimate controls in smart buildings: In case of very low or high temperature/gas leakages, trigger the alarms or corresponding controller.
WSAN: II. Wireless Sensor Actor Networks
Sensors: Passive nodes sensing from the environment with limited energy, processing, and communication capabilities. Actors: Active nodes acting on the environment with higher processing and communication capabilities, less-constrained energy resources, and mobile capability. WSAN = WSN + Actors.
WSANs vs. Wireless Sensor Networks
- Real-time requirements for timely actions: Rapidly respond to sensor input (e.g., in battlefield). To perform right action, sensor data must be valid at the time of action.
- Heterogeneous Nodes: Sensors (densely deployed) with heterogeneity for multiple events detection or multi-level energy sources; Actors (loosely deployed) with different action capabilities.
- Distributed local coordination requirements: Sensor-Actor coordination and Actor-Actor coordination.
- Nodes mobility: Specially actor nodes such as robots, ambulance, tank.
WSAN Architecture
Semi-automated architecture: Sensors communicate with a central controller which then coordinates actors. Automated architecture: Sensors communicate directly with actors, which coordinate among themselves without central control.
WSAN: III. Issues
- Self-configuration of sensor nodes.
- Energy conservation is the primary concern as in WSNs.
- Localization of sensor nodes relative to actors.
- Real-time routing β Is it possible with highly dynamic topology?
- Aggregation β It might affect in-time data delivery.
- Coordination.
- Redundancy β Exploit spatial or temporal correlation.
π‘ Why this matters: WMSN and WSAN represent advanced extensions of WSN that support multimedia and actuation capabilities, enabling real-world applications like surveillance, disaster response, and smart environments. Understanding their security, architecture, and protocol challenges is essential for designing robust systems.
β Key Takeaways
TinySec provides two security options β authentication encryption (AE) which encrypts payload and authenticates with MAC, and authentication only (AU) which only authenticates without encryption. The IV format design is critical: too long wastes bits, too short risks repetition. CBC is the preferred encryption mode for sensor networks because it self-recovers from errors and works better with repeated IVs compared to stream ciphers which have devastating failures if IV repeats. WMSN enables video/audio streaming with advantages of enlarged, enhanced, and multi-resolution views but faces challenges in QoS, source coding, and power consumption. WSAN extends WSN with actors that can act on the environment, introducing real-time requirements, heterogeneous nodes, distributed coordination, and mobility issues.
π§ Quick Revision Questions
- What are the two security options in TinySec, and what does each encrypt/authenticate?
- Why is CBC mode preferred over stream ciphers for sensor network encryption?
- What are the three keying mechanisms discussed for TinySec, and what is the drawback of each?
- What three advantages does WMSN offer over traditional WSN?
- How does WSAN differ from WSN in terms of node types, real-time requirements, and coordination?
π Lecture 39 β Bluetooth/Wireless Personal Area Networks (WPAN)
π Overview: This lecture introduces Bluetooth technology as a low-cost, low-power, short-range wireless solution for ad-hoc communication of voice and data. It covers Bluetoothβs technical features, architecture, core protocols, packet formats, and link connection types, explaining how devices form piconets and scatternets for wireless personal area networking.
ποΈ Topics Covered
This lecture covers Bluetooth introduction and definition, application areas, technical features including frequency hopping and time-division duplex, radio specifications and transmitter classes, Bluetooth topology including piconets and scatternets, standards documents and profiles, protocol architecture with core protocols (Radio, Baseband, LMP, L2CAP, SDP), packet format and access codes, inquiry and page procedures, link manager protocol functions, connection states, L2CAP services, and the two types of links between master and slave (SCO and ACL).
π Lecture Summary
What is Bluetooth?
Bluetooth wireless technology is an open specification for a low-cost, low-power, short-range radio technology designed for ad-hoc wireless communication of voice and data anywhere in the world. It serves as the ultimate headset and cordless computer connection.
Bluetooth Application Areas
Bluetooth has three main application areas: Data and voice access points for real-time voice and data transmissions, Cable replacement which eliminates the need for numerous cable attachments for connection, and Ad hoc networking where a device with Bluetooth radio can establish connection with another when in range.
Time-Division Duplex Scheme
The channel is divided into consecutive slots, each 625 ΞΌs long. One packet can be transmitted per slot. Subsequent slots are alternatively used for transmitting and receiving with strict alternation of slots between the master and the slaves. The master can send packets to a slave only in EVEN slots, and the slave can send packets to the master only in ODD slots.
Overview of Bluetooth History
Bluetooth is a short-range wireless communications technology. The name was taken from the 10th century Danish King Harald Blatand who unified Denmark and Norway. In 1994, Ericsson began a study on a wireless technology to link mobile phones and accessories. Five companies joined to form the Bluetooth Special Interest Group (SIG) in 1998, and the first specification was released in July 1999.
Technical features
| Feature | Specification |
|---|---|
| Connection Type | Spread Spectrum (Frequency Hopping) & Time Division Duplex (1600 hops/sec) |
| Spectrum | 2.4 GHz ISM Open Band (79 MHz of spectrum = 79 channels) |
| Modulation | Gaussian Frequency Shift Keying |
| Transmission Power | 1 mw β 100 mw |
| Data Rate | 1 Mbps |
| Range | 30 ft |
| Supported Stations | 8 devices |
| Data Security β Authentication Key | 128 bit key |
| Data Security β Encryption Key | 8-128 bits (configurable) |
| Module size | 9 x 9 mm |
Radio Specification
There are three classes of transmitters: Class 1 outputs 100 mW for maximum range with mandatory power control, providing the greatest distance. Class 2 outputs 2.4 mW at maximum with optional power control. Class 3 has a nominal output of 1 mW, the lowest power.
Typical Bluetooth Scenario
Bluetooth supports wireless point-to-point and point-to-multipoint (broadcast) between devices in a piconet. In a Point to Point Link, there is a master-slave relationship where Bluetooth devices can function as masters or slaves. Each piconet has a maximum capacity of 1 Mbps. A Piconet is the network formed by a master and one or more slaves (maximum 7). Each piconet is defined by a different hopping channel to which users synchronize. All devices in the piconet hop together. The masterβs ID and masterβs clock determine the frequency hopping sequence and phase. The hopping sequence is shared with all devices on the piconet. Bluetooth devices use Time Division Duplex (TDD), and the access technique is TDMA β specifically FH-TDD-TDMA.
Ad-hoc Network β the Scatternet
A Scatternet enables inter-piconet communication. Up to 10 piconets can exist in a scatternet, and multiple piconets can operate within the same physical space. This is an ad-hoc, peer-to-peer (P2P) network.
Bluetooth Standards Documents
The Core specifications provide details of various layers of Bluetooth protocol architecture, including core protocols, cable replacement and telephony control protocols, and adopted protocols. Profile specifications define the use of Bluetooth technology to support various applications.
Profiles
Bluetooth profiles include: Generic Access Profile, Service Discovery Application Profile, Cordless Telephony Profile, Intercom Profile, Serial Port Profile, Headset Profile, Dial-up Networking Profile, Fax Profile, LAN Access Profile, Generic Object Exchange Profile, Object Push Profile, File Transfer Profile, and Synchronization Profile.
Architecture
The Bluetooth architecture consists of Core protocols (Radio, Baseband, Link Manager Protocol LMP, Logical Link Control and Adaptation Protocol L2CAP, Service Discovery Protocol SDP), Cable replacement protocol (RFCOMM), Telephony control protocol (Telephony Control Specification β binary TCS BIN), and Adopted protocols (PPP, TCP/UDP/IP, OBEX, WAE/WAP).
Core Protocols
Radio defines the technical characteristics of BT radios, including the licence-free ISM band 2.4 GHz, FHSS at 1600 Hops/sec, 1 MHz channel bandwidth, GMSK modulation, tx power from 100 mw to 1 mw, and raw transmission rate of 1 Mbps. Baseband defines the procedure to communicate with other BT devices, including formation of piconets, links in a piconet (ACL or SCO), and access of transmit resources. Link Manager Protocol (LMP) is a transactional protocol between two link management entities used to setup properties of BT links, such as device authentication and learning each other's features (SCO/ACL links, packet size, power consumption mode). Host Controller Interface (HCI) is an interface through which BT devices access the lower layers of the BT protocol stack, allowing devices to pass and receive data, execute inquiries, and request authentication. Logical Link Control and Adaptation Protocol (L2CAP) shields the specifics of BT lower layers and provides a packet interface to higher layers.
Bluetooth protocols
Service Discovery Protocol (SDP) defines a service record format containing information about services provided by attributes. Attributes are composed of an ID (name) and a value, where IDs may be Universally Unique Identifiers (UUIDs). SDP defines an inquiry/response protocol for discovering services, including searching for and browsing services. RFCOMM (based on GSM TS07.10) emulates a serial-port to support a large base of legacy (serial-port-based) applications and allows multiple βportsβ over a single physical channel between two devices. Telephony Control Protocol Spec (TCS) handles call control (setup & release) and group management for a gateway serving multiple devices. Legacy protocol reuse allows reuse of existing protocols like IrDAβs OBEX or WAP for interacting with applications on phones.
Baseband
Addressing in Bluetooth uses three address types: the Bluetooth Device Address (BD_ADDR) which is a 48-bit IEEE MAC address, the Active Member Address (AM_ADDR) which is a 3-bit active slave address (with all zero being the broadcast address), and the Parked Member Address (PM_ADDR) which is an 8-bit parked slave address. This MAC address is split into three parts: the Non-significant Address Part (NAP) used for encryption seed, the Upper Address Part (UAP) used for error correction seed initialization and FH sequence generation, and the Lower Address Part (LAP) used for FH sequence generation.
Packet Structure
π Definition β Access Code: A code that starts every Bluetooth packet and comes in three types: Channel Access Code (CAC) which identifies a piconet, Device Access Code (DAC) used for paging and subsequent responses, and Inquiry Access Code (IAC) used for inquiry purposes.
Inquiry Procedure
The potential master identifies devices in range that wish to participate by transmitting an ID packet with the Inquiry Access Code (IAC) in the Inquiry state. When a device receives an inquiry, it enters the Inquiry Response state, returns an FHS packet with address and timing information, and then moves to the page scan state.
Page Procedure
The master uses the device's address to calculate a page frequency-hopping sequence. The master pages with an ID packet and the Device Access Code (DAC) of a specific slave. The slave responds with a DAC ID packet. The master then responds with its FHS packet, and the slave confirms receipt with a DAC ID before moving to the Connection state.
Channel Establishment
There are seven sub-states: Inquiry, Inquiry scan, Inquiry response, Page, Page scan, Master response, and Slave response.
Link Manager Protocol
The Link Manager carries out link setup, authentication, and link configuration. Functions include Piconet Management (channel control, master-slave switch), Link Configuration (low power mode, QoS, packet type selection), Security, Authentication, and Encryption. For Channel Control, all work related to channel control is managed by the master using a polling process. The master is the first device which starts the connection, but this role can change through master-slave role switch.
Connection State
There are four connection modes: Active Mode where the device participates actively on the transmission channel and the master regularly sends a packet to the slaves (polling) to enable slaves to send packets and re-synchronize. Sniff Mode is a low consumption mode where a Bluetooth module stays synchronized in the piconet and listens at regular intervals (Tsniff) for a short instant on specified slots. Hold Mode keeps the module synchronized with even lower consumption than Sniff mode, where only the counter on the Bluetooth chip is active. At the end of the Hold period, the module returns to active mode. Park Mode means the module is no longer an active member of the piconet but remains synchronized with the master and can listen to a broadcast channel (Beacon Channel).
L2CAP
L2CAP provides connection-oriented and connectionless data services to upper layer protocols, along with protocol multiplexing and demultiplexing capabilities. It handles segmentation and reassembly of large packets. L2CAP permits higher level protocols and applications to transmit and receive L2CAP data packets up to 64 kilobytes in length.
Links between Master and Slave
Two types of links exist: Synchronous Connection Oriented (SCO) allocates fixed bandwidth between point-to-point connection of master and slave. The master maintains the link using reserved slots and can support three simultaneous links. SCO provides bandwidth reservation/QoS, and no retransmissions are required or done in this mode. Asynchronous Connectionless (ACL) is a point-to-multipoint link between the master and all slaves. Only a single ACL link can exist, and 1, 3, or 5 slot packets are defined.
Flow Specification Parameters
QoS parameters in L2CAP define traffic flow specification indicating the performance level that the sender will attempt to achieve. These include: Service type, Token rate (bytes/second), Token bucket size (bytes), Peak bandwidth (bytes/second), Latency (microseconds), and Delay variation (microseconds).
β Key Takeaways
Bluetooth is an open, low-cost, low-power, short-range radio technology operating in the 2.4 GHz ISM band that uses Frequency Hopping Spread Spectrum with 1600 hops/sec and Time Division Duplex (TDD). Devices form piconets (one master, up to 7 active slaves) using FH-TDD-TDMA access, and multiple piconets can interconnect to form scatternets. The protocol stack includes core protocols (Radio, Baseband, LMP, L2CAP, SDP), cable replacement (RFCOMM), and telephony control protocols. Two link types exist: SCO for fixed-bandwidth, real-time traffic without retransmissions, and ACL for packet-switched, point-to-multipoint data with retransmissions possible. Bluetooth supports four connection statesβActive, Sniff, Hold, and Parkβfor power management, with Park being the lowest power mode where the device is no longer an active member but remains synchronized.
π§ Quick Revision Questions
- What is the purpose of the Inquiry procedure in Bluetooth, and what packet is transmitted by the master?
- Explain the difference between SCO (Synchronous Connection Oriented) and ACL (Asynchronous Connectionless) links in terms of bandwidth, retransmissions, and connection topology.
- What are the four connection states in Bluetooth, and how do they differ in terms of power consumption and synchronization?
- How does a piconet differ from a scatternet, and what is the maximum number of piconets that can exist in a scatternet?
- What are the three types of Bluetooth addresses (BD_ADDR, AM_ADDR, PM_ADDR), and what is the bit length of each?
π Lecture 40 β High Rate Wireless Personal Area Networks (WPAN)
π Overview: This lecture completes the discussion on Bluetooth by covering IP over Bluetooth and security, then transitions to IEEE 802.15.3, the high-rate WPAN standard. It provides a detailed examination of 802.15.3's architecture, coordination mechanisms, superframe structure, channel time management, power saving modes, and MAC frame format. Understanding this lecture is critical for comparing WPAN standards and grasping how high-rate, QoS-capable personal area networks operate.
ποΈ Topics Covered
This lecture begins with IP over Bluetooth versions 1.0 and 1.1, then discusses Bluetooth Security and presents a comparative table of all WPAN standards (802.15.1, 802.15.2, 802.15.3, 802.15.4). The main focus is an extensive overview of IEEE 802.15.3, including its motivation, topology with parent, child, and neighbor piconets, coordination processes (starting, handing over, creating child/ending piconets, association/disassociation), superframe structure with beacon, CAP, and CTAP, channel time management with dynamic channel selection, three power management modes (DSPS, PSPS, APS), and the MAC frame format with its ACK policies.
π Lecture Summary
IP Over Bluetooth
The lecture revisits the concept of running the Internet Protocol (IP) over Bluetooth, referencing both IP over Bluetooth v 1.0 and IP over Bluetooth v 1.1 as earlier specifications that enabled internet connectivity for Bluetooth devices.
Bluetooth Security
Security for the piconet is defined as one of two modes. Mode 0 (Open) means security membership is not required, and payload protection (data integrity or encryption) is not used by the MAC, though the PNC may use a list of DEV addresses to admit or deny entry. Mode 1 (Secure membership and payload protection) provides both membership security and payload protection.
WPAN Standards
A comprehensive table compares the four IEEE 802.15 WPAN standards. 802.15.1 (Bluetooth) offers 1 Mbps throughput for cell phones, PDAs, printers, and headsets with QoS suitable for voice. 802.15.2 addresses the coexistence of Bluetooth and 802.11b. 802.15.3 (High-rate WPAN) provides >20 Mbps for low-power, low-cost portable digital imaging and multimedia with very high QoS. 802.15.4 (Low-rate WPAN) operates at <0.25 Mbps for industrial, agricultural, medical, and sensor applications with relaxed QoS needs.
IEEE 802.15.3 - Overview
This high-data-rate WPAN is a potential future standard, motivated by the need for data, high-quality TV, and home cinema. It features a dynamic topology where mobile devices often join and leave the piconet with short connection times, multiple power management modes, and a secure network. It operates in the 2.4 GHz PHY with either 4 channels (high density) or 3 channels (to coexist with 802.11b). It supports 5 data rates: 11 Mbps (QPSK), 22 Mbps (DQPSK without coding), 33 Mbps (16QAM), 44 Mbps (32QAM), and 55 Mbps (64QAM). It is based on piconets in a personal space, analogous to a LAN in a larger area, where Data Devices (DEVs) establish peer-to-peer communication. A Piconet Coordinator (PNC) manages QoS requirements, power save modes, and access control. A new piconet created on the same channel as an existing PNC is called a child/neighbor piconet; if channel access is also controlled by the parent PNC, it is a dependent piconet.
IEEE 802.15.3 β Topology
The topology diagram shows a Parent Piconet Controller managing a piconet that includes Piconet Devices. A Child/Neighbor Piconet Controller may establish its own piconet, maintaining a Piconet Relationship with the parent. Data transmission is peer-to-peer. An Independent Piconet Controller operates without a parent relationship.
802.15.3 - Coordination in IEEE 802.15.3
The MAC is designed for fast connection time, ad hoc networks, QoS data transport, security, dynamic membership, and efficient data transfer.
- Starting a piconet: A DEV scans for the best channel and sends out beacons, becoming the PNC. If no channels are available, it establishes a child or neighbor piconet instead. This process does not ensure the "most capable" PNC is initially selected.
- Handing over control: When a DEV associates, the PNC checks its capabilities to see if it is more capable. The handover process maintains all existing time allocations to avoid interruption in delivery.
- Creating a child piconet: A child piconet is formed under an established (parent) piconet. It is useful for extending coverage area or shifting computational/memory requirements. It uses a distinct Piconet ID (PNID) and acts autonomously except for its dependence on a private CTA from the parent piconet.
- Ending a piconet: If the PNC stops operation with no other capable DEVs, it places a PNC Shutdown information element (IE) in the beacon. If the PNC leaves abruptly, the piconet stops. After the association timeout period (ATP) expires, a capable DEV can start a new piconet. The parent PNC can end a dependent piconet via a Disassociation Request command.
- Association and disassociation: Associating gives a DEV a unique DEVID (one octet) to save overhead compared to the 8-octet DEV address. The association process optionally provides service information and PNC capabilities. The PNC broadcasts info about all DEVs. Disassociation invalidates the DEVID, which the PNC cannot reissue until a waiting period expires.
π Definition β DEVID: A unique one-octet identifier assigned to each associated DEV, used instead of the 8-octet DEV address to save overhead. π Definition β Child Piconet: A piconet formed under an established parent piconet, using a distinct PNID and a private CTA from the parent.
IEEE 802.15.3 β Superframe
The superframe is composed of three parts:
- The beacon: Sets timing allocations and communicates management information for the piconet. It contains piconet synchronization parameters and Information Elements (IEs).
- The contention access period (CAP): Optional period used to communicate commands and/or asynchronous data using CSMA/CA with a backoff scheme.
- The channel time allocation period (CTAP): Composed of channel time allocations (CTAs), including management CTAs (MCTAs). It uses standard TDMA where DEVs have specified time windows. MCTAs may be placed at any position in the superframe by the PNC.
IEEE 802.15.3 β CAP and CTA
The CAP allows contention via CSMA/CA. The CTAP uses a standard TDMA protocol for contention-free access, enabling power saving and QoS. There are three types of CTAs:
- Private CTA: Used for dependent piconets.
- Dynamic CTA: Scheduled on a superframe-by-superframe basis.
- Pseudo-Static CTA: Only for isochronous streams. A DEV is allowed to transmit during a CTA as long as the number of consecutive lost beacons is less than mMaxLostBeacons.
Channel time management
There are three methods for communicating data between DEVs: sending asynchronous data in the CAP, allocating channel time for isochronous streams in the CTAP, and allocating asynchronous channel time in the CTAP. For dynamic channel selection, the PNC can change the piconet's channel without user intervention. To evaluate channels, the PNC can: gather info from DEVs using the Channel Status Request command, perform a passive scan, or request other DEVs to scan using the Remote Scan Request command. π‘ Why this matters: Dynamic channel selection is critical in the noisy ISM bands to maintain QoS.
Power Management
Three techniques enable DEVs to turn off for one or more superframes:
- Device synchronized power save (DSPS) mode: Allows DEVs to wake up and exchange traffic at the same time, making it easy for other DEVs to determine when a DSPS DEV is available.
- Piconet-synchronized power save (PSPS) mode: Allows DEVs to sleep at intervals defined by the PNC. The DEV sends a request to the PNC to enter this mode.
- Asynchronous power save (APS) mode: The only responsibility is to communicate with the PNC before the end of its ATP to preserve membership.
MAC Frame format
The MAC frame format includes fields for frame control, destination and source IDs, fragmentation control, stream index, and the frame payload. A key component is the ACK Policy.
- NO ACK: Used for frames that don't require guaranteed delivery, where retransmission would arrive too late or an upper layer is handling ACKs.
- Immediate-ACK (Imm-ACK): Each frame is individually ACKed upon reception.
- Delayed-ACK (Dly-ACK): The source sends multiple frames without intervening ACKs. The ACKs are grouped into a single response frame sent when requested by the source. This decreases overhead compared to Imm-ACK while still verifying delivery. π Definition β Dly-ACK (Delayed ACK): An ACK policy where multiple data frames are sent before a single aggregated ACK is requested, reducing overhead. π Example: In a video stream, using Imm-ACK for each video frame might cause high overhead. Using Dly-ACK, the source could send 5 video frames, then request a single ACK frame that acknowledges all 5 frames, reducing the total number of transmitted frames.
β Key Takeaways
The IEEE 802.15.3 High-Rate WPAN is designed for high-throughput, low-latency multimedia applications, operating at up to 55 Mbps in the 2.4 GHz band. Its MAC relies on a Piconet Coordinator (PNC) to manage a star topology with peer-to-peer data, using a superframe structure composed of a Beacon, optional CAP (CSMA/CA), and CTAP (TDMA with CTAs). Key coordination features include selective PNC handover, child piconet creation for coverage/capability extension, and dynamic channel selection to avoid interference. Power saving is achieved through three distinct modes (DSPS, PSPS, APS), and efficient data transfer is supported by three ACK policies (No ACK, Imm-ACK, Dly-ACK). The 1-octet DEVID replaces the 8-octet hardware address to reduce overhead.
π§ Quick Revision Questions
- What are the three parts of an IEEE 802.15.3 superframe, and what is the primary purpose of each?
- Explain the difference between a child piconet and a neighbor piconet, and when a dependent piconet is created.
- What is the advantage of using a DEVID (1 octet) instead of the DEV's hardware address (8 octets)?
- Compare the three power save modes (DSPS, PSPS, APS) in terms of how a DEV determines its sleep/wake schedule.
- A source DEV sends a burst of 10 data frames and wants to verify delivery with minimal overhead. Which ACK policy should it use, and why not the other two?
π Lecture 41 β IEEE 802.15.4/ZigBee
π Overview: This lecture introduces ZigBee and the IEEE 802.15.4 standard for low-power, low-data-rate wireless mesh networking. It covers the architecture, device types, topologies, addressing, channel access mechanisms, and superframe structure, emphasizing applications in monitoring and control.
ποΈ Topics Covered
The lecture begins with an overview of ZigBee, including its definition, market features, and target markets, followed by the ZigBee/802.15.4 architecture and general characteristics. It then details IEEE 802.15.4 basics, device types (FFD vs RFD), topologies (star, peer-to-peer), network extension, device addressing, PHY layer overview, frequency bands and data rates, general radio specifications, and channel access mechanisms (slotted/unslotted CSMA/CA). The data transfer model and superframe structure are also discussed.
π Lecture Summary
What is ZigBee Alliance?
An organization with a mission to define reliable, cost effective, low-power, wirelessly networked, monitoring and control products based on an open global standard. The alliance provides interoperability, certification testing, and branding.
IEEE 802.15.4: What is ZigBee?
ZigBee is a standard for mesh networking, providing reliability through meshed connectivity. It is designed for low power applications, enabling very long battery life. It offers a low data rate (20-250 Kb/sec, depending on band), is very secure with AES-128 encryption available, and is self configuring, allowing ad hoc networks and ease of installation.
ZigBee/IEEE 802.15.4 Market Feature
Key features include: low power consumption, low cost, low offered message throughput, support for large network orders (<= 65k nodes), low to no QoS guarantees, and flexible protocol design suitable for many applications.
ZigBee/802.15.4 Architecture
The architecture is layered, with the PHY and MAC layers defined by IEEE 802.15.4, and the NWK (Network) layer, APL (Application) layer, and ZDO (ZigBee Device Object) defined by the ZigBee Alliance. The Security Service Provider is also part of the architecture.
ZigBee/802.15.4 Technology: General Characteristics
- Data rates of 250 kbps, 20 kbps and 40 kpbs.
- Star or Peer-to-Peer operation.
- Support for low latency devices.
- CSMA-CA channel access.
- Dynamic device addressing.
- 16 channels in the 2.4GHz ISM band, 10 channels in the 915MHz ISM band, and one channel in the European 868MHz band.
IEEE 802.15.4 Basics
802.15.4 is a simple packet data protocol for lightweight wireless networks. Channel access is via Carrier Sense Multiple Access with collision avoidance (CSMA/CA) and optional time slotting.
IEEE 802.15.4 Device Types
There are two different device types:
- A full function device (FFD) can operate in three modes: Device, Coordinator, and PAN coordinator.
- A reduced function device (RFD) can only operate in Device mode.
FFD vs RFD
- Full function device (FFD): Can operate in any topology, is network coordinator capable, and can talk to any other device.
- Reduced function device (RFD): Limited functionality, typically used as an end device.
Extending ZigBee Networks
A ZED (ZigBee Extension Device) is a ZigBee router with a wire interface. It joins two or more radio disjoint PANs and provides a βwormholeβ within a single PAN, offering a low cost, high reliability link within the radio network. It βextendsβ the ZigBee network layer.
Device Addressing
Each independent PAN will select a unique PAN identifier. Addressing modes include:
- Star: Network (64 bits) + device identifier (16 bits)
- Peer-to-peer: Source/destination identifier (64 bits)
IEEE 802.15.4 PHY Overview
PHY functionalities include:
- Activation and deactivation of the radio transceiver
- Energy detection within the current channel
- Link quality indication for received packets
- Clear channel assessment for CSMA-CA
- Channel frequency selection
- Data transmission and reception
Frequency Bands and Data Rates
The standard specifies two PHYs:
- 868 MHz/915 MHz DSSS PHY (11 channels): 1 channel (20 Kb/s) in European 868 MHz band, 10 channels (40 Kb/s) in 915 MHz ISM band.
- 2450 MHz DSSS PHY (16 channels): 16 channels (250 Kb/s) in 2.4 GHz band.
π Definition β DSSS: Direct Sequence Spread Spectrum, a modulation technique used to spread the signal over a wider frequency band.
General Radio Specifications
- Transmit Power: Capable of at least β3 dBm.
- Receiver Sensitivity: -85 dBm (2.4 GHz) / -91 dBm (868/915 MHz).
- Link quality indication: May be implemented using signal to noise ratio estimation or receiver energy detection.
Channel Access Mechanism
Two types of channel access mechanisms, based on network configuration:
- In non-beacon-enabled networks β unslotted CSMA/CA channel access mechanism.
- In beacon-enabled networks β slotted CSMA/CA channel access mechanism, where the superframe structure is used.
CSMA/CA Algorithm
Each device shall maintain three variables for each transmission attempt:
- NB: Number of slots the CSMA/CA algorithm is required to backoff while attempting the current transmission.
- BE: The backoff exponent, related to how many backoff periods a device shall wait before attempting to assess a channel.
- CW: (A special design variable).
Data Transfer Model
- Data transferred from device to coordinator: The device sends data to the coordinator.
- Data transferred from coordinator to device: The coordinator sends data to the device.
Superframe
- In the CFP (Contention Free Period), a GTS (Guaranteed Time Slot) may consist of multiple slots, all assigned to a single device, for either transmission (t-GTS) or reception (r-GTS).
- In the CAP (Contention Access Period), each βcontention slotβ is of 20 symbols long.
β Key Takeaways
ZigBee and IEEE 802.15.4 are designed for low-power, low-data-rate wireless control and monitoring networks, offering mesh networking for reliability. The standard defines two device types: FFD (full function) and RFD (reduced function), supporting star and peer-to-peer topologies. Channel access uses either slotted or unslotted CSMA/CA, depending on beacon mode. The superframe structure includes both a Contention Access Period (CAP) and a Contention Free Period (CFP) with Guaranteed Time Slots (GTS). Key features include AES-128 security, dynamic addressing, and operation in the 868 MHz, 915 MHz, and 2.4 GHz frequency bands.
π§ Quick Revision Questions
- What are the two device types defined in IEEE 802.15.4, and what are their capabilities?
- Explain the difference between slotted and unslotted CSMA/CA channel access mechanisms.
- What are the key functionalities of the PHY layer in IEEE 802.15.4?
- Describe the three variables (NB, BE, CW) maintained for each CSMA/CA transmission attempt.
- What is the purpose of a GTS in the superframe structure, and how is it assigned?