Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04W—WIRELESS COMMUNICATION NETWORKS
  • H04W74/00—Wireless channel access
  • H04W74/02—Hybrid access
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
  • H04L9/40—Network security protocols
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04W—WIRELESS COMMUNICATION NETWORKS
  • H04W28/00—Network traffic management; Network resource management
  • H04W28/02—Traffic management, e.g. flow control or congestion control
  • H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information

Definitions

• This invention generally relates to wireless radio systems and more particularly to an improved recurring MAC superframe structure and beaconing method for radio networks.
• the MAC protocol in one form is designed for operation in the 2.4 GHz ISM band with a frequency hopping radio. Any system that uses this band in the US must comply with the FCC rules. Additionally, if the band is unlicensed, the system must also be able to operate in the presence of other ISM band radio systems, and interference sources, e.g. microwave ovens.
• the MAC protocol as known in one form is a hybrid protocol which combines both TDMA and CSMA/CA access mechanisms. Hybrid MAC protocols can offer features which ensure good performance under a wide range of conditions.
• the MAC protocol in one form uses a superframe, which incorporates two contention-free periods (CFPs) and a contention period.
• the start of the superframe is the point at which a station begins to hop to a new channel and ends immediately before the station starts to hop to the next channel.
• the duration of the superframe is fixed and is the same as the dwell or hop period.
• the access mechanism used during each CFP is TDMA, whilst the access mechanism used during the contention period is CSMA/CA.
• Each of the contention free periods is divided into a number of pairs of fixed length slots, two per voice connection.
• the first slot in each pair may be used to transmit voice data from the Control Point to a node (downlink) and the second is used to transmit voice data from a node to the Control Point (uplink).
• MAC protocols in the context of the present invention can be categorized into two types: distributed and centralized.
• Distributed MAC protocols such as the WiMedia MAC
• centralized MAC protocols are desirable, for example, to support the scenario with a high density of devices and to support improved QoS guarantee.
• each type of protocol has its advantages in some particular application scenarios, but may be sub- optimal in other cases. Therefore a flexible MAC protocol that operates in different modes and supports scenarios that require distinct features would be able to support a wider range of applications, and therefore would have a much higher market.
• current MAC protocols can only support either centralized operation or distributed operation, primarily due to the limitations of existing MAC superframe structure and related operations.
• Distributed MAC protocols such as IEEE 802.11 DCF (Distributed Coordinated Function) and 802. l ie EDCA (Enhanced distributed Channel Access), have salient features for supporting P2P (peer-to-peer) applications scenarios. They support direct link setup, avoid point of failure and facilitate flexible multi-hop channel sharing. However, it is not straightforward to support QoS strict real-time applications minimize energy consumption and address the hidden terminal problem under the context of distributed approaches.
• the WiMedia MAC provides a distributed beaconing protocol and a distributed reservation protocol (DRP) to address such problems. But the beaconing protocol requires each device to own a beacon slot in each beacon period and send a beacon in every superframe, which leads to scalability problems due to the overhead of beacon slots in every superframe. In other words, to support high node density applications becomes very difficult due to the increasing overhead of the beacon slots with the increase of the node density.
• HCF Hybrid
• Coordination Function are better to support QoS, high-density applications, and energy- saving.
• Centralized MAC protocols provide deterministic channel access controlled by a coordinator (access point or cluster-head) device, which improves channel usage efficiency and enables more efficient energy- saving schemes.
• a coordinator access point or cluster-head
• the centralized architecture cannot really enable direct P2P link setup, nor avoid a single point of failure and support flexible multi-hop channel sharing.
• a beacon is transmitted immediately after the hop. This Beacon is used to maintain network synchronization, control the format of the superframe and manage when each node should transmit and receive data.
• the CFP (contention Free Period) at the end of the superframe is used for the initial transmission of the voice data, whilst the CFP at the start of the superframe is used for the optional retransmission of any data which was not received or was incorrectly received.
• the dwell period is fixed at 20ms to provide acceptable performance with respect to latency.
• the length of the dwell period also means that each voice data message contains 20 ms of ADPCM data (640 bits), equivalent to an extended DECT B-field and 48 bits of control data, equivalent to the DECT A-field.
• each packet transmitted includes the necessary MAC and PHY headers.
• the MAC can provide either 4 voice connections with a large enough CFP at the start of the frame to enable all the voice data to be retransmitted, or a larger number of connections (e.g. 6), but in this case the CFP at the start of the frame is only large enough for the retransmission of data from two connections.
• the CFP in which initial transmission occurs and the CFP in which retransmission occurs are separated by a frequency hop, giving frequency diversity which is particularly important given the environment in which the protocol will operate.
• the service slot is used by voice nodes to communicate with the Control Point.
• Each voice data packet transmitted by a node includes in the packet header a piggyback acknowledgement of the last voice data message received by the node. i.e. in the uplink packet, the voice node acknowledges the downlink packet sent by the control point.
• This system allows the control point to determine prior to a hop which voice data transmissions were lost, to determine the retransmissions required, and advertise these retransmissions in the Beacon at the start of the next superframe.
• Each voice data packet can only be retransmitted once.
• the time between the two CFPs, the contention period, is used for data transmissions using a CSMA/CA protocol similar to that specified in the 802.11 standard [802.11].
• the MAC uses a slotted contention scheme, acknowledgement and retransmission of data messages and a fragmentation scheme to improve performance.
• the CSMA/CA period occupies the whole of the superframe, with the exception of the space required for the hop and Beacon, maximizing data throughput.
• the data nodes can create an ad-hoc network in which control of the network is distributed between all the nodes.
• the use of beacon in management The primary function of the Beacon is to enable all nodes to synchronize to the hopping pattern of the network.
• the Beacon transmitted by the Control Point is also used to manage the network during the contention free periods.
• the Control Point Beacon can include a list of active voice connections (and therefore slot assignments), retransmission slot assignments for the current superframe, connection status information and paging information.
• Control Point performs "connection packing" to eliminate unused slots and maximize the contention period and therefore maximize data throughput
• each node schedules the transmission of an ad-hoc beacon during each dwell period.
• a node uses its address to determine when it should send an ad-hoc beacon preventing collision of ad-hoc beacons from different nodes. If a node receives messages from two different nodes before it is due to transmit its own beacon, then it cancels transmission of the beacon
• the service slot is used by voice nodes to send management messages to a Control Point, e.g., to request a connection from the Control Point. Since there is only one service slot it is possible for two nodes to transmit at the same time and for their transmissions to collide.
• Each management message is explicitly acknowledged by the Control Point in the CPB, and if there is no acknowledgement a node performs a random backoff across a number dwell periods before re-sending the message. When closing a connection, the node transmits a management in its voice slot.
• This invention presents a new MAC superframe structure and a set of related operations, which are the corner stones of a MAC protocol for flexible wireless systems.
• the unified MAC protocol enabled by this invention could support flexible operation in either distributed mode or centralized mode, and support seamless transition from one mode to another adaptively.
• the invention in one form could allow a distributed network and a centralized network to co-exist harmoniously, as well as multiple centralized networks .
• the IEEE 802.11 standard and its 802. l ie amendment do support both distributed operation (mandatory) and point-coordinated operation (optional) modes, the two modes require totally different MAC structure and operations. Therefore, the transition from one to another can not be seamless in 802.11 networks. In most cases, a device only operates in one mode, for example, the mandatory distributed mode. Coexistence is also an open problem in 802.11 networks, since QoS can not be fully guaranteed in a point-coordinated network, if it co-exists with a distributed or another centralized network.
• the present invention in one form proposes a new flexible MAC structure that harmoniously supports multiple operation modes and enables a seamless transition between operation modes.
• the invention in one form resides in a flexible wireless system that supports both centralized and distributed modes of MAC protocols, using a new MAC superframe structure which comprises: a new recurring MAC superframe including a beacon period; a data/sense/sleep period for data selective communication, sleep and channel sensing for detecting primary users in cognitive systems; and a signaling window used for exchanging network entry messages and channel reservation requests.
• the flexible wireless system expediently includes beacon operation, a peer beacon device in distributed mode and a master beacon device in centralized mode which are used to selectively participate in the beacon operation, as well as a slave beacon device associated with the master device.
• the system may include controlled channel access for the beacon period, wherein the channel access is reservation based.
• the signaling window may be located at the end of the MAC superframe structure.
• the invention in a second form resides in a flexible wireless system that supports both centralized and distributed modes of MAC protocols, which uses a new MAC superframe structure comprising: a new recurring MAC superframe including a beacon period; a data/sense/sleep period for data selective communication, sleep and channel sensing for detecting primary users in cognitive systems; a signaling window used for exchanging network entry messages and channel reservation requests, the system including beacon operation, a peer beacon device in distributed mode and a master beacon device in centralized mode which are used to selectively participate in said beacon operation, as well as a slave beacon device associated with said master device.
• the invention resides in a method for a wireless system that flexibly supports both centralized and distributed modes of MAC protocols, which uses a new MAC superframe structure the method comprising: deploying a new recurring MAC superframe including using a beacon period; using a data/sense/sleep period for data selective communication, sleep and channel sensing for detecting primary users in cognitive systems; deploying a signaling window used for exchanging network entry messages and channel reservation requests; the method including a beacon operation using a peer beacon device in distributed mode and a master beacon device in centralized mode which are used to selectively participate in the beacon operation, as well as using a slave beacon device associated with the master device.
• FIG 1 illustrates a reference network architecture and device type in the context of the present invention.
• FIG 2 illustrates an exemplary reference MAC Structure embodying the invention.
• a Subnet is defined as a collection (or group) of devices which are under the management of one entity (e.g., network administrator) and share a common MAC protocol. If the medium access in the subnet is controlled by a single device, a subnet is defined as a Centralized Subnet. On the other hand, if the medium access in the subnet is coordinated in the distributed manner, a subnet as a Distributed Subnet. As illustrated in FIG 1, as applied in the present invention, there are three basic types of devices.
• a device in a distributed subnet is called a Peer Device; the subnet coordinator in a centralized subnet is called a Master Device; and, a device other than subnet coordinator in the centralized subnet is called a Slave Device.
• a device can only operate as one of the three types, i.e., a device can either be a Master, Slave or Peer device.
• two subnets as are referred to Neighboring Subnets if the two subnets share the same channel and at least one active device from the first subnet is in the transmission range of the second subnet.
• the proposed MAC protocol follows a recurring Superframe structure, which consists of a beacon period (BP), data/sense/sleep period (DSSP) and a signaling window (SW).
• BP beacon period
• DSSP data/sense/sleep period
• SW signaling window
• the signaling window and beacon period are used for broadcasting/exchanging control/management information and their sizes (in time slots) are dynamically adjustable.
• All the devices should keep awake during beacon period and signaling window in order to capture all the control/management information which may be relevant to every device.
• a device may exchange data, monitor one or more channels (needed in cognitive networks), or go to sleep mode during Data/Sense/Sleep period.
• a device is treated as a beaconing device if it owns a beacon slot in a beacon period BP and regularly transmits beacons.
• BP beacon period
• a beaconing device Different from WiMedia, not every device is required to be a beaconing device, which allows the flexibility and scalability. Whether a device should become a beaconing device for purposes of this invention will depend on the following considerations:
• a peer device should be a beaconing device.
• a master device must be a beaconing device. In other words, a master device must own one dedicated beacon slot in the beacon period. There could be multiple master devices in the same network, each of which controls a group of slave devices. In such case, each master device shall own one beacon slot in the beacon period.
• a slave device is normally a non-beaconing device, which does not own a beacon slot in the beacon period. But in certain scenarios, a slave device could be a beaconing device, e.g., to enable co-existence and reduce a hidden terminal problem.
• the considerations above can be used to help establish a beaconing backbone across the connected subnets. With the beaconing backbone and the dedicated use of beacon slots, those devices in a coordination role or in real-time and intensive transmission can easily guarantee control information (including bandwidth reservation information) reliably and timely delivered, thus improving QoS support and system reliability.
• BPST Beacon Period Start Time
• superframe number The BP starting time and superframe number are initiated by the first device establishing the beacon period, which could be a master device or a peer device.
• the merging of BPST and superframe structure is necessary if two disconnected subnets become connected.
• Beacon period (BP) operation The channel access method for beacon period is reservation based, specifically TDMA based. Beacon period is divided into multiple equal beacon slots, numbered from zero and increased one by one. The starting time of the Superframe is equivalent to the starting time of the first beacon slot. Each beaconing device owns one beacon slot, sends a beacon in its own beacon slot and listens to other beacon slots. This is similar as WiMedia.
• a new beaconing device should preferably select the smallest available beacon slot in the BP as its own beacon slot. For example, if a device is the very first device which initiates BP, it should choose beacon slot zero as its own beacon slot number. A beaconing device should regularly send a beacon in its own beacon slot.
• a beaconing device should use the beacon to advertise its own superframe number, device type (illustrated in Tablel), subnet ID (which could be a name string, e.g., configured by subnet owner), the SW length, in addition to what is defined in WiMedia standard for example, Beacon Period Occupancy IE (BPOIE, which includes BP length), DRP Availability IE, PCA Availability IE, Traffic Indication Map (TIM) IE, Identification IE.
• BPOIE Beacon Period Occupancy IE
• TIM Traffic Indication Map
• Beacon period length is adjustable between minimum BP length (BPj 111n , e.g. one beacon slot) and maximal BP length (BP max ).
• the BP length is of the minimum length by default.
• the beacon period may be extended.
• the beacon period may be downsized and the beacon slots may be shifted to lower numbered slots.
• BP extension is to ensure that each beaconing device be aware of the BP adjustment request and confirm such adjustment request. For example, every beaconing device should confirm/update/advertise such BP adjustment in its beacon.
• a signaling window is an adjustable time window that is used for exchanging control or management information, for example, network entry messages, channel reservation requests and traffic indication.
• a signaling window is preferably placed at the end of a Superframe. However, it may be positioned in another location, e.g., after BP, according to system preference.
• Any device may use the signaling window to send control/management information on demand. Different from beacon period, the whole signaling window is shared by all the devices depending on the opportunity; thus improving channel efficiency for signaling.
• the advantages of using a reserved signaling window instead of other randomly available MAS in DSSP to exchange control information are energy-saving and reliability. For example, a device can go to sleep mode during DSSP without missing control message. Although a device can still use any available MAS in DSSP for exchanging control information, it may require all the intended receivers to keep awake during DSSP, which reduces energy efficiency.
• MASs in DSSP may not be available during "peak data traffic time", which could cause intolerable delay to critical control messages, such as channel-switch-messages to protect primary users in cognitive radio systems.
• Signaling window duration is adjustable between minimum SW length (SW m1n ) and maximal SW length (SW max ).
• Any beaconing device in the network may ask to extend the current signaling window if the signaling window becomes overloaded. Many ways can be used to judge whether the signaling window is overloaded, such as observing collision probability, channel utilization ratio and measurement reports from other devices. If the signaling window gets overloaded, a beaconing device may include an information element in its beacon to request such SW extension. Every beaconing device receiving such request should extend the SW accordingly.
• the channel access method for signaling window is contention based. Slotted aloha or back-off based carrier sensing medium access (CSMA) could be used for the contention. For slotted aloha method, based on the fact that the maximal signaling message length is much less than the maximum length of a regular Medium Access Slot (MAS), the signaling slot length should be smaller than the regular MAS slot length.
• CSMA carrier sensing medium access
• the use policy of medium access slots (MAS) in DSSP should follow either Reservation Access or Prioritized Contention Access (PCA) or Group-PCA.
• the reservation access and PCA can be applied to both distributed subnet and centralized subnet.
• the Group- PCA is applied to the centralized subnet only.
• the use policy is published by beaconing devices and it can be updated on a Superframe -basis.
• a beaconing device should always publish its own view of use policy for every MAS in DSSP. Accordingly, a master device should announce all the reservation related with itself as well as its associated slave devices.
• a medium access slot marked as reservation can be accessed preferably only by the reservation owner.
• a medium access slot marked as PCA available is open to all the devices in the network.
• the Group-PCA is also proposed, which is only open to a specific subnet, for example a master device and its slave devices.
• the master device should make reservation, label the owner suitably and mark the reserved MASs as Group-PCA.
• the master device may have higher priority than slave devices to access medium, e.g., to send Poll messages.
• the proposed invention could serve as a basis for future WiMedia UWB standard, IEEE 802.11, Cognitive wireless networks, and IEEE 802.15 wireless systems, but the implementation is not limited thereto.

Landscapes

• Engineering & Computer Science (AREA)
• Computer Networks & Wireless Communication (AREA)
• Signal Processing (AREA)
• Computer Security & Cryptography (AREA)
• Mobile Radio Communication Systems (AREA)
• Small-Scale Networks (AREA)

Priority Applications (5)

Applications Claiming Priority (2)

Publications (1)

Family

ID=40462841

Family Applications (1)

Country Status (8)

Cited By (10)

Families Citing this family (42)

Citations (3)

Family Cites Families (7)

• 2008
  • 2008-12-04 TW TW097147174A patent/TWI482456B/zh active
  • 2008-12-05 CN CN200880119571XA patent/CN101889469B/zh active Active
  • 2008-12-05 JP JP2010536578A patent/JP5479356B2/ja active Active
  • 2008-12-05 US US12/746,279 patent/US8467357B2/en active Active
  • 2008-12-05 ES ES08855846T patent/ES2391920T3/es active Active
  • 2008-12-05 EP EP08855846A patent/EP2232938B1/en active Active
  • 2008-12-05 WO PCT/IB2008/055128 patent/WO2009072089A1/en not_active Ceased
  • 2008-12-05 KR KR1020107014829A patent/KR101549415B1/ko active Active

Patent Citations (3)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events